Interleukin (IL)-33 is dispensable for Schistosoma mansoni worm maturation, egg excretion, and the maintenance of egg-induced pathology in intestines of mice


 Background: Schistosomes are trematode worms that dwell in their definitive host’s blood vessels, where females lay eggs that need to be eliminated in the environment with host excreta to maintain their life cycle. Both worms and eggs require type 2 immunity for their maturation and excretion, respectively. However, immune molecules that orchestrate such immunity remain unclear. IL-33 is one of the epithelium-derived cytokines that induce type 2 immunity in tissues. This study aimed at determining its role in the maturation, reproduction, and excretion of S. mansoni eggs, and in the maintenance of egg-induced pathology in the intestines of mice.Methods: Using S. mansoni-infected IL-33-deficient (IL-33-/-) and wild-type (WT) mice, worm morphology, reproduction, and egg excretion were studied at different time points of infection. IL-5 and IL-13 production in spleens and mesenteric lymph nodes were measured. Tissue histology was performed on the terminal ilea of non- and infected mice.Results: Morphology-wise, worms from IL-33-/- and WT mice at the fourth and sixth weeks of infection did not differ. The worms' reproduction, expressed as eggs per worm pair, as well as the excretion of eggs, expressed as the number of eggs in intestinal tissues, did not differ between IL-33-/- and WT mice. In the sixth week of infection, IL-33-/- mice presented impaired type 2 immunity in intestines, characterized by decreased production of IL-5 and IL-13 in mesenteric lymph nodes and fewer inflammatory infiltrates with fewer eosinophils in the ilea. Besides, there was no difference between IL-33-/- and WT mice in the levels of IL-25 and TSLP in intestinal tissues.Conclusions: Despite its ability to initiate type 2 immunity in tissues, IL-33 alone seems dispensable for S. mansoni maturation, reproduction, and egg excretion. The transient impairment of type 2 immunity observed in the intestines, but not spleens, highlights the importance of IL-33 over IL-25 and TSLP in initiating, but not maintaining, locally induced type 2 immunity in intestinal tissues in schistosome infection. Further studies are needed to decipher the role of each of them in schistosomiasis and clarify the possible interactions that might exist between them.


Background
Schistosomes are blood-dwelling trematode worms that affect over 250 million people in the world, of which 201.5 million live in sub-Saharan Africa (1). Of several schistosome species that exist, three, namely Schistosoma haematobium, S. japonicum and S. mansoni, are the main cause of schistosomiasis in humans (2). They cause urogenital and hepato-splenic schistosomiasis, respectively (2). S. haematobium worms live in perivesical vein plexuses, and S. japonicum and S. mansoni live in mesenteric veins where females lay hundreds to thousands of eggs per day (3). Half of the eggs are washed away to the liver by blood ow from the intestines, and most of the remaining are retained in intestinal tissues, of which only a few succeed in reaching the intestinal lumen (4,5) to be eliminated in the environment with host's feces.
By their excretory-secretory products (ESP) such as interleukin-4 (IL-4)-inducing principle of S. mansoni eggs (IPSE/α1) (6,7) and omega-1 (ω1) (8) from S. mansoni and their homologs from S. haematobium (9) and S. japonicum (10), tissue-trapped eggs elicit strong and vigorous type 2 cell-mediated immunity that induces perioval granuloma formation and leads to brosis (11,12), pathological characteristics of the patent schistosome infection. While this immune response is bene cial for the host, especially in the liver where it protects hepatic cells from toxic effects of egg-derived ESP, it also plays a major role in the development of liver pathology (13). In contrast, in addition to being protective for and yet smiting the host with the pathology, granulomas in intestines play a bene cial role for the parasite, as they were found to favor the escape of eggs from the host through the intestinal wall (14).
Eggs are not the sole inducers of type 2 immunity in schistosomiasis, as studies have reported type 2 immune responses during prepatent schistosomiasis infection before egg deposition by female worms begins (15,16). The type 2 immunity during the prepatent schistosome infection was later found to be essential for the maturation of the worms, as injection of IL-4, the T helper 2 (Th2) polarizing cytokine, in schistosome-infected recombination activating gene (RAG)-de cient mice, in which schistosome worms fail to mature and reproduce due to the lack of functional CD4 + T cells, restored their maturation and egg deposition (17).
