Inhibition of Colitis by IL-25 Associates with Induction of Alternatively Activated Macrophages

Abstract

Background

Interleukin (IL)-25, a Th2-related factor, inhibits the synthesis of inflammatory cytokines by macrophages and attenuates experimental colitis in mice. The mechanism underlying the counterregulatory effect of IL-25, however, remains unknown. Since Th2-cytokines can abrogate inflammatory pathways by inducing alternatively activated macrophages (AAMs), we evaluated whether AAMs are involved in the IL-25-mediated anticolitic effect.

Methods

AAM-related markers were evaluated in peritoneal and lamina propria mononuclear cells of mice with or without 2,4,6-trinitrobenzenesulphonic acid (TNBS)-induced colitis treated with IL-25 and/or neutralizing IL-4, IL-13, and transforming growth factor beta 1 (TGF-β1) antibodies. Peritoneal AAMs induced in vivo by injecting mice with IL-25 were transferred to mice with TNBS colitis. Finally, we assessed the in vitro effect of IL-25 on the alternative activation of peritoneal F4/80+ cells.

Results

IL-25 enhanced the expression of AAM-related markers in F4/80⁺ cells infiltrating the peritoneum and colon of naïve and colitic mice. Peritoneal F4/80⁺ cells isolated from IL-25-treated mice reduced the severity of TNBS colitis when injected intraperitoneally to recipient mice. Since IL-25 did not directly induce AAM in vitro and in vivo in mice, IL-25 administration enhanced the expression of IL-4, IL-13, and TGF-β1, which are known to promote AAM differentiation, we finally assessed whether such cytokines were involved in the IL-25-driven AAM induction. Blockade of IL-4, IL-13, and TGF-β1 with neutralizing antibodies in mice did not inhibit the stimulatory effect of IL-25 on AAM gene expression.

Conclusions

The IL-25-mediated anticolitic effect is associated with induction of AAMs, a subset of macrophages with antiinflammatory properties.

IL-25, Crohn's disease, colitis, macrophages

Crohn's disease (CD) and ulcerative colitis (UC), the main forms of inflammatory bowel disease (IBD) in humans, are chronic relapsing disorders of the intestine.¹ The cause of IBD is unknown, but evidence suggests that, in both CD and UC, the tissue damage is due to an abnormal immune reaction within the intestinal wall, which is directed against luminal bacterial antigens.^2,3 Analysis of inflammatory and antiinflammatory pathways in the intestine of IBD patients has also shown that the deregulated immune response is characterized by an altered balance between inflammatory and regulatory cytokines. For instance, the excessive production of inflammatory T helper (Th)1- and Th17-related cytokines in CD tissue and Th2- and Th17-cytokines in UC associates with a defective activity of transforming grow factor (TGF)-β1.^4,–7 In this context, we have recently shown that IBD-related inflammation is marked by a diminished production of interleukin (IL)-25,⁸ a Th2-type cytokine produced by several cell types and involved in the negative regulation of inflammatory pathways in the gut.^8,–12 These observations suggest that the decreased synthesis of IL-25 can contribute to sustaining the mucosal inflammation in IBD. Mechanisms by which IL-25 suppresses immune-inflammatory responses in the gut are not yet known. However, the fact that monocytes/macrophages express functional IL-25 receptors^8,13 suggests that these cells can be involved in the IL-25 antiinflammatory effects. Indeed, monocytes/macrophages can be either proinflammatory or antiinflammatory, depending on the functional phenotypes these cells acquire in response to distinct subsets of cytokines and other tissue-derived signals.¹⁴ For example, following activation with interferon-gamma (IFN-γ), the signature cytokine of Th1 cell responses, or microbial pathogen-associated molecular patterns, monocytes/macrophages are induced to produce high amounts of inflammatory cytokines and reactive oxygen species necessary to eradicate invading organisms and tumor cells. These cells are termed classically activated macrophages (CAMs) and can also cause immunopathology if their activation is not adequately controlled.¹⁴ Monocytes/macrophages can also undergo a different activation program, referred to as “alternative activation.” Alternatively activated macrophages (AAMs) produce antiinflammatory molecules and various components of extracellular matrix and contribute to the host response to parasites, fibrosis in granulomatous diseases, angiogenesis, tissue repair, and healing.^15,16 AAMs have been shown to attenuate experimental inflammation in the gut¹⁷ and to be induced by antibodies against tumor necrosis factor alpha (TNF-α),¹⁸ thus confirming the regulatory properties of these cells. Since AAM differentiation is preferentially induced by Th2 cytokines,¹⁹ we here investigated the possibility that AAMs could be involved in the IL-25-mediated anticolitic effect.

