Dietary non-starch polysaccharides impair immunity to enteric nematode infection

Background The influence of diet on immune function and resistance to enteric infection and disease is becoming ever more established. Highly processed, refined diets can lead to inflammation and gut microbiome dysbiosis, whilst health-promoting dietary components such as phytonutrients and fermentable fibres are thought to promote a healthy microbiome and balanced mucosal immunity. Chicory (Cichorium intybus) is a leafy green vegetable rich in fibres and bioactive compounds that may promote gut health. Results Unexpectedly, we here show that incorporation of chicory into semisynthetic AIN93G diets renders mice susceptible to infection with enteric helminths. Mice fed a high level of chicory leaves (10% dry matter) had a more diverse gut microbiota, but a diminished type-2 immune response to infection with the intestinal roundworm Heligmosomoides polygyrus. Furthermore, the chicory-supplemented diet significantly increased burdens of the caecum-dwelling whipworm Trichuris muris, concomitant with a highly skewed type-1 immune environment in caecal tissue. The chicory-supplemented diet was rich in non-starch polysaccharides, particularly uronic acids (the monomeric constituents of pectin). In accordance, mice fed pectin-supplemented AIN93G diets had higher T. muris burdens and reduced IgE production and expression of genes involved in type-2 immunity. Importantly, treatment of pectin-fed mice with exogenous IL-25 restored type-2 responses and was sufficient to allow T. muris expulsion. Conclusions Collectively, our data suggest that increasing levels of fermentable, non-starch polysaccharides in refined diets compromises immunity to helminth infection in mice. This diet-infection interaction may inform new strategies for manipulating the gut environment to promote resistance to enteric parasites. Supplementary Information The online version contains supplementary material available at 10.1186/s12915-023-01640-z.

properties may aid in a healthy gut environment and improve immunity and disease resistance [3].
Regular consumption of green vegetables has been associated with improved immune function and less enteric pathogen infection, due to high contents of vitamins, fibres, and bioactive phytochemicals such as polyphenols [4]. Chicory (Cichorium intybus) is a leafy vegetable widespread across Europe and Asia, where it is grown on an industrial scale, due to its high content of fructans (a mix of fructooligosaccharides and inulin) in the roots (up to 40% dry matter) [5]. Within the plant, the content of nutrients varies considerably; the content of inulin is negligible in the leaves, but high in the roots [6]. The health-promoting effects of inulin and its ability to alter GM composition are well known, and consequently, most studies on chicory have focused on the root part of the plant [7,8]. However, the leaves are also widely consumed as a health-promoting food in humans, reportedly having hepatoprotective and antidiabetic activity, and are traditionally used to treat diarrhoea and vomiting [9]. Recently, anti-inflammatory effects of chicory leaves in several rodent models of autoimmune inflammation have been demonstrated [10,11].
Chicory leaves are both a rich source of fibre, which may act as prebiotic substrate for the GM, and bioactive secondary metabolites such as sesquiterpene lactones (SL) [12]. In livestock such as pigs, chicory forage consumption has been associated with higher lactobacilli:coliform ratios, most likely resulting from increased intake of soluble fibres [13,14]. Moreover, the high concentrations of SL and other bioactive phytochemicals may reduce intestinal infections. Parasitic gastrointestinal worms (helminths) infect more than a billion people worldwide, and are also ubiquitous in livestock, and it is well known that chicory has anti-parasitic properties [15,16]. Notably, grazing animals fed chicory have less worms [17,18], an effect that we recently showed derives from the direct anti-parasitic properties of SL found within the plant [19]. Thus, nutraceuticals based on chicory leaves may hold great promise for reducing inflammation, promoting a healthy GM, and limiting enteric pathogens.
It is becoming increasingly apparent that there is considerable crosstalk between host dietary components and the intestinal immune system, which may have important implications for host responses to gut pathogens. Immunity to intestinal helminths is critically dependent on type-2 immune mechanisms such as the release of cytokines (IL-4, IL-13) from T-cells or innate lymphoid cells and mucus production in the gut epithelium [20]. Differences in the abundance of macronutrients (e.g. proteins), fibres, phytochemicals and vitamins may markedly affect the host mucosal response to infection [21][22][23].
Given the potential anti-parasitic and anti-inflammatory properties of the phytochemicals in chicory [24,25], we hypothesized that consumption of this plant may significantly alter the host response to infection, thus providing a tractable system for dissecting the effects of phytonutrient intake on host-parasite interactions. To this end, we developed a rodent model whereby chicory leaves are incorporated into semisynthetic AIN93G rodent diets and fed to helminth-infected mice. We aimed to determine if dietary chicory could reduce helminth infection, and what impact chicory had on anti-helminth immune responses and infection-induced changes in the GM. We report here that, in contrast to chicory's well known antiparasitic properties in livestock, in this model chicory inclusion enhanced enteric helminth infection. We found that this unexpected result stems from non-starch polysaccharides (NSP) derived from chicory, which are lower in semisynthetic mouse diets. This increased level of NSP promoted a type-1 immune environment and susceptibility to infection, which could be reversed by exogenous administration of a type-2 cytokine. Our findings shed light on the dynamic interaction between dietary components and host immunity in the context of a pathogenic infection.

