Enteric helminth-induced type-I interferon signalling protects against pulmonary virus infection through interaction with microbiota

Abstract

Helminth parasites have been reported to exert beneficial immune modulatory roles in allergic and autoimmune conditions and detrimental effects in tuberculosis and some viral infections. Their role in co-infection with respiratory viruses is not clear. Here, we investigated the effects of strictly enteric helminth infection with Heligmosomoides polygyrus on respiratory syncytial virus (RSV) infection in a mouse model. H. polygyrus infected mice showed significantly less disease and pulmonary inflammation after RSV infection, associated with reduced viral load. Type 2 and adaptive immune responses were not essential since protection against RSV was maintained in IL-4Rα^-/- and RAG1^-/- mice. Importantly, H. polygyrus infection upregulated expression of IFN-β and IFN stimulated genes (ISG) in both the duodenum and the lung, and its protective effects were lost in IFNAR1^-/- and germ-free mice, revealing essential roles for type I IFN signaling and microbiota in H. polygyrus induced protection against RSV. In conclusion, we demonstrate that a strictly enteric helminth infection can have protective antiviral effects in the lung through induction of a microbiota-dependent type I IFN response.

Introduction

Respiratory syncytial virus (RSV) is a major respiratory pathogen. It infects nearly all infants by the age of 2 years (1), but does not induce lasting immunity and leads to recurrent infections throughout life. It is estimated that worldwide, 33.4 million children under the age of 5 experience RSV lower respiratory tract infection (LRTI) annually and 10% of these require hospitalisation, resulting in up to 199,000 deaths (2–4). There is also major morbidity and mortality due to RSV in the elderly (5). Currently, there is no effective vaccine available for RSV, and treatment is limited to supportive care. Severe RSV LRTI is associated with and thought to be due to severe pulmonary inflammation.

In addition, severe RSV infection during infancy has also been associated with increased risk for asthma development. There is substantial evidence indicating that children hospitalized with RSV-bronchiolitis, are more likely to experience recurrent wheezing episodes for a prolonged period of time after recovery from this illness (6–9).

Helminths infect approximately 3 billion people worldwide. It has long been proposed that infection with helminths could suppress the development of immune-mediated disease, as in countries where their prevalence is high the prevalence of asthma, allergy and autoimmune conditions has been found to be correspondingly low (10). Intestinal helminths in particular have been of major interest due to their ability to modulate host immune and inflammatory responses to foreign antigens (11–16) and several clinical trials have been carried out or are underway, assessing their utility as therapeutic agents in inflammatory bowel disease, multiple sclerosis and asthma(17).

Helminth infections rarely occur in isolation and co-infections are very common with varying effects such as reduced pathogen control and increased disease, as reported for HIV infection and tuberculosis (18–21). Recent experimental models in mice report reactivation of systemic latent γ-herpesvirus and reduced control of enteric norovirus replication (22, 23) indicating that in these systems, helminth infection suppresses anti-viral immunity resulting in increased viral replication. However, the impact of helminth infection on respiratory viruses is not well understood. Clinical data are lacking, but mouse models suggest reduced influenza-induced pathology in helminth co-infection (24, 25).

Here, we investigated whether infection with the strictly enteric murine helminth Heligmosomoides polygyrus would change the course of disease and inflammation during RSV infection. This is the first study to demonstrate protective effects of helminth infection on RSV infection and reveals a novel mechanism of type I IFN induction by enteric helminth infection at a site distant from the gut.

Results

H. polygyrus protects against RSV disease and inflammation as a consequence of reduced viral load

Mice were infected with H. polygyrus, and 10 days later, when adult worms emerge into the lumen of the gut, mice were infected with RSV. H. polygyrus coinfection protected against RSV-induced weight loss (Fig 1A) and reduced RSV-induced increases in enhanced pause, indicative of deterioration in baseline lung function (Fig 1B).

Fig. 1. H. polygyrus infection attenuates RSV disease and inflammation and reduces RSV viral load.