Materials and Methods

In Vivo Studies

All reagents were from Sigma-Aldrich (Milan, Italy) unless specified. Six to 8-week-old wildtype Balb/c mice were purchased from Harlan (Udine, Italy) and hosted in conventional animal facilities at the University of Rome Tor Vergata (Italy). Naïve mice were injected intraperitoneally (IP) with recombinant mouse IL-25 (10 μg/mouse, R&D Systems, Minneapolis, MN) dissolved in phosphate-buffered saline (PBS) or vehicle (PBS, 200 μL/mouse) and sacrificed at 24, 48, 72, and 96 hours. The dose of IL-25 we selected for this study is the same as that we previously used to suppress experimental colitis in mice.⁸ In some experiments, mice were injected IP with a neutralizing IL-4 and/or IL-13 antibody (500 μg/mouse, R&D Systems) 6 hours before and 6 hours after IL-25 administration. In further studies, mice were injected IP with a neutralizing TGF-β1 antibody (600 μg/mouse, R&D Systems) 1 hour before and 48 hours after IL-25 administration. For induction of trinitrobenzene sulfonic acid (TNBS) colitis, 2.5 mg of TNBS in 50% ethanol (150 μL/mouse) was administered to lightly anesthetized mice through a 3.5F catheter inserted into the rectum. Controls were mice treated with 150 μL of 50% ethanol. In these studies, recombinant mouse IL-25 (10 μg/mouse) was injected IP 24 hours after the administration of TNBS. Weight changes were recorded daily, and tissues were collected for RNA analysis, isolation of lamina propria mononuclear cells (LPMC), or were embedded in OCT for histological analysis.

F4/80⁺ cells were isolated from the peritoneum of mice treated with IL-25 or PBS by washing the peritoneal cavity with cold PBS followed by incubation of the resulting cell population with F4/80-coated beads and magnetic sorting according to the manufacturer's instructions (Invitrogen, Milan Italy). F4/80⁺ cells were then checked for their purity by flow cytometry (> 97%) and transferred to naïve Balb/c mice by IP injection (0.75 × 10⁶ cells/mouse). Two days later, these mice received intrarectal administration of TNBS. To monitor F4/80⁺ cell migration in vivo, F4/80⁺ cells isolated from the peritoneum of mice treated with IL-25 or PBS were either left unlabeled or labeled with the fluorescent membrane marker chloromethyl-benzamidodialkylcarbocyanine (CM-Dil) (Invitrogen) as previously described²⁰ and injected IP to naïve mice. Mice were then killed 24, 48, or 72 hours later; peritoneal and mesenteric lymph node cells and colonic LPMC were harvested and analyzed by flow cytometry. Experiments were approved by the local Ethics Committee.

Cell Isolation and Culture

LPMC were isolated from the colons of naïve mice treated with IL-25 or PBS as previously described⁸ and analyzed for arginase1 (Arg1), found in the inflammatory zone-1 (FIZZ1), YM1, and inducible nitric oxide synthase (INOS) RNA expression by real-time polymerase chain reaction (PCR). An aliquot of LPMC was also used to purify F4/80⁺ cells by magnetic sorting using anti-F4/80-coated beads. The resulting cell preparations (>95% F4/80⁺ cells as assessed by flow cytometry) were analyzed for the content of RNA transcripts for Arg1, FIZZ1, YM1, and INOS by real-time PCR. Additionally, LPMC were isolated from control and colitic mice, treated with PBS or IL-25, resuspended in RPMI 1640, supplemented with 10% inactivated fetal bovine serum (FBS), penicillin (100 U/mL), and streptomycin (100 μg/mL) (Life Technologies-GibcoCRL, Milan, Italy) (complete medium), seeded in 48-well culture dishes with LPS (100 ng/mL) or PGN (10 μg/mL) for 6 hours. At the end, cells were used to extract RNA.

Peritoneal F4/80⁺ cells isolated as indicated above were resuspended in RPMI 1640 medium and cultured (0.5 × 10⁶/mL) with or without IL-25 (50 ng/mL) and/or IL-4 (50 ng/mL) and/or IL-13 (10 ng/mL) or IL-21 (50 ng/mL) (all from R&D Systems) for 3–24 hours. In some experiments, murine bone marrow-derived macrophages (BMMs) were prepared by standard procedures using anti-F4/80-coated beads, cultured (1 × 10⁶/mL) in DMEM + 10% FBS + 10 ng/mL GM-CSF (R&D Systems) and stimulated with or without the above cytokines for 24 hours.