Chicory supplementation effects immune responses to Heligmosomoides polygyrus infection, but not parasite burdens
To examine in detail how chicory influenced the course of a helminth infection in mice, we fed mice either an AIN93G control diet, or the AIN93G diet supplemented with either 1 or 10% dried chicory leaves, during infection with the small intestinal roundworm Heligmosomoides polygyrus. At 14 days post-infection (p.i.), neither faecal egg counts or worm burdens were different between dietary treatment groups (p > 0.05; Fig. 1A,B). Infection resulted in significant eosinophilia and goblet and Paneth cell hyperplasia in the jejunum, but these parameters were unaffected by diet (Fig. 1C). Moreover, mice infected for 30 days with H. polygyrus and fed either AIN93G or the 10% chicory-supplemented diet also had equivalent worm burdens (Additional file 1: Fig. S1). Thus, in this model dietary chicory did not exert direct anti-parasitic activity, nor influence the pathological response to infection in the small intestinal mucosa. To investigate whether chicory may influence the development of adaptive immune responses to infection, we quantified T-cell profiles in the mesenteric lymph nodes (MLN) at day 14 p.i. H. polygyrus infection in mice fed the AIN93G diet increased the proportion of Th2 (CD4 + GATA3 + ) T-cells in the MLN (p < 0.001) with no changes in the proportion of Th1 (CD4 + Tbet + ) or T-regulatory (CD4 + Foxp3 + ) T-cells ( Fig. 2A). Interestingly, we noted that in infected mice fed the 10% chicory diet, there was a substantial increase (p < 0.01 for interaction between diet and infection) in Th1 cells following infection that did not occur in the other groups, together with a small, non-significant decrease in Th2 cells ( Fig. 2A). Thus, the Th2: Th1 ratio in infected mice fed the 10% chicory diet was significantly skewed towards a Th1 profile compared to control-fedinfected mice, which had a highly polarized Th2 profile characteristic of helminth infection (Fig. 2B). There was no effect of diet on T-regulatory cell proportions in either uninfected or H. polygyrus-infected mice ( Fig. 2A). To confirm further the Th1 polarization, we performed qPCR on tissue from the proximal small intestine at day 14 p.i. to investigate the expression of Dclk, Duox2, Ifng, Il10, and Gpx2. Gene expression confirmed the chicorymediated polarization towards Th1 cells revealed by T-cell phenotyping. Expression of the tuft cell marker Dckl1, which was highly induced by H. polygyrus infection, was significantly suppressed by chicory in a dosedependent manner (Fig. 2C). Similarly, infection-induced expression of Duox2 and Gpx2 was also attenuated by dietary chicory, albeit not significant (Fig. 2C). In contrast, expression of the Th1/Treg-related genes Ifng and Il10 was either unaffected or tended to be increased by chicory supplementation (Fig. 2C). Collectively, these data show that chicory did not lower H. polygyrus infection, and, in fact, appeared to promote a Th1 immune response which is normally associated with helminth persistence.