Emerging evidence indicates that the induction of type 2 immunity in tissues is not solely dependent on IL-4 as through activation of group 2 innate lymphoid cells (ILC2), epithelium-derived cytokines IL-25, IL-33 and thymic stromal lymphopoietin (TSLP) also induce type 2 immunity. Activated ILC2, in turn, produce abundant amounts of type 2 effector cytokines IL-4, IL-5, IL-9 and IL-13 (18)(19)(20)(21) and, through the expression of class II major histocompatibility complex (MHC II) (22) and OX40L (23), interact with CD4 + T cells to potentiate such type 2 immune responses. Also, ILC2 were found to initiate the adaptive type 2 immunity in an IL-4-independent manner by inducing IL-13-dependent activation and migration of DCs to the draining lymph nodes where they polarize naïve CD4 + T cells into Th2 cells (24).
However, the role of ILC2-activating cytokines IL-25, IL-33 and TSLP in schistosomiasis remains less understood. Focusing on the liver pathogenesis during S. japonicum infection, two studies showed that IL-33 contributes to the pathology development via induction of type 2 immune responses in infected mice (25,26). Indeed, studies have shown that IL-33 plays a critical role in the development of liver pathology through ILC2-derived IL-13-induced activation of hepatic stellate cells (HSC) (27) and alternative activation of macrophages (M2) (26). Moreover, Yu et al. (25) found that injection of exogenous IL-33 into S. japonicum-infected mice led to increased worm burden at the sixth week of infection without affecting their fecundity, suggesting that IL-33 might play a role in the migration and maturation of schistosome worms. Whether endogenous IL-33 plays a role in schistosome maturation and reproduction is not known.
Because IL-33 is known to induce and/or amplify M2 polarization of macrophages (26,(28)(29)(30), known to be essential for the excretion of schistosome eggs (14), we thought that in addition to contributing to the maturation of schistosome worms through induction of type 2 immunity during prepatent schistosome infection, IL-33 may also play a critical role in the excretion of S. mansoni eggs across the intestinal tissues. We hypothesized that IL-33 de ciency will impair the maturation and reproduction of S. mansoni worms, and the expulsion of their eggs across the intestinal wall, leading to accumulation of eggs in intestinal tissues of IL-33 −/− mice (31,32), and that type 2 immunity would be impaired in the absence of IL-33. Here we show that IL-33 is dispensable for the maturation and reproduction of S. mansoni worms, as well as for the excretion of their eggs across the intestinal wall of infected mice. Also, our ndings support the idea that IL-33 might be most potent in initiating, but not maintaining, type 2 immunity in tissues. To maintain initiated type 2 immunity, IL-33 might need the synergy of IL-25 and TSLP and/or of CD4 + Th2-derived effector cytokines.

Methods
Parasite, mice, and infection A Puerto Rican strain of S. mansoni was maintained in the laboratory by passage between Biomphalaria glabrata snails and ICR mice. BALB/cCrSlc (hereinafter referred to as BALB/c) were purchased from Japan SLC (Hamamatsu, Shizuoka, Japan) and maintained in speci c pathogen-free conditions at Nagasaki University animal facilities. Kindly provided by Professor Satoshi Uematsu (Osaka City University Graduate School of Medicine, Osaka, Japan), IL-33 −/− mice on BALB/c background were bred and maintained in the same conditions as for WT BALB/c at Nagasaki University animal facilities. All mice were provided with water and food ad libitum. Female mice aged 8-12 weeks were subcutaneously infected (33) with 50 and 35 freshly shed S. mansoni cercariae, respectively for nine and 12 weeks. Mice were sacri ced every three weeks from week six, except for worm morphology assessment where mice were also sacri ced at week four of infection. To assess the production of IL-25, IL-33 and TSLP in intestinal tissues during S. mansoni infection, WT BALB/c mice were infected with S. mansoni cercariae as described above and sacri ced every week from week zero to week four, then every two weeks to week 12 of infection.

