IL-33 Contributes to Schistosoma japonicum-induced Hepatic Pathology through Induction of M2 Macrophages

Interleukin (IL)-33 is involved in T helper (Th)2-biased immune responses in mice infected with Schistosoma, but the precise mechanism remains to be elucidated. Herein, we investigated the role of IL-33 and its receptor ST2L in hepatic granuloma pathology induced by Schistosoma japonicum infection. We found that IL-33 induced the increased production of IL-5 and IL-13 from splenocytes and liver mononuclear cells (MNCs) of infected mice. The infected mice developed significantly higher number of ST2L-expressing cells in spleen and liver. Most of the ST2L-expressing cells in liver were F4/80+ macrophages, indicating the key role of macrophages in the response to IL-33. However, the liver MNCs in male-only worm infection had a poor response to IL-33, though elevated serum IL-33 was observed. ST2L+F4/80+ cells were lower in male-only worm infection than that of mixed infection. IL-33 and soluble egg antigen (SEA) upregulated ST2L expression on macrophages in vitro and ST2L-expressing macrophage displayed MHCII-CD11b+M2 phenotype. Macrophage deletion significantly attenuated IL-33-induced type 2 immunity and egg granuloma formation during S. japonicum infection. These data demonstrate that IL-33 contributes to hepatic granuloma pathology through induction of M2 macrophages during S. japonicum infection.

IL-33, a ligand for the orphan IL-1 family receptor ST2L, has been implicated in initiating, amplifying, and maintaining Th2 responses and M2 macrophages [11][12][13] . IL-33 and ST2L are constitutively expressed in healthy liver and their expression levels are increased in hepatic ischemia/reperfusion 14 . Recently, Arshad et al. reported that IL-33 was constitutively expressed in liver vascular endothelial cells and sinusoidal endothelial cells 15 . It has been demonstrated that the hepatocyte is a major source of IL-33 in acute murine hepatitis 16,17 . However, little is known regarding the role of IL-33 in the formation of granuloma in the liver during S. japonicum infection.
We recently reported that mice infected with the nematode parasite, Angiostrongylus cantonensis, displayed increased expression of IL-33 18 . Given the importance of IL-33 in type 2 responses, the current study was undertaken to investigate the role of IL-33 in S. japonicum infection. Herein, we examined the role of IL-33 in regulating host immune responses and liver pathology during S. japonicum infection. We observed that infected mice developed significantly higher number of ST2L-expressing cells in spleen and liver. Most of the ST2L-expressing cells in liver were F4/80 + macrophages, indicating that macrophages are the dominant cell-type responsive to IL-33. However, the liver MNCs in mice with male-only worm infection displayed a poor response to IL-33, though elevated serum IL-33 was observed. The numbers of ST2L + F4/80 + cells were lower in male-only worm infections than in mixed infection. Additionally, IL-33 and soluble egg antigen (SEA) up-regulated ST2L on macrophages and ST2L + macrophage I -CD11b + M2 phenotype. Our studies further demonstrate that hepadisplayed an MHCItic granuloma formation was significantly attenuated in mice following macrophage depletion during S. japonicum infection.

S. japonicum infection drives IL-33 expression.
Previously, our studies strongly suggested that IL-33 was important for type 2 immune responses induced by parasites 18 . To investigate whether IL-33 is also involved in the pathogenesis of schistosomiasis, we measured serum levels of IL-33 obtained from patients infected with S. japonicum. As shown in Fig. 1a, significantly higher levels of IL-33 were observed in patients infected with S. japonicum compared to uninfected controls (p < 0.05).  20) or patients with S. japonicum infection (n = 23) was determined by ELISA (*p < 0.05). (b) Liver mononuclear cells (MNCs) from naive mice or from S. japonicum-infected mice at week 7 post-infection were activated with IL-33 (10 ng/ml) or ConA (5 μg/ml) for 72 hours. Supernatants were collected for IL-13 ELISA assay. (c) Splenocytes from naive mice or from S. japonicum-infected mice at week 7 postinfection were activated with SWA (50 μg/ml), SEA (20 μg/ml), IL-33 (10 ng/ml) or ConA (5 μg/ml) for 72 hours. Supernatants were collected for IL-13 and IL-5 ELISA. Data represent observations from 3 independent experiments and are shown as mean ± standard error of the mean (SEM), *p < 0.05 and **p < 0.01, by the Student t test.