Dietary chicory increases diversity and modulates Heligmosomoides polygyrus-induced changes in the gut microbiota
Chicory contains a number of putative prebiotic constituents that may modify the host GM [13]. To determine if the chicory-mediated changes in immune function in H. polygyrus-infected mice were accompanied by changes in the GM, caecal digesta samples taken at day 14 p.i. were analysed by 16S rRNA gene amplicon-based sequencing. Interestingly, the mice fed the 10% chicory diet had a significantly more diverse GM than the other groups Shown is mean ± S.E.M for worm burdens, and median for faecal egg counts. C Eosinophil, goblet cell, and Paneth cell responses in jejunal tissue (cells/mm 2 tissue). Shown is geometric mean ± 95% confidence intervals. ***p < 0.001 by two-way ANOVA with Tukey post hoc testing. n = 6 per group (Fig. 3A). Furthermore, samples from the three dietary groups clustered distinctly based on unweighted UniFrac distance metrics (Fig. 4B). Samples from mice fed the 10% chicory diet were most divergent from mice fed the AIN93G control diet, with mice fed the 1% chicory diet clustering closer with controls (intermediate) ( Fig. 3B; p < 0.05 by permutational multivariate ANOVA (PER-MANOVA)). Uninfected and H. polygyrus-infected mice also clustered into two distinct groups ( Fig. 3C; p < 0.05 by PERMANOVA). Notably, when plotting Unweighted UniFrac distance metrics based on the 3 × 2 factorial design, 6 distinct clusters were apparent, indicating profound interactions between the treatments and suggesting that chicory modulated the H. polygyrus-induced changes in the GM (Fig. 3D).
Analysis of composition of microbes (ANCOM) identified 8 bacterial taxa that were significantly differently distributed between the groups (Fig. 3E). Independently of diet, H. polygyrus infection decreased the abundance of Turicibacter spp. and increased the abundance of a zOTU (zero-radius Operational Taxonomic Unit) closely related to Limosilactobacillus reuteri (Fig. 3E). The effects of chicory were markedly dependent on the inclusion level in the diet. Ten percent chicory had profound effects on the relative abundance of Coriobacteriaea and Clostridia spp. (both increased) and Allobaculum spp. (decreased), but these effects were not evident with 1% chicory. Significant interactions were observed between diet and infection for the abundances of Desulfovibrio spp. and zOTUs closely related to Akkermansia muciniphila and Bifidobacterium pseudolongum. Ten percent chicory tended to increase the abundance of Desulfovibrio spp. in uninfected mice, but decreased it in infected mice. The abundance of A. muciniphila was increased by infection in control-fed mice and those fed 10% chicory, but not those fed 1% chicory. Finally, the abundance of B. pseduolongum was markedly increased by 1% chicory, particularly in uninfected mice, but was strongly suppressed by 10% chicory. Collectively, these data indicate that supplementing an AIN93G diet with 10% chicory substantially increases the α-diversity and abundance of specific/several zOTUs, and significantly modulates H. polygyrusinduced changes, in the caecal GM.