Worm Morphology And Number
Adult S. mansoni worms were obtained by portal vein perfusion, xed with 4% neutral buffered formalin (NBF) (34), and their morphology was assessed under the inverted light microscope at 40x magni cation and their number was counted. Brie y, the portal vein was cut at its base under the liver, then the left cardiac ventricle was perfused with 30 mL of saline citrate (7.5 g of sodium citrate and 8.5 g of sodium chloride in milliQ) (33), followed by perfusion with 30 mL of PBS. The mesenteric veins were thoroughly checked for manual retrieval of worms that failed to wash out during perfusion.

Tissue Egg And Eggs Per Worm Pair Numbers
Livers and intestines were harvested and digested with 4% potassium hydroxide (KOH) at 37 °C for 14 hours. Brie y, livers were weighed and digested with 10 mL of 4% KOH, and intestines were cleansed of fecal matters, opened longitudinally, thoroughly washed with PBS, weighed then digested as for livers (33). After digestion, samples were centrifuged for 5 minutes at 2000 rpm and room temperature. Eggs were counted in 50 µL of thoroughly mixed pellet suspension under the light microscope at 40x magni cation and related to the organ weight. The number of eggs per worm pair was obtained by dividing the total number of tissue eggs per mouse by the number of worm pairs from the same mouse.

Production of soluble egg antigen ( Sm SEA)
Frozen eggs were thawed on ice, resuspended in ice-cold PBS containing 10 µg/mL of Leupeptin, then homogenized on ice by using handheld sterile glass Te on homogenizer. The homogenate was subjected to ve cycles of freeze (-80 °C) and thaw (on ice), incubated at 4 °C overnight with rotation, then

Tissue Cytokines
Small intestines were harvested in ice-cold PBS. After removal of fecal matters, intestines were opened along their axis, abundantly washed with ice-cold PBS, cut into small pieces and homogenized in 3 mL of ice-cold HBSS (containing 10 µg/mL of leupeptin and 0.1 mM/mL of PMSF) with gentleMACS Octo Dissociator (Miltenyi Biotec GmbH, Bergisch Gladbach, Germany) using the Protein_01 program. The homogenates were centrifuged for 20 minutes at 20,000 g and 0 °C (35). The supernatants were aliquoted and stored at -30 °C until use. The concentrations of IL-25, IL-33, and TSLP were measured in the supernatants by ELISA according to the manufacturer's instructions (DuoSet ELISA, R&D Systems, Inc., Minneapolis, USA).

Histology
A one cm-long fragment of terminal ileum was cut from each mouse, close to the secum of both infected and non-infected mice, cleaned from fecal matters and xed in 10% NBF until use. Samples were sent to the Division of Cell Function Research Support, Biomedical Research Support Center at Nagasaki University School of Medicine, for tissue processing. Slide-embedded Hematoxylin and Eosin (H&E)stained tissue sections were scanned at 40x magni cation using Aperio CS2 Scanner (Leica Biosystems Imaging, Vista, California, USA), and digital images were analyzed using Aperio ImageScope version 12.4.3 software (Leica Biosystems Imaging, Vista, California, USA). The number and size of granuloma areas were, respectively, counted and measured. The intestinal wall thickness was measured at three different places. The abundance of eosinophils in in ammatory in ltrates was visually appreciated. All the measurement results were compared between IL-33 −/− and WT. Results IL-33 de ciency does not affect S. mansoni worm maturation, reproduction, and egg excretion Schistosome worms are characterized by their dependence on the host immune system, particularly type 2 immunity, for their maturation, reproduction, and egg excretion (14,17,34,36), pointing to the importance of type 2 immunity in the biology of schistosomes. Because IL-33 is known to induce type 2 immunity independently of IL-4 (37), and that a recent study (25) reported increased S. japonicum worm numbers after injection of exogenous IL-33 into infected mice, we thought that its de ciency might compromise the maturation and reproduction of S. mansoni worms. Thus, we compared the morphology and number of worm pairs between IL-33 −/− and WT mice. We infected IL-33 −/− and WT BALB/c mice with S. mansoni cercariae and sacri ced them at indicated post-infection time points (Fig. 1a). As shown in Fig. 1b, morphology-wise, there was no difference between worms recovered from IL-33 −/− and WT mice. Although the number of worm pairs was signi cantly higher in IL-33 −/− mice at the ninth week of infection (Fig. 1c), their fecundity, expressed by the number of eggs per worm pair, did not differ between IL-33 −/− and WT mice during the whole infection course (Fig. 1d).