Histopathological Analysis and Immunofluorescence

Colonic frozen sections were stained with hematoxylin and eosin (H&E) or antimouse F4/80 antibody (Biolegend, San Diego, CA) followed by incubation with a highly sensitive biotinylated secondary antibody (Dako, Glostrup, Denmark) and the tyramide signal amplification kit (Perkin Elmer, Waltham, MA). For TNBS-induced colitis, the colitis score was calculated using a previously validated method in which the degree of inflammation on microscopic cross-sections of the colon can grade semiquantitatively from 0 to 5.^21,22

RNA Extraction, cDNA Preparation, and Real-time PCR

RNA was extracted using the TRIzol reagent according to the manufacturer's instructions (Invitrogen). A constant amount of RNA (500 ng/sample) was retrotranscribed into complementary DNA (cDNA), and 1 μL of cDNA/sample was then amplified using the following conditions: denaturation 1 minute at 95°C, annealing 30 seconds at 58°C for Arg1, Fizz1, YM1, IL-6, IL-12p35, IL-12p40, and TNF-α and at 60°C for INOS and β-actin, followed by 30 seconds of extension at 72°C. Primer sequences were as follows: Arg1 sense: 5-GTC-TGG-CAG-TTG-GAA-GCAT-C-3; Arg1 antisense: 5-TGG-TTG-TCAG-GGG-AGT-GTT-G-3; Fizz1 sense: 5-CCA-ATC-CAG-CTA-ACT-ATC-CC-3; Fizz1 antisense: 5-TGG-TCC-AGT-CAA-CGA-GTA-AG-3; YM1 sense: 5-CAA-GGC-TGC-TAC-TCA-CTT-C-3; YM1 antisense: 5-CAG-CAC-TCT-TTC-CAA-TGT-C-3; INOS sense: 5′-AAG-GCT-CTG-TTC-TGT-TAG-GC-3′; INOS antisense: 5′-CTT-CTG-CTC-CAA-ATC-CAA-CG-3′; TNF-α sense: 5′-ACC-CTC-ACA-CTC-AGA-TCA-TC-3′; TNF-α antisense: 5′-GAG-TAG-ACA-AGG-TAC-AAC-CC-3′; IL-6 sense: 5′-AGC-CAG-AGT-CCT-TCA-GAG-AG-3′; IL-6 antisense: 5′-GAT-GGT-CTT-GGT-CCT-TAG-CC-3′; IL-4 sense: 5′-ACA-GGA-GAA-GGG-ACG-CCAT-3′; IL-4 antisense: 5′-AAG-CAC-CTT-GGA-AGC-CCT-AC-3′; IL-13 sense: 5′-GAG-CAA-CAT-CAC-ACA-AGA-CC-3′; IL-13 antisense: 5′-AAT-CCA-GGG-CTA-CAC-AGA-AC-3′; IL-12p35 and IL-12p40 were evaluated using commercially available TaqMan probes (Applied Biosystems, Foster City, CA). β-Actin (sense: 5-AAG-ATGCCC-AGA-TCA-TGT-TTG-AGA-CC-3; antisense: 5-AGC-CAG-GTC-CAGACG-CAG-GAT-3) was used as an internal control. Real-time PCR was performed using the IQ SYBR Green Supermix (Bio-Rad Laboratories, Milan, Italy). Gene expression was calculated using the ddCt algorithm.

Flow Cytometry

To characterize peritoneal and colonic lamina propria cell infiltrates, freshly obtained peritoneal cells and LPMC samples were preincubated with Mouse BD Fc Block purified antimouse CD16/CD32 mAb 2.4G2 (Becton Dickinson, Milan, Italy) at 4°C for 15 minutes, followed by incubation for 30 minutes at 4°C with the following monoclonal antibodies: CD11b FITC (1:50 final dilution, Becton Dickinson), CD4 PeRCP (1:50 final dilution, Becton Dickinson), CD8 FITC (1:50 final dilution, Becton Dickinson), DX5 PE (1:50 final dilution, Becton Dickinson), F4/80 APC (1:50, final dilution, Invitrogen), CD206 FITC (1:50 final dilution, eBioscience, Clone Number MR5D3, Frankfurt, Germany), and isotype control IgGs (1:50 final dilution, Becton Dickinson). Data were acquired with FACSCalibur and analyzed using CellQuest software (BD PharMingen, Milan, Italy).

Statistical Analysis

Student's t-test, analysis of variance (ANOVA) test, or nonparametric Kruskal–Wallis test were used to calculate statistical significance between groups.

Results

Intraperitoneal Administration of IL-25 to Mice Enhances Peritoneal Infiltration of CD11b-expressing F4/80⁺ Cells and Increases the Gene Expression of AAMs

In initial studies we characterized the peritoneal cell infiltrate of mice injected IP with IL-25 or PBS. The percentages of CD4⁺, CD8⁺, and DX5⁺ cells did not significantly change over time following IL-25 treatment (Fig. 1a–c). By contrast, IL-25 administration led to a marked increase in the fraction of CD11b-expressing F4/80⁺ cells, which was evident at 72 hours and became statistically significant at 96 hours (Fig. 1d).