Dietary chicory increases Trichuris muris burdens in the caecum and creates a polarized type-1 environment compared to AIN93G diets
As we observed an effect of chicory on H. polygyrusinduced immune responses and GM composition, but not on worm burdens, we next asked whether chicory may have a stronger modulatory effect on a caecumdwelling parasite, at the main site of microbial fermentation in the gut. As the 10% chicory diet induced the strongest effect on immune function and the GM, we tested the effect of this diet in mice infected with the whipworm T. muris. C57BL/6 mice were trickle-infected with three doses of 20 T. muris eggs which induces a mixed Th1/Th2 response and models naturally acquired whipworm infections [26,27]. Strikingly, 5 weeks after the commencement of trickle infection, mice fed the chicory-supplemented diet had significantly higher worm burdens than mice fed the AIN93G diet (Fig. 4A).
To explore the immunological basis underlying the effect of chicory on T. muris burdens, we performed high-throughput Fluidigm-based qPCR to measure expression of a panel of immune and mucosal barrierrelated genes in caecum tissue of infected mice. Mice fed either AIN93G or the 10% chicory diet clustered into two distinct groups based on their transcriptional response (Fig. 4B). We observed a broad downregulation of genes involved in Th2 immune function and mucosal barrier defences, and a strong upregulation of genes involved in Th1 and Th17 immune function ( Fig. 4C-D). Genes that were significantly upregulated in the mice fed chicory included Ifng, Nos2, and Tnf, whilst downregulated genes involved Il4, Retnlb, and Defa3 (Fig. 4C). The caecal microbiota composition differed between the dietary groups (p = 0.055 based on Bray-Curtis Dissimilarity metrics; Fig. 5A), but not on unweighted UniFrac distance metrics (p = 0.25; data not shown). We noted that, consistent with the experiments in H. polygyrus-infected mice, chicory caused an expansion of A. muciniphila, and the Coriobacteriaceae family (Fig. 5B). In contrast to H. polygyrus-infected mice, in this experiment the abundance of Allobaculum spp. and zOTUs corresponding to B. pseudolongum were higher in chicory-fed mice. Notably, mice fed the chicory-supplemented diet had significantly lower levels of Lactobacillus spp. (Fig. 5B). To further explore a role for chicory modulating the response to T. muris, we fed mice either AIN93G or 10% chicory-supplemented diets and infected them with a single dose of 300 T. muris eggs, which typically stimulates a highly skewed type-2 response and worm expulsion around 21 days post-infection (p.i.) [28]. Indeed, all mice fed the AIN93G diet had completely cleared their infection at day 21 p.i., whereas 3/6 of chicory-fed mice harboured worms, albeit at low levels (range 4-14 worms; Fig. 5C). Thus, the effect of chicory was less pronounced during single-dose acute infection than with the low-dose trickle infection which may reflect the high level of immunity induced by this infection regime [28]. Taken together, these data indicate that inclusion of 10% chicory into a purified diet altered the gut environment and increased T. muris infection during trickle infection by inhibiting protective type-2 immune mechanisms.

Dietary pectin promotes Trichuris muris infection and impairs type-2 immunity
To investigate the underlying cause of the impaired type-2 response and increased T. muris burdens in chicory-fed mice, we first performed a detailed chemical composition of the different diets. Chemical analysis showed that the diets with chicory inclusion (particularly at 10% inclusion) were enriched in dietary fibre, especially NSP, compared to the control diet ( Table 1). The 10% chicory diet contained twice the amount of non-cellulosic NSP as the purified AIN93G diet, including the presence of arabinose, galactose, and uronic acids that were absent in the control diet. This prompted us to examine the role of NSP in immunity to T. muris. Uronic acids are the monomeric backbone of pectin, a major component of the primary cell walls of plants, and a prebiotic source of fermentable fibre for the GM [13,29].
To determine if an increased pectin content could be responsible for the observed effects on anti-helminth immunity, we fed mice either an AIN93G diet or the same diet supplemented with 5% pectin during a trickle T. muris infection, i.e. 3 doses of 20 eggs weekly. Notably, pectin-fed mice had significantly higher worm burdens (Fig. 6A). Pectin-fed mice also had significantly higher serum levels of T. muris-specific IgG2a (a marker of a Th1 response), but interestingly, also a trend for higher levels of the Th2-related IgG1 isotype (Fig. 6B). Furthermore, pectin-fed mice had increased levels of Ifng and Nos2 expression, and decreased levels of Retnlb and Il13 expression, in caecal tissue (Fig. 6C). Thus, dietary pectin supplementation closely resembled the observed effects of chicory in impairing immunity to T. muris. Moreover, when mice were given a single dose of 300 eggs, a significant effect of pectin was also observed. Mice fed the AIN93G diet had largely cleared their infection at day 21 p.i., whereas pectin-fed mice all still harboured worms, indicating expulsion was incomplete (Fig. 6D). Consistent with this, levels of T. muris-specific IgE were significantly lower in pectin-fed mice, whereas IgG1 and IgG2a levels were unaffected (Fig. 6E). Finally, gene expression profiling of caecum tissue also revealed decreased expression of Retnlb in pectin-fed mice (Fig. 6F). Overall, these data indicate that inclusion of a non-cellulosic NSP source is sufficient to increase T. muris burdens and impair type-2 immunity during both trickle and acute, high-dose infection.