It is known that IL-33 induces M2 polarization of macrophages (26,(28)(29)(30) and that M2 are essential for the excretion of schistosome eggs across intestinal tissues (14). It has also been reported that due to the impairment of their expulsion across the intestinal wall, more schistosome eggs accumulate in intestinal tissues (31,32). To verify whether IL-33 plays a critical role in the excretion of eggs, we compared the number of eggs in intestinal tissues between IL-33 −/− and WT every three weeks from week six to week 12 of infection. No statistical difference was found between both mouse genotypes in the number of tissue eggs in the small intestines (Fig. 1e), indicating that IL-33 may be dispensable for the excretion of schistosome eggs. Because several studies reported a pathogenic role for IL-33 in egg-induced liver pathology by an increased number of liver tissue eggs (25,26), we sought to see whether IL-33 de ciency would be associated with a decreased number of eggs in liver tissues of IL-33 −/− compared to WT mice.
Unexpectedly, we found no difference in liver egg numbers between both mouse genotypes (Fig. S1). Together, these data indicate that IL-33 is dispensable for the maturation and reproduction of S. mansoni worms, and the excretion of their eggs across the intestinal tissues. IL-33 de ciency is associated with transitory impairment of type 2 immunity in mesenteric lymph nodes of S. mansoniinfected mice Compared to IL-25 and TSLP, IL-33 is known to be stronger in inducing type 2 immunity through the activation of ILC2 and macrophages (38,39). Therefore, we sought to assess whether IL-33 de ciency would impair type 2 immunity in intestines. We isolated immune cells from MLNs of S. mansoni-infected IL-33 −/− and WT mice, stimulated them with SmSEA for 72 hours, and measured IL-5 and IL-13 cytokines by ELISA. As expected, IL-33 de ciency impaired the production of IL-5 and IL-13 in MLNs of infected mice in response to stimulation with SmSEA at six weeks of infection. However, this impairment was not sustained during the infection course as it disappeared in subsequent infection time points (Fig. 2a,b). To verify whether this impairment was limited to intestines or was systemic, we isolated immune cells from spleens of infected mice and measured IL-5 and IL-13 in the supernatants after 72 hours of stimulation with SmSEA. Surprisingly, we found that while IL-33 de ciency did not affect the production of IL-5 in spleens during the infection course (Fig. 2c), it was rather associated with increased production of IL-13 at six weeks of infection (Fig. 2d).

IL-33 de ciency transiently attenuated egg-induced pathology in intestines of S. mansoniinfected mice
Although schistosome worms also induce type 2 immunity (15,16), eggs remain the most potent inducers of type 2 immunity and the main cause of the pathology in the liver and intestines of infected de nitive hosts (11)(12)(13). Studies have reported a pathogenic role for IL-33 in liver pathology during schistosome infections (25)(26)(27). While none of them reported on the role of IL-33 in the intestinal pathology development, studies related to in ammatory bowel diseases reported controversial roles for IL-33 in the development and/or exacerbation of these diseases. Some of these studies reported a protective role for IL-33 (40), others incriminated it in the development or exacerbation of these diseases (41,42). Thus, we thought to see whether IL-33 de ciency would compromise the development of egginduced pathology in intestinal tissues of infected mice. As shown in Fig. 3a and b, IL-33 de ciency was transiently associated with attenuated type 2 in ammatory responses in terminal ilea of IL-33 −/− mice compared to WT mice, characterized by less in ltration of intestinal tissues by in ammatory cells and wall thickness similar to naïve mice at the sixth week of infection. Moreover, the in ammatory in ltrates contained fewer eosinophils in IL-33 −/− mice than in WT in the sixth week of infection (Fig. 3a). Both mouse genotypes did not differ in the granulomas number and areas (Fig. 3c,d). Together, these data suggest that IL-33 may be important in initiating but not maintaining type 2 immunity at mucosal barriers than systemically and that it is dispensable for the maintenance of schistosome egg-induced pathology in intestines, as due to persistence of egg-derived ESP which strongly induce type 2 immunity, alternative compensatory mechanisms might have been triggered to compensate its absence. These results prompted us to speculate that IL-25 and TSLP expression might be upregulated in this setting, to compensate for the absence of IL-33 in later infection time points in intestines.