Figure 1.

IL-25 enhances peritoneal infiltration of macrophages and enhances the expression of AAM-related markers. (a–d) Mice were injected with PBS (control, CTR) or IL-25 (10 μg/mouse) and sacrificed at the indicated timepoints. Total intraperitoneal cells were then isolated and characterized for the expression of CD4, CD8, DX5, and CD11b/F4/80 by flow cytometry. Numbers indicate the percentages of positive cells and are mean ± SEM of three experiments analyzing in total samples from nine mice per group (control vs. IL-25-treated mice, *P = 0.01). (e) Percentages of CD206-expressing F4/80⁺ cells in intraperitoneal cell samples isolated from control (CTR) or IL-25-treated mice as described in (a–d). Data indicate mean ± SEM of three experiments in which nine mice per group were used (control vs. IL-25 treated mice, *P = 0.01). Right inset: representative dotplots showing the percentage of CD206-expressing F4/80⁺ cells in intraperitoneal cell samples isolated from mice treated as indicated in (a–d). Numbers indicate the percentages of positive cells in the selected areas. One of three representative experiments in which similar results were obtained is shown. (f,g) IL-25 enhances RNA expression of AAM-related markers. Mice were injected with PBS or IL-25 (10 μg/mouse). After 72 hours mice were sacrificed and total or F4/80⁺ IP cells were isolated and analyzed for Arg1, Fizz1, Ym1, and INOS RNA expression by real-time PCR. Levels are normalized to β-actin. Data indicate the mean ± SD of three independent experiments analyzing in total samples from nine mice per group (note the log scale of the y-axis; control vs. IL-25-treated mice, *P < 0.03).

To examine whether IL-25 enhances the expression of AAM-related markers, intraperitoneal cells were also stained with F4/80 and CD206, the mannose receptor, a specific marker of AAMs.²³ IL-25-treated mice exhibited a significant increase in the fraction of F4/80-positive cells expressing CD206 as compared to mice treated with PBS (Fig. 1e). Moreover, we collected intraperitoneal cells from the two groups of mice at day 3 and analyzed the expression of additional markers of AAMs, such as Arg1, Fizz1, and YM1 by real-time PCR. The same samples were analyzed for the expression of INOS, a marker of CAMs. Arg1, Fizz1, and YM1 RNA expression was significantly increased in IL-25-treated mice as compared to control mice, while the expression of INOS remained unchanged (Fig. 1f). To exclude that the higher expression of AAM genes seen in IL-25-treated mice was due to differences in the number of macrophages infiltrating the peritoneum, the content of transcripts for these markers was also analyzed in F4/80⁺ cells isolated at day 3. A marked increase in AAM-related markers was seen in F4/80⁺ cells isolated from IL-25-treated mice (Fig. 1g).

Intraperitoneal Administration of IL-25 Leads to Enhanced Expression of AAM-related Markers in the Colon

To determine whether IP administration of IL-25 altered the intestinal lamina propria cell infiltrate, we isolated colonic LPMC from mice treated with IL-25 or PBS and analyzed them by flow cytometry. No significant change in the fraction of T (both CD4 and CD8) cells, NK cells, and CD11b-expressing F4/80⁺ cells was seen following IL-25 administration (Fig. 2a–d). To confirm these data, F4/80⁺ cells were also analyzed by immunofluorescence. The representative pictures shown in Figure 2e show that IL-25 did not alter the intestinal accumulation of F4/80⁺ cells. By contrast, whole colonic explants and F4/80⁺ LPMC isolated from the colons of IL-25-treated mice exhibited a significant increase in the expression of AAM-related markers while INOS remained unchanged (Fig. 2f,g).

Figure 2.

Mice injected IP with IL-25 exhibit enhanced colonic expression of AAM-related markers. (a–d) Percentages of CD4⁺, CD8⁺, DX5⁺, and CD11b/F4/80⁺ cells in LPMC isolated from the colons of naïve mice treated with PBS (CTR) or IL-25 (10 μg/mouse) and sacrificed at the indicated timepoints. Analysis was performed by flow cytometry. Data indicate mean ± SEM of three experiments analyzing in total at least nine mice per group. No statistical difference was observed in the two groups of mice. (e) Representative immunofluorescence pictures showing F4/80⁺ cells in colonic sections obtained from control or IL-25-treated mice. Mice were treated as indicated above and sacrificed at 72 hours (original magnification, ×40 or ×400 in the upper left inset). (f,g) Mice were injected with PBS or IL-25 (10 μg/mouse). After 72 hours, whole mucosal explants and colonic F4/80⁺ LPMC were analyzed for Arg1, Fizz1, Ym1, and INOS RNA transcripts by real-time PCR. Levels are normalized to β-actin and indicate the mean ± SD of three independent experiments analyzing in total cell samples of at least nine mice per group (control vs. IL-25-treated mice, *P < 0.02, **P < 0.04). [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