Scientific RepoRts | 6:29844 | DOI: 10.1038/srep29844 Since IL-33 regulates type-2 cytokines 19 , we examined its role in modulating cytokine production in BALB/c mice infected with S. japonicum and euthanized at week 7 post-infection. Splenocytes were cultured in the presence or absence of schistosome worm antigen (SWA), schistosome egg antigen (SEA), IL-33 or Concanavalina (Con A) for 72 hours and supernatants assayed by ELISA. We found that IL-33 significantly increased levels of IL-5 and IL-13 in splenocytes obtained from infected mice (Fig. 1c), whereas splenocytes isolated from uninfected mice did not respond to IL-33 in vitro; neither IL-5 nor IL-13 were detected. Interestingly, IL-33 induced much higher levels of IL-5 and IL-13 from splenocytes compared to SWA and SEA. We also observed that there was a significant increase in IL-13 production by IL-33 stimulated liver cells in infected mice compared with liver cells obtained from uninfected mice (Fig. 1b).
IL-33 promotes type 2 cytokine production and liver pathology only in mixed male and female infection. We next determined the role of IL-33 in the formation of liver granuloma. BALB/c mice were infected with conventional mixed male and female S. japonicum where eggs were laid (mixed sex infection) or with a male-only infection which there were no eggs (male-only infection). Granuloma formation in the liver with mixed sex infection is shown in Fig. 2a. We then compared the expression level of IL-33 in serum from mixed sex infection and male-only infection mice at 4, 7, and 10 weeks post-infection. Although, a significant increase in IL-33 was detected at weeks 7 and 10 post infection (P < 0.01, as compared to week 4 post infection), there were no significant difference in the serum levels of IL-33 between mixed sex infection and male-only infection mice at week 7 post infection (Fig. 2b). However, IL-33 induced higher levels of IL-13 production in spleen and liver of mice with mixed infected than in male-only infection (Fig. 2b). Taken together, these observations suggest that SEA regulates the expression of the receptor of IL-33, although the expression of IL-33 itself is independent on SEA.
Macrophages are the major cell-type-responsive to IL-33 in the spleen and liver granuloma of S. japonicum-infected mice. IL-33 targets cells of immune system via ST2L 13 . In order to detect ST2L expression on cells from spleens and livers during S. japonicum infection at different time points, BALB/c mice were infected and euthanized at weeks 4, 7, and 10 post-infection, Splenocytes and liver MNCs were prepared for FACS analysis. We demonstrated that ST2L + cells were upregulated on both splenocytes and liver MNCs after infection (Fig. 3a). Since previous data demonstrated that T cells producing type 2 cytokines express ST2L on their surface both in vitro and ex vivo 20 , we first detected ST2L expression on CD4 + T cells during infection. However, we found that the frequency of ST2L-expressing CD4 + cells was increased in liver granuloma (10 ± 2% of CD4 + cells vs 0.6 ± 0.1% in control livers) with no significant upregulation in the spleen (data not shown). These observations indicate that the receptor of IL-33 was upregulated on cells other than CD4 + cells, The supernatants were collected and IL-13 production was detected by ELISA. Data indicate mean ± SEM of 3-5 mice per group; *p < 0.05 and **p < 0.01, by the Student t test.
which led us to further identify which cell population was the dominant IL-33-responsive cell in schistosome egg granuloma.
In order to define the cells in schistosome egg granulomas responsible for expressing ST2L, the infected BALB/c mice were euthanized at week 7 post infection and single-cell suspension of liver and spleen were stained with surface markers against macrophages, T cells and B cells as well as ST2L. We demonstrated that specific ST2L could be expressed in different cells. More than 75% of ST2L + cells were F4/80 + macrophages in liver granuloma, while 25% of ST2L-expressing cells were either T or B cells as well as non-T or B cells (Fig. 3b). More than 75% of ST2L-expressing cells were F4/80 + macrophages in spleen. These observations indicate that macrophages are the major source of ST2L in the spleen and liver granulomas and the type 2 cytokines induced by IL-33 are dependent on the macrophage during S. japonicum infection.