There is no change of IL-25 and TSLP production in the absence of IL-33 in intestines of infected mice Individually or synergistically, IL-25, IL-33 and TSLP are known to induce tissue type 2 immune responses in different homeostatic and pathologic conditions (38,39,43,44). Besides, the existence of possible interactions between these cytokines was raised (39,44). We, therefore, reasoned that, due to IL-33 de ciency, there might be compensatory changes of IL-25 and/or TSLP production in S. mansoni-infected IL-33 −/− mice compared to WT mice. As shown in Fig. 4, there was no statistically signi cant difference in the levels of IL-25 and TSLP in the small intestinal tissues between IL-33 −/− and WT mice, indicating that there is no compensatory changes of IL-25 and TSLP production in the absence of IL-33. Although the levels of IL-25 and TSLP expression in intestinal tissue homogenates tended to increase with S. mansoni infection, compared to naïve mice, infected mice did not produce that much of these cytokines to reach a statistically signi cant difference (Fig. 4a,b).
Studies have reported an increase in IL-33 levels in the sera of individuals with S. japonicum infection compared to non-infected (26). In mice, this increase, which starts around week four of infection, reaches its peak around week eight (25), corresponding with the oviposition period. This may indicate that schistosome eggs are the major inducers of IL-33 release in schistosome infection settings. At the same time, because of its functional redundancy with IL-25 and TSLP (45), we sought to know the kinetics of these cytokines production during an S. mansoni infection. Thus, we infected only WT BALB/c mice with S. mansoni cercariae and checked for the release of IL-25, IL-33 and TSLP in their intestinal tissues. While the levels of IL-33 remained constantly higher even in naïve mice, levels of IL-25 and TSLP tended to increase with oviposition (Fig. S2), indicating that S. mansoni eggs may induce the release of IL-25 and TSLP but not of IL-33 in the intestines of infected mice.

Discussion
Studies have shown that both schistosome worms and eggs induce type 2 immunity, which is essential for their maturation, reproduction, and egg excretion (14)(15)(16)(17). De ciency in CD4 + Th2 cells and their effector cytokines IL-4 and IL-13 was shown to substantially decrease or completely abrogate egg excretion as a consequence of impaired worm maturation or failed signaling by type 2 effector cytokines (14,(46)(47)(48). Moreover, accumulating evidence suggests that ILC2 and their activating cytokines IL-25, IL-33 and TSLP induce adaptive type 2 immunity independently of IL-4 (22)(23)(24). Besides, while there is no report on the induction of IL-25, IL-33 and TSLP release by migrating schistosomula, it has been reported that schistosome eggs induce and/or enhance the production of IL-25 and IL-33 (49,50), indicating that schistosome eggs may be orchestrating their excretion by inducing the production of alarmin cytokines IL-25, IL-33 and TSLP to trigger type 2 immunity, essential to their excretion. In this study we report that IL-33 de ciency does not affect the maturation of worms as on both early (four weeks after infection) and late (six weeks after infection) worm maturation, the morphology did not differ between worms recovered from IL-33 −/− and WT mice, suggesting that IL-33 may not be required for schistosome worm maturation.
Consequently, despite a higher number of worm pairs at the ninth week of infection in IL-33 −/− mice, the number of eggs per worm pair, as well as the number of tissue eggs did not differ between the mouse genotypes. Although we did not determine the worms' lengths, the proportion of females in pairs (17,51), and the number of eggs in feces (14), based on the morphology of worms (34,52) and the intestinal tissue egg numbers (31,32) as indications for worm maturation and egg excretion, respectively; to the best of our knowledge, this is the rst study that has attempted to look at the role of IL-33, as a potent initiator of type 2 immunity necessary for schistosome worm maturation, in the maturation of S. mansoni worms and the excretion of their eggs.