Since we previously showed that IL-25 prevents and cures experimental colitis when injected IP into mice,⁸ we next determined whether the therapeutic effect of IL-25 on TNBS colitis was associated with enhanced expression of AAM genes in the colon. As expected, mice with TNBS colitis and receiving a single IP administration of PBS exhibited significant weight loss from the first day of TNBS treatment (Fig. 3a). At day 4, these mice had lost more than 30% of their initial body weight (Fig. 3a), and there was nearly 50% of mortality (Fig. 3b, left panel). By contrast, mice treated with IL-25 regained weight (Fig. 3a) and manifested less mortality (Fig. 3b, left panel). Consistently, histological examinations of colonic tissues as well as blinded histological scoring of colitis in the different groups were significantly reduced in IL-25-treated mice as compared to PBS-treated mice (Fig. 3b, right panel).

Figure 3.

Enhanced expression of AAM genes in the colons of IL-25-treated mice during TNBS-colitis. (a) IL-25 attenuates TNBS-mediated colitis. Mice were injected IP with PBS or IL-25 (10 μg/mouse) 24 hours after the TNBS administration and sacrificed at day 4. Body weight was recorded daily; each point represents the cumulative mean weight ± SD of five animals per group. Data refer to three independent experiments (TNBS+PBS-treated mice vs. TNBS+IL-25-treated mice, *P = 0.04, **P = 0.01). (b) Percent of survival (left panel) and histological scoring (right panel) of mice treated as indicated in (a). Data indicate mean ± SEM of three separate experiments in which at least 15 mice per group were used *P = 0.02, **P = 0.03. (c) IL-25 enhances RNA expression of AAM-related markers. Mice were treated as indicated in (a). Four days after TNBS administration, whole mucosal explants and colonic F4/80⁺ LPMC were analyzed for Arg1, Fizz1, and Ym1 RNA expression by real-time PCR. Levels are normalized to β-actin and indicate the mean ± SD of three separate experiments in which at least 15 mice per group were used (TNBS+PBS-treated mice vs. TNBS+IL-25-treated mice, *P < 0.02). (e) IL-25 administration increases the percentage of CD206-expressing F4/80⁺ LPMC in mice with TNBS-colitis. Representative dotplots showing the percentage of CD206 and/or F4/80-expressing cells in total LPMC isolated from mice treated as indicated in (a). Numbers in the quadrants indicate the percentages of positive cells One of three representative experiments in which similar results were obtained is shown.

Next colonic mucosal explants and F4/80⁺ LPMC of IL-25-treated mice with TNBS colitis were analyzed for the expression of AAM-related markers by real-time PCR. Mice receiving IL-25 exhibited a significant increase in the expression of Arg1, Fizz1, and YM1 as compared to mice with TNBS colitis and treated with PBS (Fig. 3c,d). These findings were substantiated by the demonstration that the percentage of CD206-expressing F4/80⁺ LPMC was increased in IL-25-treated mice (Fig. 3e).

Overall, these results indicate that IP administration of IL-25 leads to a marked increase in the colonic expression of AAM genes.

IL-25 Renders Colonic LPMC Unresponsive to Bacterial Stimulation in Terms of Inflammatory Cytokine Production

Unlike CAMs, AAMs do not produce inflammatory cytokines following bacterial stimulation.²⁴ To confirm that the in vivo IL-25-induced macrophage response is skewed towards a regulatory phenotype, LPMC were isolated from the colons of mice with TNBS colitis that had been given IL-25 or PBS the day after the induction of colitis and cultured with PGN or LPS. After 6 hours, the RNA transcripts for IL-12 p40 and p35 subunits, TNF-α and IL-6, were evaluated by real-time PCR. LPMC isolated from mice with TNBS colitis treated with IL-25 showed significantly reduced expression of the cytokines in response to LPS and PGN as compared with cells isolated from animals treated with PBS (Fig. 4a–d).

Figure 4.

IL-25 administration reduces the cytokine response of colonic LPMC to LPS and PGN in vitro. LPMC were isolated from the colons of mice with TNBS colitis, receiving PBS or IL-25 the day after induction of colitis, and stimulated in vitro with PGN or LPS for 6 hours. RNA expression for IL-12p40, IL-12p35, TNF-α, and IL-6 was evaluated by real-time PCR. Levels are normalized to β-actin and indicate the mean ± SD of all experiments. *P = 0.03, **P = 0.04.