IL-25 treatment restores immunity to Trichuris muris in pectin-fed mice
We postulated that the inclusion of dietary NSP impaired worm expulsion by promoting a type-1 immune response at the expense of type-2 immunity. However, the possibility remained that chicory or pectin may increase worm burdens by an alternative mechanism, e.g. by acting as an additional nutrient source for the parasites. We reasoned that if impaired immunity was responsible, then exogenous addition of a type-2 polarizing factor should overcome the dietary factors to promote expulsion. IL-25 is an alarmin released by epithelial cells during allergies and helminth infection, which can drive a multi-faceted type-2 response and restore immunity in genetically susceptible mice when administered exogenously [30]. Therefore, mice were fed pectin-supplemented diets during a high-dose T. muris infection, with or without IL-25 treatment. Consistent with previous data, mice fed an AIN93G diet had cleared the infection at day 21 p.i., whereas expulsion was incomplete in pectin-fed mice administered vehicle control (Fig. 7A). Importantly, IL-25 treatment completely restored expulsion (Fig. 7A). Measurement of serum antibodies showed that IL-25 treatment boosted IgE responses that were impaired by dietary pectin (Fig. 7B), whilst analysis of MLN cells showed that pectin diminished Th2 (GATA3 + ) T-helper responses, but IL-25 treatment effectively restored a Th2 dominance within the MLN (Fig. 7C). IL-25 treatment also boosted Retnlb expression in the caecum, but Ifng, Nos2, and Il13 expression was not different (Additional file 2: Fig. S2). Collectively, these data suggest that strengthening the host type-2 response was sufficient to abrogate the effect of the dietary pectin, suggesting that the role of dietary NSP in impairing anti-helminth immunity is by skewing the type-1/type-2 response in favour of a type-1 environment which promotes worm survival.