The standard co-infection protocol was used as follows: female BALB/c mice were given 200 H. polygyrus L3 larvae by oral gavage at day -10 or left naive. At day 0, 6x10⁵ PFU (A, B) or 4x10⁵ PFU (C, D, E) RSV or UV-inactivated RSV was administered intranasally, (A) Mice were weighed daily and percentage of original weight is shown; (B) Lung function was assessed by whole body plethysmography (WBP) measuring enhanced pause (Penh); (C,D) Samples were taken at the indicated time points after RSV infection for flow cytometric analysis. Numbers of CD3⁺ CD8⁺ T cells (C) and of MHCII⁺CD11b⁺CD11c⁺ conventional dendritic cells (D) per right lung lobe are shown; (E) Lungs were harvested on days 3, 4 and 6 post RSV infection and plaque assays performed.

All data are depicted as mean ± SEM. Data in A & B pooled from 2 independent experiments, total n=8 per group, in C,D & E from 2 independent experiments, total n=6 per group per time point. Statistical significance of differences between RSV infected groups was determined by two-way ANOVA with Bonferroni’s post hoc test. *P<0.05, **P<0.01,***P<0.001.

RSV infection in the mouse model induces pulmonary inflammation with cellular infiltration, specifically of NK cells, CD8⁺ T cells and conventional DCs (cDC) (26, 27). In mice co-infected with H. polygyrus, RSV-induced increases in NK cell, B cell (Supp Fig 1A & B) and CD8⁺ T cell numbers were absent while the increase in cDC numbers was significantly reduced (Fig 1C&D). Early pro-inflammatory cytokine production of IL-6 and TNF-α on day 2 was induced to a significantly lower level in H. polygyrus infected mice compared to those infected with RSV alone (Supp Fig 1C & D). IFN-γ increased with RSV infection, but was not suppressed in co-infected mice, indicating selective inhibition of a pathway independent of IFN-γ (data not shown).

Given these changes in RSV-induced signs of disease, we asked whether H. polygyrus suppresses the immune response or directly alters RSV infectivity. Lung viral titers, as assessed by plaque assay, peaked at day 4 post-infection. H. polygyrus coinfection reduced viral titre without changing the kinetics of replication (Fig 1E). In C57BL/6 mice ex-vivo plaque assays are unreliable, therefore we tested the effects of co-infection in C57BL/6 mice by measuring expression of the RSV L gene in the lung by RT-PCR, as an indicator of viral load. L gene expression was, again, significantly reduced in H. polygyrus infected mice in this strain (Supp Fig 1E).

These findings demonstrate a potent inhibition of RSV-induced disease, early pro-inflammatory cytokine production and recruitment of a broad range of immune cells to the lung in H. polygyrus co-infection presumably due to an initial reduction in viral infection.

Type 2 and adaptive immune responses are not required for H. polygyrus-induced protection against RSV infection

Type 2 immune responses are crucial during most helminth infections, aiding in wound healing and immunity to helminth infection (28–31). IL-4Rα-deficient mice cannot respond to IL-4 or IL-13 signals, and present strongly diminished type 2 immune responses (32). Consistent reductions in RSV titers were observed in H. polygyrus co-infected IL-4Rα^-/- mice, similar to those seen in wild type BALB/c mice (Fig 2A).

Fig. 2. Type 2 and adaptive immune responses are not required for the attenuation of RSV viral titres.

The standard co-infection protocol was followed (A) in BALB/c IL-4Rα deficient mice and (B) in BALB/c RAG1-deficient mice. Lungs were harvested on day 4 of RSV infection and plaque assays performed to determine titres. Data in panel A are pooled from 2 individual experiments, total n=4-8 per group. Data in panel B are representative of 2 independent experiments, n=3-4 per group. Statistical significance between groups was determined by one-way ANOVA with Tukey’s post hoc test. *P<0.05, **P<0.01.

All data are depicted as mean ± SEM.