Transfer of IL-25-induced AAMs to Mice Attenuates TNBS Colitis

To prove that in vivo in mice IL-25 promotes an immune response characterized by the presence of macrophages with antiinflammatory properties, mice were injected IP with IL-25 or PBS and 3 days later sacrificed. Peritoneal F4/80⁺ cells were sorted, labeled with the fluorescent dye CM-Dil, and injected IP into untreated mice. TNBS colitis was induced in these mice 2 days after F4/80⁺ cell administration. By flow cytometry, we initially showed that F4/80⁺ cells injected IP migrated into the mesenteric lymph nodes and colonic lamina propria regardless of whether these cells had been primed or not in vivo with IL-25 (Fig. 5a). Interestingly, transfer of IL-25-primed but not control (PBS) F4/80⁺ cells to mice was associated with a significant attenuation of TNBS colitis, as substantiated by histological analysis of colonic sections (Fig. 5b,c). Consistently, mice with TNBS colitis, which had received F4/80⁺ cells from IL-25-treated mice, exhibited diminished expression of IL-12 (p40 and p35), IL-6, TNF-α, and INOS (Fig. 5d). In contrast, colitic mice injected IP with IL-25-primed F4/80⁺ cells had elevated levels of Arg1, Fizz1, and YM1, and this was evident in both colonic tissue and F4/80⁺ LPMC samples (Fig. 5e,f).

Figure 5.

(a) Flow cytometric analysis of F4/80⁺ cells isolated from the peritoneum of control or IL-25-treated mice, either left unlabeled or labeled with CM-Dil, and injected IP into naïve mice. Recipient mice were sacrificed after 72 hours and cells were isolated from colonic lamina propria (LPMC), mesenteric lymph nodes (MLNs) and peritoneal cavity (IPC). Representative dotplots show the percentages of unlabeled and CM-Dil-labeled cells in the various compartments. One of three representative experiments in which similar results were obtained is shown. (b) H&E-stained-colonic sections taken from control (ETOH) or colitic mice injected IP with control or IL-25-primed F4/80⁺ cells. Mice were sacrificed at day 4. (c) Histological scoring of colonic sections in each group. Data indicate the mean ± SD of all experiments in which at least 14 mice for group were considered (* P = 0.02). (d) Mice with TNBS-colitis which had been injected with IL-25-primed F4/80⁺ cells exhibit a diminished expression of INOS and proinflammatory cytokines. Mice were sacrificed at day 3 following TNBS administration and colonic samples were analyzed for IL-12p40, IL-12p35, TNF-α, IL-6, and INOS RNA expression by real-time PCR. Levels are normalized to β-actin and indicate the mean ± SD of all experiments (mice transferred with PBS primed F4/80⁺ cells vs. mice transferred with IL-25 primed F4/80⁺ cells; *P < 0.01). (e,f) Mice with TNBS colitis that had been injected with IL-25-primed F4/80⁺ cells have enhanced expression of AAM-related markers. Studies were conducted as indicated in (b) and RNA transcripts for Arg1, Fizz1, and Ym1 were evaluated by real-time PCR in whole mucosal explants (e) and colonic F4/80⁺ LPMC (f). Levels are normalized to β-actin and indicate the mean ± SD of three separate experiments in which at least four mice per group were used (note the log scale of the y-axis; mice transferred with PBS-primed F4/80⁺ cells vs. mice transferred with IL-25-primed F4/80⁺ cells; *P < 0.03). [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

IL-25 Does Not Directly Induce AAMs In Vitro

To determine whether IL-25 induces directly the differentiation of AAMs, we isolated peritoneal F4/80⁺ cells from untreated mice, cultured them with IL-25, IL-4, or IL-13 for different times, and then analyzed AAM-related markers by real-time PCR. No significant increase in AAM-associated markers was seen in IL-25-stimulated F4/80⁺ cells. In the same experiments IL-4 and IL-13, both used as a positive control for the induction of AAMs, increased the RNA expression of Arg1, Fizz1, and YM-1 (Fig. 6a–c). Consistently, flow cytometric analysis of peritoneal cells isolated from naïve mice and cultured in vitro with Th2 cytokines showed that IL-4 and IL-13, but not IL-25, enhanced the fraction of CD206-expressing F4/80⁺ cells (Fig. 6d). Moreover, IL-25 did not potentate the IL-4 or IL-13-induced AAM gene expression (Supporting Fig. 1). In the same cell cultures, no increase in AAM gene expression was seen following IL-21 stimulation (Supporting Fig. 2). Finally, we showed that IL-4 and IL-13, but neither IL-21 nor IL-25, enhance AAM gene expression in BMM cultures (Supporting Fig. 3)

Figure 6.