Discussion
Bioactive dietary components play a large role in regulating gut health and resistance to enteric pathogen infections. Our recent work with chicory has identified that SL have a clear anthelmintic activity, and it is known that SL (especially those derived from chicory) are antiinflammatory and antibacterial compounds [31,32].
Thus, we examined the effects on parasitic infection and gut microbiota in mice. Unexpectedly, rather than reducing parasite burdens, we found that chicory increased whipworm infection, and in both T. muris and H. polygyrus models, inclusion of chicory in the diet resulted in a significant muting of the helminth type-2 response. Thus, in our mouse model, it may be possible that an accompanying high fibre content overruled the potential anthelmintic effects we expected to find due to the SL content of the chicory in the diet.
Incorporation of chicory into AIN93G diets was done so as to keep the diets iso-caloric and balanced for crude protein content, but the diets thus differed in their fibre content, suggesting a possible mechanism for the observed effects. Whilst cellulose content was higher in the chicory-supplemented diets, we have previously shown that cellulose content in the diet does not result in increased T. muris infections [28], suggesting that the most likely cause was the increase in non-cellulosic polysaccharides in the chicory diet; the latter was confirmed by experiments with pure pectin substituting for chicory. Whilst the difference in non-cellulosic polysaccharide content between the AIN93G diet and the 10% chicory-supplemented diet was marginal (an increase from 1.8% to 3.6%), it was clearly sufficient to have a strong effect on T. muris infection. We cannot rule out that other NSP sugar residues such as galactose may also play a role in this response, or that they may be synergistic effects of different NSP within complex plant material in regulating the response to helminth infection.
The exact mechanisms whereby dietary NSP impair type-2 immunity remains to be established, but it is telling that in H. polygyrus-infected mice, high levels of chicory caused increased α-diversity in the GM. A species-rich GM may produce a plethora of metabolites that influence the activity of immune cells [33]. For example, species richness has been correlated with the production of metabolites such as secondary bile acids which promote the production of type-17 cytokines such as IL-22, but may potentially inhibit the expression of type-2 anti-helminth immunity [34,35]. Consistent with this hypothesis, germ-free mice have markedly higher Th2 responses than conventional mice and appear to be somewhat more resistant to H. polygyrus [36,37]. This may suggest that an increasingly rich GM may progressively shift the mucosal immune response towards more of a type-1 or type-17 environment and downplay Th2 responses. It is also notable that germfree mice are refractory to infection with T. muris, due to a lack of microbes that are essential for egg hatching and larval development, thus indicating that dietary manipulation of the microbial environment in the gut may directly modulate T. muris infection [36]. More studies are needed to unravel these mechanisms.
Interestingly, we have recently shown that semisynthetic mouse diets supplemented with pure inulin, a highly fermentable fructan polymer, also impaired type-2 immune responses to T. muris and prevented worm expulsion [28]. Notably, in these experiments, inulin intake was associated with impaired goblet and mast cell hyperplasia. This may suggest that a range of prebiotic substrates, including both highly refined polymers such as inulin and pectin, and crude plant material such as chicory leaves, can modulate the response to helminths in a negative way. The effect of these prebiotic fibres on other immune cell populations such as innate lymphoid cells requires further investigation. However, we have also noted that including inulin in a large animal model of helminth infection (Trichuris suis in pigs) did not have a type-1 polarizing effect, and in fact tended to promote Th2 responses [38]. The reasons for these discrepancies between studies have not yet been resolved, but may be related to differences in diet composition between the semisynthetic nature of compositionally defined mouse diets and the complex nature of pig diets which moderates the effects of the inulin inclusion due to a preexisting and higher level of fermentable fibre in the basal diet. Consistent with this, it has in some mouse models been shown that colitis can be worsened by inclusion of high levels of inulin, but only when incorporated into semisynthetic diets and not unrefined mouse chow [39].
Thus, there appears to be a complex trilateral relationship between diet, the GM and mucosal immune function during pathogen infection. Elucidating the underlying mechanisms of this interaction would be highly valuable for the development of targeted nutritional interventions to fine-tune immune response in the context of different gut infections. Moreover, the ability of dietary components to regulate the type-1/type-2 immune balance in the gut further opens up the possibility for nutritional interventions to correct aberrant autoimmune responses. Indeed, pectin consumption in mice has been linked to reduced severity of food allergies and asthma, which would be consistent with our data showing a suppression of type-2 immune function [40,41].

Conclusions
We have shown here a novel role for plant NSP in mediating susceptibility to enteric helminth infection. An increased understanding of the factors underlying this effect of diet may allow development of new tools to manipulate the gut environment to promote resistance to enteric pathogens.

Mice
In all experiments, female C57BL/6 mice (aged 6-8 weeks; Envigo) were used. Mice were kept in individually ventilated cages with sawdust, nesting material, and ad libitum water and feed. In all experiments, mice were allowed to adapt to their respective diets for 2 weeks prior to infection.

Parasite infection, IL-25 treatment, and necropsy
H. polygyrus and T. muris were propagated as previous described [28,42]. For H. polygyrus-infected mice, an additional 1 cm cut from the proximal end of the jejunum was collected and stored in 10% natural buffered formalin (4% formaldehyde) for histology, and cell numbers enumerated by a microscopist (blinded to the treatment groups). Fresh digesta samples were collected from the caecum and snap frozen at − 80 °C for GM analyses. Furthermore, the MLN were collected and stored in RPMI 1640 media (Sigma-Aldrich). The MLNs were supplemented with 10% foetal calf serum (Sigma-Aldrich) and stored on ice for flow cytometry (see below). Worm count was performed by manually picking as the mice were euthanized, and faeces taken from the colon was used for egg counts by the modified McMaster technique [43].