Expression of ST2L on macrophages in the liver from male-only infected mice was impaired. Schistosoma infection results in a prominent Th2 response when egg production is initiated 4 . We found that more macrophages were recruited to the livers in mixed infection than in male-only infection (Fig. 4a). Next, we compared ST2L-expressing macrophages in the livers of mixed infection with male-only infection. As shown in Fig. 4b, there were more ST2L + macrophages in the liver of mixed infection than male-only infection at weeks 4, 7, and 10 post infection. Using quantitative RT-PCR, we detected significantly increased expression of ST2L in the livers of mixed infected mice compared to male-only infected mice (Fig. 4c). Furthermore, we found that mRNA levels of Th2 cytokines (IL-13 and IL-10) were significantly higher in mixed infected mice than male-only infected mice (Fig. 4e). Conversely, the expression of IFN-γ, a Th1 cytokine, was lower in mixed infected mice than male-only infected mice at week 7 post-infection (Fig. 4e). Taken together, these data indicate that SEA enhances ST2L expression and promotes type 2 immune responses.
IL-33 enhances the polarization of macrophage toward M2 phenotype 12,21 . Consistent with these findings, the expression of Ym-1 and Fizz1, two molecules associated with M2, were much higher in livers of mixed infected mice than male-only infected mice. Higher CCL24 expression was also observed in livers of mixed infected mice compared to male-only infected mice (Fig. 4d).

Both SEA and IL-33 promote ST2L expression and enhance the polarization of AAM in vitro.
Given the well-established contribution of IL-33/ST2L to type 2 immune responses 12,18 and the expression of ST2L on macrophages 21 , we reasoned that IL-33 might be involved in AAM macrophage activation. To characterize the effect of IL-33 on macrophage activation, we prepared liver MNCs and cultured MNCs with IL-33(10 ng/ ml), SEA (20 μg/ml), and SWA (50 μg/ml) for three days before analyzing the number of ST2L-expressing macrophages. We demonstrated that both IL-33 and SEA caused an increase in the levels of ST2L expression on macrophage compared to SWA (Fig. 5a,b). We further examined the phenotype of ST2L + and ST2L − macrophage and found that ST2L + macrophage displayed significantly increased levels of CD11b and low levels of MHCII (Fig. 5c), displaying an M2 phenotype 22 . Macrophage depletion in vivo impairs IL-33-induced type 2 immunity. After establishing that macrophages were capable of responding to IL-33 and producing type 2 cytokines, we investigated if macrophages were essential for the in vivo induction of type 2 immunity in response to IL-33. Two groups of mice were treated with clodronate-containing liposomes to deplete macrophages or PBS as controls at 4-6 weeks after infection. Clodronate treatment decreased the number of tissue macrophages, as shown by the significantly decreased expression of F4/80 + cells in spleen. Correspondingly, the IL-33-induced IL-13 cytokine response in spleen (Fig. 6c) was significantly attenuated in mice treated with clodronate compared to PBS-treated controls, although, there was no significant difference in serum IL-33 levels between these two groups (data is not shown). In addition, there was a significant reduction in liver egg granuloma in mice treated with clodronate compared to PBS-treated controls (Fig. 6a,b). Interestingly, we found that ST2L + CD4 + cells were decreased in both spleen and liver of mice treated with clodronate compared to PBS-treated controls (Fig. 6d). Overall, these data indicate that macrophages are essential for the development of type 2 immunity induced by IL-33.

Discussion
IL-33 contributes to the initiation of type-2 immune responses following infection with several pathogens and allergens. In liver, IL-33 plays an important role in viral and immune cell mediated pathology 15,23 . In schistosomiasis, the immune response against schistosome eggs results in granuloma formation in liver and initiates the Th2-biased immunity that prevents acute mortality of host but ultimately causes liver fibrosis 4 . Here, we investigated the role of IL-33 during S. japonicum infection. We demonstrated increased levels of IL-33 in the serum in both infected human and mice, consistent with a recent study reporting an association of IL-33 overexpression with S. mansoni dependent liver damage 17 . Therefore, we examined the possibility that this cytokine also plays a role in regulating type 2 responses and/or alternative activation of macrophage and coincident development of egg granuloma pathology during S. japonicum infection.