IL-25 fails to promote the in vitro differentiation of AAMs. (a) Transcripts for Arg1, Fizz1, and YM1 were evaluated in F4/80⁺ cells isolated from the peritoneum of naïve mice and stimulated in vitro with IL-25, IL-4, or IL-13 for 3 hours (a), 6 hours (b), or 18 hours (c). Levels are normalized to β-actin and indicate the mean ± SD of three independent experiments (*P = 0.03; **P = 0.04). (d) F4/80⁺ cells isolated from the peritoneum of naïve mice were cultured in vitro with the indicated cytokines for 48 and 72 hours. The percentages of CD206-expressing F4/80⁺ cells were then evaluated by flow cytometry. Numbers indicate the percentages of positive cells in the selected areas. One of three representative experiments in which similar results were obtained is shown.

Blockade of IL-4, IL-13, and TGF-β Does Not Inhibit the In Vivo Stimulatory Effect of IL-25 on AAM Gene Expression

Previous studies have shown that overexpression of IL-25 in mice results in enhanced production of IL-4 and IL-13.¹⁰ Thus, we assessed whether these two cytokines were involved in the IL-25-mediated upregulation of AAM genes. As expected, peritoneal cells isolated from mice receiving IL-25 exhibited enhanced IL-4 and IL-13 RNA expression (Fig. 7a). However, mice injected with a neutralizing IL-4 and/or IL-13 antibody exhibited no decrease in the IL-25-induced RNA expression of AAM-related markers as compared to mice treated with a control antibody (Fig. 7b). Both antibodies, however, were active, as in the same experiments they significantly reduced the expression of IL-4 and IL-13, respectively (Supporting Fig.4).

Figure 7.

IL-25 enhances IL-4 and IL-13 RNA expression. (a) Mice were injected with PBS or IL-25 (10 μg/mouse) and sacrificed at the indicated timepoints. Total intraperitoneal cells were isolated from control (CTR) or IL-25-treated mice. IL-13 and IL-4 RNA expression was analyzed by real-time PCR. Data indicate mean ± SEM of all experiments (control vs. IL-25-treated mice, *P = 0.01). (b) Administration of a neutralizing IL-4 and/or IL-13 antibody to mice did not affect the IL-25-induced RNA expression of AAM-related markers. Mice were injected IP with a neutralizing IL-4 and/or IL-13 antibody (500 μg/mouse) 6 hours before and 6 hours after IL-25 administration. At day 3, intraperitoneal F4/80⁺ cells were isolated and analyzed for Arg1, Fizz1, and Ym1 transcripts by real-time PCR. Levels are normalized to β-actin and indicate the mean ± SD of three independent experiments in which at least two mice per group were used (*P < 0.01).

TGF-β1 is also able to favor the differentiation of AAMs.²⁵ Therefore, we next assessed the involvement of this cytokine in the IL-25-induced AAM gene upregulation. F4/80+ cells isolated from the peritoneum of IL-25-treated mice contained elevated levels of TGF-β1 transcripts (Supporting Fig. 5a). However, administration of a neutralizing TGF-β antibody to these mice did not modify the IL-25-induced RNA expression of AAM-related markers as compared to mice treated with a control antibody (Supporting Fig. 5b).

Discussion

Initially considered as a factor that positively regulates Th2 cell-mediated immunity,^10,11 IL-25 has recently been shown to control additional immune pathways due to its ability to target several cell types. For example, IL-25 inhibits the synthesis of inflammatory cytokines by CAMs and, in vivo in mice, administration of IL-25 prevents and cures experimental colitides.^8,12 It has also been shown that IL-25 is constitutively produced in the human and mouse intestine and that its expression is markedly reduced in IBD.^8,12 Taken together, these findings support the antiinflammatory action of IL-25, and suggest that defects in IL-25 production can contribute to amplify and maintain pathogenic responses in the gut. In the present study we have begun to dissect mechanism(s) whereby IL-25 counterregulates inflammatory reactions in the gut. We focused our work on macrophages, because our previous study showed that these cells are one of the major targets of IL-25 in the gut.⁸ In particular, we aimed to determine whether the antiinflammatory effect of IL-25 was mediated by AAMs, because these cells are inducible by Th2 cytokines¹⁹ and are known to restrain the inflammatory function of CAMs and favor the resolution of detrimental immune reactions.^14,15