Experimental diets
The chicory used for the experimental diets was C. intybus cv. "Spadona" (DSV Ltd., Denmark) sown as a pure sward (7 kg seeds/ha) in May 2019 and harvested mid-June 2019 at the experimental facilities of the University of Copenhagen (Taastrup, Denmark, 55_6704800N, 12_2907500E). Approximately 12 kg of fresh chicory leaves were harvested and dried using a constant airflow. The dry and ground leaves were incorporated into a purified AIN93G diet at the expense of starch and casein, with balanced crude protein and metabolizable energy content, and processed into pellets. The pelleting procedure did not involve heating above 30 °C. Three diets were prepared (Table 1): a control diet (standard AIN93G), and two experimental diets with 1 and 10% chicory (Table 1). In addition, a diet with 5% citrus peel pectin (Sigma-Aldrich) in place of starch was also formulated (Table 1) using the same pelleting procedure. All diets were prepared by Sniff Spezialdiäten GmbH, Germany.

16S rRNA gene amplicon sequencing
DNA was extracted from caecum digesta using a commercial Bead-Beat Micro AX Gravity kit (A&A Biotechnology), used in accordance with the manufacturer's guidelines. Before extraction, samples were lysed in lysis buffer supplemented with lysozyme (4000 U) and mutanolysin (50 U) and incubated at 50 °C for 20 min. DNA concentration was measured on Varioskan ® (Thermo Fisher Scientific). The V3 region on the 16S rRNA gene was amplified using the universal forward primer 338 F (5′-ACT CCT ACG GGA GGC AGC AG-3′) and reverse primer 518 R (3′-ATT ACC GCG GCT GCTGG-5′) (0.5 μL of each/sample, at 10 μM concentration) that included Nextera ™ (illumina CA, USA) compatible overhangs. In addition to this, 5 μL/sample 5 × PCRBIO HiFi buffer (PCR Biosystems©, UK), 0.25 μL/sample PCRBIO HiFI Polymerase (PCR Biosystems ©, UK), and 1 μL/sample BSA buffer (Sigma) and formamide was added. 17.75 μL DNA was added after being diluted to 1/500 using sterilized Milli-Q ® water, resulting in a total volume of 25 μL. Standard PCR cycling was used: One initial 95 °C denaturation for 2 min followed by 33 cycles of 95 °C for 15 s, 55 °C for 15 s, and 72 °C for 20 s, ending with one final elongation step at 72 °C for 4 min. The PCR1 products were cleaned using magnetic beads on a Biomek 4000 Workstation © (Beckman Coulter, CA, USA). A second PCR using products from the initial PCR, incorporated standard Nextera Illumina barcodes. Two microliters of initial PCR product plus 4 μL of primer P5 and P7, 5 μL/sample PCRBIO HiFi buffer (PCR Biosystems©, UK), and 0.25 μL/sample PCRBIO HiFI Polymerase (PCR Biosystems ©, UK) was pooled to a total volume of 25 μL. One initial 95 °C denaturation for 1 min was followed by 13 cycles of 95 °C for 15 s, 55 °C for 15 s, and 72 °C for 15 s, ending with one final elongation step at 72 °C for 5 min. The concentration was measured with 1 × Qubit dsDNA HS assay Kit (Invitrogen, CA, USA) on Varioskan ® (Thermo Fisher Scientific, MA, USA), and finally individual barcodes were added to the samples, before sequencing on an Illumina NextSeq platform [28].

Microbiota analysis
Quality control of reads, de-replicating, purging from chimeric reads, and constructing zero-radius Operational Taxonomic Units (zOTU) was conducted with the UNOISE pipeline [44] and taxonomically assigned with Sintax [45]. Taxonomical assignments were obtained using the Greengenes (13.8) 16S rRNA gene database. Data were normalized to 7000 reads (contained within the most indigent sample), and samples below this sequencing depth excluded. PERMANOVA was used to evaluate group differences based on weighted and unweighted UniFrac distance matrices, or Bray-Curtis dissimilarity metrics, and taxa-level differences were assessed by ANCOM. Principal Coordinates Analysis (PCoA) was performed on unweighted UniFrac or Bray-Curtis distances.