We found that both liver MNCs and splenocytes from S. japonicum-infected mice became IL-33-responsive and induced the production of IL-5 and IL-13. Since IL-33 exerts its cytokine activity through a signaling receptor known as ST2L, we examined ST2L expression in liver using mixed and single male infection model. It was determined that ST2L + cells were upregulated on both splenocytes and liver MNCs during infection with mixed sex but not in mice with male-only infection, suggesting that higher ST2L was induced by SEA rather than by SWA.
IL-33 targets many immune cells involved in the pathogenesis of disease. Fort et al. described a population of non-T/non-B cells that released the type 2 cytokines IL-4, IL-5, and IL-13 in response to IL-25 24 . Later, several research groups 16,25,26 reported the same populations (Lin -sca1 + Thy1 + ST2L + ) but named them differently (namely "natural helper cells", "nuocytes", and "innate helper type 2 cells"). Previous studies showed that IL-33 also targets T cells to amplify Th2 cytokines. Additionally, the macrophage is one of IL-33-responsive cells that can induce the production of IL-13. All these IL-33-responsive cells express ST2L and produce IL-5 and IL-13 in response to IL-33. We explored ST2L expression during S. japonicum infection and found that the dominant cells which responded to IL-33 were macrophages though many cells (including T cells, B cells and non-T/non-B cells) in the liver also expressed ST2L and responded to IL-33. Consistent with high ST2L expression on macrophages, the liver MNCs produced higher level of IL-13 when there are a higher number of macrophages. Of interest, are the differences in the response to IL-33 between mixed and male only-infection. IL-33 protein expression levels were enhanced both after mixed infection and male-only infection while the liver cells of the latter displayed a poor response to IL-33 and a reduction in Th2 cytokine production. Comparison of ST2L expression in LMNC reveals that ST2L + macrophages are characteristic of the higher levels of IL-13 production during mixed infection. Thus, increased expression of ST2L expressed in the liver correlated with more severe type 2 inflammatory responses during S. japonicum infection.
Schistosome eggs induce AAM-rich granuloma and Th2-biased immunity that prevents acute mortality but ultimately causes liver fibrosis 5 . IFN-γ and TLR pathways orient macrophage function towards the M1 phenotype whereas IL-4 and IL-13 polarize macrophages function towards the M2 phenotype. Recently, several reports demonstrated that IL-33/ST2L also plays important roles in the amplification of M2 polarization 12,27 . Therefore, in the micro-environmental milieu is an important determinant of macrophage function during inflammation. We wondered whether SEA could polarize macrophage into AAM. Therefore, we cultured liver MNCs with IL-33 or SEA in vitro and found that both IL-33 and soluble egg antigen (SEA) increased the expression of ST2L on macrophages in contradistinction to SWA. Further analyses demonstrated that ST2L + macrophages displayed the MHCII -CD11b + M2 phenotype, indicating that IL-33 and SEA could enhance AAM polarization. Furthermore, more ST2L + macrophages were observed in the liver and spleen at week 7 than at week 4 post infection and there were more ST2L + macrophages in the liver of mixed infection than male-only infection at weeks 4, 7, and 10 post infection. Zhu et al. showed that SWA favors the generation of M1 macrophages, whereas SEA preferentially promotes M2-polarized phenotype during S. japonicum infection 28 . Taken together, SEA could upregulate ST2L expression and promotes macrophage responsive to IL-33 to produce IL-13. Both IL-13 and IL-33 further promote the polarization of macrophages toward M2 phenotype. Therefore, more and more ST2L + M2 macrophages appear in the liver. In addition, the deletion of macrophage in S. japonicum infected mice resulted in down-regulation of ST2L expression in T cells and decreased IL-13 production. These data indicate that IL-33 promotes type 2 immunity and helminth-induced hepatic granuloma pathology through induction of M2 macrophages. It should be noted that since SEA and IL-33 increase expression of ST2L it raises a question as to whether both have a synergistic or additive effect. Alternatively, SEA may induce IL-33 expression in hepatic cells and ST2L expression in macrophages. The delineation of these pathways remains to be seen.