In a study by Yu et al. (25), it was reported that the injection of S. japonicum-infected mice with exogenous IL-33 increased the number of worms recovered at the sixth week of infection, and exacerbated the liver pathology by increasing the number and size of liver granulomas. This may simply mean that as endogenous IL-33 plays a role in the development of egg-induced liver pathology (25)(26)(27), injecting exogenous IL-33 would exacerbate its pathogenic effects. In the present study, we could not nd any statistically signi cant difference in the number of eggs per worm pair between IL-33 −/− and WT mice. These results corroborate the ones reported by Yu et al. (25) as they did not nd a difference in the number of eggs per female worm. This indicates that IL-33 alone may have a negligible role to play in worm maturation, reproduction, and egg excretion.
While studies related to in ammatory bowel diseases reported controversial roles for IL-33 in the development and/or exacerbation of these diseases, with some reporting a protective role for IL-33 (30,40), and others incriminating it in the development or exacerbation of these diseases (41,42), to the best of our knowledge, no report had been made on the role of this cytokine in the intestinal pathology during schistosomiasis. Although IL-33 seemed dispensable for S. mansoni worm maturation and the excretion of their eggs, we sought to know whether it may play a signi cant role in the development of egg-induced pathology in the intestines of infected mice as it does in the liver (25)(26)(27). We found that the absence of IL-33 transiently impaired type 2 immunity in small intestines of IL-33 −/− mice, but not in their spleens, characterized by impaired production of IL-5 and IL-13 cytokines in MLNs in response to stimulation with SmSEA and attenuated egg-induced in ammation in IL-33 −/− mice ilea at the sixth week of infection. These results are in line with ndings by Vannella et al. (45) who, although focused on the role of alarmin cytokines IL-25, IL-33 and TSLP in the development and maintenance of type 2 cytokine-driven in ammation and brosis in lungs and liver, found that single ablation of these cytokines had no signi cant ameliorating effect on the liver pathology. However, when all three cytokines were ablated, a signi cant improvement of the pathology could be observed in the early phase of the infection, pointing towards the existence of functional redundancy between these cytokines. Findings from the present study tread in the same direction as the absence of IL-33 did not affect the pathology development, nor the number and size of granulomas in the intestines of IL-33 −/− mice in time points beyond the sixth week of infection. However, the difference between the study by Vannella et al. (45) and ours is that we started our observation at the sixth week of infection, when the egg-induced type 2 immunity is still at its start, while Vannella et al. (45) started their observation at the ninth week of infection when the egg-induced type 2 immunity has already reached its peak. We think that it could have been possible for them to notice a signi cant difference between IL-33 de cient mice and WT at an earlier stage of the infection, as observed in our study.
Despite its sharing of functional redundancy with IL-25 and TSLP (45), IL-33 remains the most potent of all three in inducing type 2 immunity (38,39,53). In addition to inducing type 2 immunity by itself, IL-33 can also potentiate the type 2 immunity induced by the two other cytokines, IL-25 and TSLP (54). Of all the cells that respond to IL-33, ILC2 and Th2 are the most important as through their production of abundant amounts of type 2 cytokines IL-4, IL-5 and IL-13, they play the most important role in cellmediated effector type 2 immunity (55,56), characterized by, among others, accumulation of M2 macrophages and eosinophils in affected tissues. Although dispersed in all tissues, ILC2 are more abundant in lungs and intestinal tissues (57), where they are the rst to be activated by IL-33 and migrate to local draining lymph nodes to initiate the adaptive type 2 immunity (24,58). Thus, it is understandable that the absence of IL-33 in IL-33 −/− mice at the early stage of the patent infection might have left them inactivated, leading to impaired type 2 immunity (58) as seen in the present study. In addition to acting through ILC2 and Th2 cells, IL-33 also acts directly on eosinophils, inducing their expansion and activation (59,60). Therefore, its absence in IL-33 −/− mice can explain the small number of eosinophils in the in ammatory in ltrates in the sixth week of infection (61). However, due to the persistence of eggderived ESP (6-10) as eggs keep accumulating in the tissues, and to the fact that IL-25 and TSLP can induce type 2 immunity independently of IL-33 (50,(62)(63)(64), alternative mechanisms leading to activation of both innate and adaptive type 2 immunity, including taking over of ILC2 activation by IL-25 and TSLP and Th2-dependent effector pathways, might have been activated to compensate the absence of IL-33. Together, these alternative mechanisms may have led to improved type 2 immunity in time points beyond the sixth week of infection.