RNA extraction and quantitative PCR
RNA was extracted using a commercial miRNAeasy Mini Kit (Qiagen) following the manufacturer's guidelines. Briefly, tissue was homogenized in Qiazol lysis reagent using a gentleMACS Dissociator (Miltenyi Biotec, Germany) and filtered in an RNAeasy spin column (Qiagen) including on-column DNAase treatment. Afterwards, concentration and purity were measured using a Nan-oDrop ND-1000 spectrophotometer (NanoDrop Technologies, DE, USA). First-strand cDNA, including gDNA removal, was synthesized using a commercial Quanti-Tect Reverse Transcription Kit (Qiagen) according to the manufacturer's instructions. Quantitative PCR was performed using PerfeCTa SYBR Green Fastmix (Quantabio) on a AriaMx Real-time PCR System (Agilent, US under the following conditions: 2 min at 95 °C followed by 40 cycles of 5 s at 95 °C and 20 s at 60 °C, and finished with 30 s at 95 °C, 30 s at 65 °C, and 30 s of 95 °C again. Primer sequences are listed in Additional file 3: Table S1. The ΔΔCT method was used to calculate fold changes, relative to the housekeeping gene Gapdh. Samples that failed to amplify (CT ≥ 35 in housekeeping gene) were excluded from further analysis.

Fluidigm analysis
A dynamic in-house gut-immunity panel based on genes involved in immunity, gut microbiota signaling, and gut barrier functions was used [46]. Primers were designed to span an intron if possible and yield products around 75-200 nucleotides long using primer 3 (http:// bioin fo. ut. ee/ prime r3/) or primer blast (https:// www. ncbi. nlm. nih. gov/ tools/ primer-blast/) with standard settings [47,48]. Primer sequences are listed in Additional file 4: Table S2. qPCR was performed using the Biomark HD system (Fluidigm Corporation) on a 96.96 IFC using the manufacturer's instructions. The pre-amplification (TaqMan PreAmp, Thermo Fisher Scientific) and the following Exonuclease 1 (NEB Biolabs) treatment were performed on 8 × diluted cDNA for 17 cycles using a 250-nM pool of the selected primers. Pre-amplified and exonuclease treated cDNA was diluted 8 × before qPCR. Melting curve was assessed in the associated software for multiple peaks and the -RT sample was checked for gDNA background. One primer assay (MVP1) was included as an extra check for gDNA contamination [49]. Primer efficiency was calculated from a calibration curve made from a 5 × dilution row of a pool of undiluted pre-amplified and exonuclease treated cDNA. Primer assays with efficiencies between 80 and 110% and R 2 > 0.98 were accepted for further analysis. All Fluidigm qPCR data processing was performed in Genex6 (multiD Analysis AB). Sixty-nine candidate genes and 8 reference genes were assessed. The reference genes were analysed for stable expression using the geNorm and NormFinder algorithms [50,51]. Actb, Sdha, Tbp, and Ywhaz were most stable and used as reference genes. Candidate gene expression in cycle of quantification values (Cq) were normalized to the reference genes, cDNA duplicates were averaged, relative expression of the lowest expressed were set to 1, and the data were log2 transformed before statistical analysis.

ELISA
T. muris-specific IgG1, IgG2a, and IgE were measured using excretory/secretory antigens and anti-mouse monoclonal antibodies. T. muris antigens were produced by culture of adult worms at 37 °C, followed by collection of culture media and concentration over a 10-kDa Amicon Filter (Sigma-Aldrich). Plates were coated with 5 µg/mL of antigen, before the addition of serially diluted sera followed by biotinylated rat anti-mouse IgG1 (clone A85-1), IgG2a (clone R19-15), or IgE (clone R35-118). Horseradish peroxidase was added, followed by 3,3' ,5,5'-tetramethylbenzidine substrate, and the plates read at 450 nm after the reaction was stopped by the addition of 0.1 M sulfuric acid.