In summary, we have demonstrated an important role for IL-33/ST2L-mediated type 2 immune responses and inflammation in the liver during S. japonicum infection. Selective ablation of IL-33 with a consequent diminution of inflammatory response in the liver may represent a potential therapeutic approach to hepatic granuloma pathology caused by S. japonicum.

Patients.
Human studies were carried out in accordance with The Code of Ethics of the World Medical Association (Declaration of Helsinki) and approved by the Human Research Ethics Committee of the Sun Yatsen University in China. The patients received an explanation regarding the scope of the study, such as objectives, procedures, and potential risks, and signed an informed consent statement before inclusion in the study. Sera were collected from patients with acute schistosomiasis who had contact with infected water, had eggs in the stool and a typical clinical manifestation.
Parasite preparation and infection of mice. BALB/c mice (6 to 8 weeks old) were obtained from the Center of Experimental Animals, Sun Yat-sen University (Guangzhou, China). Animal protocols were approved by the Animal Care and Use Committee of Sun Yat-sen University. All animal experiments were performed in accordance with Chinese animal protection laws and with permission from the Institutional Review Board. Oncomelania hupehensis harboring S. japonicum cercariae (Chinese mainland strain) were purchased from the Jiangsu Institute of Parasitic Diseases (Wuxi, China). Mice were percutaneously infected with male cercariae for worm-only infections or male and female cercariae for egg-laying infections by placing a glass slide carrying 20 ± 2 S. japonicum cercariae on their abdomen for 15 minutes. After 4, 7, and 10 weeks, mice were sacrificed to obtain their livers and spleen (n = 4-6/group). SWA and SEA were prepared and diluted in PBS (pH 7.2) to a final concentration of 10 mg/mL, as previously described 25,26 . The protein concentration of SEA and SWA was determined using a bicinchoninic acid (BCA) Protein Assay kit (BioRad, Richmond, CA). Liver MNCs were isolated from the livers of mice using modifications of procedures previously described 28 . Briefly, mouse livers were minced and re-suspended in PBS (pH 7.2) and centrifuged at 50 × g for 5 minutes. Supernatants containing liver MNCs were collected and washed in PBS. The cell pellets were re-suspended in 37% Percoll solution (Sigma-Aldrich) and gently overlaid onto 70% Percoll. After centrifuged (850 × g) with the off-brake setting for 30 minutes at room temperature, the inter-phase containing liver MNCs were collected and washed twice in PBS. These cells were subsequently used for stimulation and surface staining.
ELISA assay. Enzyme-linked immunosorbent assay (ELISA) was performed to measure the production of cytokines in cell culture supernatants and in the sera of mice. IL-5, IL-13, and IL-33 were measured by using ELISA kits (eBioscience, San Diego, CA) according to the manufacturer's instructions.
Histopathology. Formalin-fixed, paraffin-embedded liver tissues from uninfected controls and mice infected at 4, 7, and 10 weeks post infection were sectioned and stained with H&E to assess histopathology including granuloma formation.

Macrophage depletion.
Macrophage depletion studies were done in Balb/c mice using liposome-encapsulated clodronate injection as previously described 21,27 . Mice were percutaneously infected with male and female cercariae by placing a glass slide carrying 20 ± 2 S. japonicum cercariae on their abdomen for 15 minutes. From weeks 4 to 6, mice received 4 tail vein injections of 100 μL/10g body weight of 1 mg/ml PBS or clodronate (CLOD)-containing liposomes. Mice were sacrificed 24 hours after the fourth injection, at which time plasma, spleen and liver tissue were collected.

Statistical analysis.
All statistics were carried out using Prism (GraphPad Software), and P values were obtained by using two-tailed Student's t-test for two group comparisons. The data were expressed as mean ± SEM or mean ± SD. Values of p < 0.05 were considered as statistically significant.