Studies have pointed to the existence of possible interactions between IL-25, IL-33 and TSLP (39,44). In one study, anti-IL-33 treatment and TSLP receptor de ciency blocked the infection-induced expression of IL-25 in lung epithelial cells, and ex vivo treatment of ILC2 with TSLP increased their expression of IL-25 and IL-33 receptors (44). In another, it was noted that IL-25 shared with IL-33 many activities on macrophages without having additive effects, pointing toward the possible existence of common downstream signaling pathways for their biological activities (39). Therefore, we thought that IL-33 de ciency might be associated with a modi ed production of IL-25 and TSLP in the intestines of S. mansoni-infected IL-33 −/− mice. Our results showed no modi cation of intestinal production of IL-25 and TSLP as their levels in intestinal tissue homogenates did not differ between mouse genotypes, meaning that although they can, individually or synergistically, induce type 2 immunity, the absence of one may not affect the others in the schistosome infection settings or intestines. The nature and conditions of occurrence of the interactions between IL-25, IL-33 and TSLP pointed out by the above-mentioned studies (39,44) remain to be clari ed.
Studies in humans and mice have reported the increase of IL-33 levels in the sera of individuals and animals infected with S. japonicum infection (26). Also, these increased levels of IL-33 in serum peaked around the eighth week of infection in mice (25), relatively corresponding to the peak of egg-induced immune responses, suggesting that through their ESP, eggs may be the main inducers of IL-33 release in schistosome infections. Indeed, Hams et al. (49,50) reported that injection of S. mansoni eggs or recombinant form of their derived components, namely ω1, induced the production of IL-25 and IL-33, respectively in the lungs and fat tissue. Whether eggs in intestinal tissues induce the production of IL-33, IL-25 and TSLP is not known. By measuring the levels of these alarmin cytokines in the intestinal tissue homogenates during S. mansoni infection in WT BALB/c mice, we found that IL-33 levels remained constantly higher, even in non-infected mice. In contrast, IL-25 and TSLP levels uctuated over the infection course, peaking around the tenth week of infection, with TSLP having much lower levels than IL-25. The start of an increase in levels of IL-25 and TSLP tended to correspond to the oviposition, suggesting that this latter might be inducing the release of IL-25 and TSLP, but not of IL-33. In their study, Flamar et al. (58) recently reported that IL-33 expression was high in small intestines of naïve mice, corroborating our ndings. This indicates that IL-33 is constantly expressed in high amounts in mouse intestinal tissues.

Conclusion
To the best of our knowledge, this is the rst study that looked at the role of IL-33 in the maturation and reproduction of S. mansoni worms, as well as in the excretion of their eggs across intestinal tissues toward the lumen. It is also the rst study to report on the role that IL-33 may play in the maintenance of egg-induced type 2 immunity in intestines. It showed that IL-33 is dispensable for the maturation and reproduction of S. mansoni and the excretion of their eggs. Furthermore, due to transient impairment of type 2 immunity observed in the intestines but not spleens, this study highlights the importance of IL-33 over IL-25 and TSLP in initiating, but not maintaining, locally induced type 2 immunity in intestinal tissues in schistosome infections. These results corroborate previously reported ndings that IL-25, IL-33 and TSLP might be sharing a partial functional redundancy in their ability to maintain the tissue-induced type 2 immunity. Their combined or sequential ablation might be the best option to decipher the role of each of them in schistosomiasis and clarify the possible interactions that might exist between them.  The morphology of worms recovered from both mouse genotypes at weeks four (upper panels) and six (lower panels) was assessed under the microscope at 40x magni cation. Representative photographs are presented here. (c) The number of worm pairs from both IL-33-/-and WT mice was compared between the mouse genotypes. (d) The number of eggs per worm pair was determined as an indication of worm fecundity. Groups were compared using unpaired two-tail t-test with Welch's correction. (e) The number of eggs in intestinal tissue was used to indicate the egg excretion status (14,32). Experiments were replicated at least three times. Data are representative of 2 independent experiments with similar results and are presented as mean with SEM. Mann-Whitney test was used for the comparison of mouse groups at p<0.05 signi cance. wpi=week post-infection.

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download.