By analyzing a panel of molecules expressed selectively by AAMs, we initially showed that, in vivo in mice, IL-25 enhanced the expression of AAM-related markers in peritoneal and colonic F4/80⁺ cells. This later finding occurred also in mice with TNBS colitis, and paralleled the IL-25-mediated attenuation of colitis. While our study was ongoing, it was shown that infection of mice with the tape-worm Hymenolepis diminuta increases markers indicative of AAM differentiation, and that in vitro-induced AAMs reduce colonic inflammation when administered to mice.¹⁷ Since there is no technical procedure to selectively deplete AAMs, we could not definitively prove that AAMs mediate the anticolitic effect of IL-25. We preliminarily adopted the clodronate-liposome technique to deplete AAMs, but mice treated with clodronate-liposomes developed a severe and often lethal TNBS colitis regardless of whether they received IL-25 (unpubl. obs.). This may be due to the possibility that clodronate-liposomes may kill mucosal cell types (e.g., phagocytes and neutrophils), which are necessary to counter-regulate inflammation. To circumvent these difficulties, we used an alternative approach and determined whether F4/80⁺ cells isolated from the peritoneum of IL-25-treated mice and transferred to recipient mice attenuated the severity of TNBS colitis. Initially we showed that fluorescence-labeled peritoneal F4/80⁺ cells isolated from IL-25 and control-treated mice migrated equally into the colon of mice following IP administration, thus indicating that IL-25 does not alter the migratory capacity of F4/80⁺ cells. Interestingly, transfer of cells isolated from IL-25-treated mice inhibited the ongoing TNBS colitis. Real-time PCR analysis of colonic samples taken from these mice showed that transfer of IL-25-primed F4/80⁺ cells was associated with a marked reduction in the production of IL-12, IL-6, TNF-α, and INOS and an upregulation of AAM-associated markers, thus supporting the antiinflammatory role of IL-25-activated macrophages. Our data are consistent with the demonstration that TNBS colitis is driven by bacteria-induced activation of CAMs and that treatment of mice with compounds interfering with the inflammatory function of monocytes/macrophages results in disappearance of the intestinal inflammatory lesion.^26,27

In a final set of functional studies, we determined whether IL-25 was able to orchestrate directly the differentiation of AAMs. Stimulation of peritoneal and bone-marrow F4/80⁺ cells with IL-25 did not enhance AAM gene expression, thus confirming our recent study showing that IL-25 is unable to induce human blood monocytes to differentiate in AAMs.²⁸ These findings, however, are not surprising, as it has previously been reported that other cytokines, like IL-21, can favor the in vivo expansion of AAMs despite their inability to induce AAM genes in cultured macrophages.²⁹ We also showed that IL-4 and IL-13 enhanced AAM-related markers in peritoneal F4/80⁺ cells and that both these cytokines were upregulated in peritoneal cells of IL-25-treated mice. However, blockade of IL-4 and/or IL-13 with neutralizing antibodies did not affect the ability of IL-25 to induce AAMs in vivo, arguing against a role for these cytokines in the IL-25-mediated AAM expansion. Moreover, our data would seem to suggest that the stimulatory effect of IL-25 on AAM genes is not mediated by TGF-β1, since blockade of this cytokine with a neutralizing antibody did not inhibit the IL-25-mediated induction of Arg1, Fizz1, and YM1. Thus, the critical question remains as to how IL-25 promotes the peritoneal and colonic expansion of AAMs in mice. Studies with labeled macrophages revealed that a single IP administration of IL-25 to mice did not enhance the migration of such cells to the colon. Nonetheless, we cannot exclude the possibility that IL-25 may selectively promote the recruitment of AAMs in both the peritoneum and colon, perhaps by regulating the synthesis of chemokines, which preferentially attract AAMs. Studies are now in progress to address this issue.

Little is known of the frequency and functional activity of AAMs in human immune-mediated pathologies, including IBD. Hunter et al¹⁷ have recently shown that patients with active CD have reduced numbers of CD68⁺CD206⁺ cells in the gut as compared to controls. This finding fits well with the recent demonstration that IL-25 expression is downregulated in IBD⁸ and the observation made in this study that IL-25 is an important inducer of AAM genes in vivo. Therefore, it would be relevant to establish whether enhancing IL-25 expression or activity in IBD tissue associates with increased expression of AAM markers and attenuation of inflammatory pathways.

In conclusion, the present study shows that, in vivo in mice, IL-25 promotes an immune response that enhances the expression of AAM genes with the downstream consequence of attenuating the ongoing inflammation in the gut.

Supporting Information

Additional Supporting Information may be found in the online version of this article.

Please note: Wiley-Blackwell are not responsible for the content or functionality of any supporting materials supplied by the authors. Any queries (other than missing material) should be directed to the corresponding author for the article.

Author notes

Reprints: Giovanni Monteleone, Dipartimento di Medicina Interna, Università Tor Vergata, Via Montpellier 1, 00133 Rome, Italy e-mail: )

Supported by the “Fondazione Umberto di Mario,” Rome, the Broad Medical Research Program Foundation (No. IBD-0242), and Giuliani SpA, Milan, Italy.

Declaration: G.M. has filed a patent entitled “A treatment for inflammatory diseases”“ (patent No. 08154101.3), while the remaining authors have no conflicts of interest to disclose.