To determine if other adaptive immune responses are required for H. polygyrus-mediated protection against RSV infection we used RAG1-deficient mice, which lack all T and B cells. Once again, RSV titers were significantly suppressed in both RAG1^-/- mice and wild type controls following co-infection with H. polygyrus (Fig 2B). Together, these observations show that adaptive immune responses and IL-4Rα-dependent type 2 cytokine responses are not required for the protective effect of H. polygyrus on RSV infection

H. polygyrus infection induces expression of type I IFN, IFN stimulated gene in both the duodenum and the lung

Type I IFNs are major players in the initial response to viral entry into the mucosa (33). Since the adaptive and type 2 immune responses were not essential for the protection against RSV infection, we hypothesised that H. polygyrus enhances the mucosal innate immune response conferring an antiviral state. 2’5’ oligoadenylate synthetase (OAS) and viperin are two of many IFN stimulated genes (ISG) which have been found to play an protective role in RSV infection and can be driven by type I IFN signaling (34–37). Gene expression of IFN-β, viperin, and OAS1A was increased in the duodenum from day 3 post- H. polygyrus infection (Supp Fig 2). Importantly, expression of these genes was also increased in the lung at day 10-14 of H. polygyrus infection (Fig 3A-C), and remained increased 1 hr after RSV infection (Fig 3 D-F). At 12 hrs after RSV infection expression of these genes increased further to similar levels irrespective of H. polygyrus co-infection (data not shown). This data suggests that upregulation of Viperin, OAS1a and IFN-β may underpin H. polygyrus-induced protection against RSV infection in the lung.

Fig. 3. H. polygyrus induces type I IFN and associated gene expression in the lung and type I IFN signalling is essential for H. polygyrus-induced protection against RSV-infection.

BALB/c mice were given 200 L3 H. polygyrus larvae or left naïve. Half of the large left lung lobe was placed in Trizol and RT-PCR was performed for expression levels of IFN-β, OAS1a, Viperin in lung (A-C) comparing H. polygyrus infected to naïve mice. The standard co-infection protocol was followed in BALB/c mice (D-F) and C57BL/6 mice or (G) IFNAR1 deficient mice. 1 hour after RSV infection half of the large left lung lobe was placed in Trizol and RT-PCR was performed for expression levels of (D) IFN-β (E) OAS1a (F) Viperin. (G) 3 days post-RSV infection half of the large left lung lobe was placed in Trizol and RT-PCR was performed for expression of RSV L gene. All results were normalised to 18S expression and represented as fold change in expression over naïve controls (A-C), UV-RSV controls (D-F), or C57BL/6 RSV controls (G). Data are depicted as mean ± SEM. Data are pooled in A-F from 2 independent experiments, total n=6-8 per group and in G from 2 individual experiments, total n=10 per group. Statistical significance of differences between groups was determined, A-C & G by two-way ANOVA with Bonferroni’s post hoc test and in D-F by one-way ANOVA with Tukey’s post hoc test. *P<0.05, **P<0.01, ***P<0.001, NS = non-significant.

H. polygyrus-induced protection against RSV infection requires type I IFN receptor signaling

Since ISG, including Viperin and OAS1a are expressed upon type I IFN receptor signalling, we used IFNAR1-deficient mice which fail to signal in response to IFN-α and IFN-β. In IFNAR1^-/- mice suppression of RSV load in H. polygyrus coinfection was lost, implying an essential role for this pathway in protection (Fig 3G).

The cathelicidin CRAMP is upregulated during H. polygyrus infection but is not required for expression of type I IFN and ISGs

Cathelicidins are small, cationic antimicrobial peptides, of which there is only one present in humans and mice: LL-37 and CRAMP respectively. LL-37 has been shown to have direct antiviral activity against RSV (38) and has been linked to type I IFN production in DCs (39, 40) which has been implicated in wound healing (41). Interestingly, mCRAMP expression was also found to be upregulated in both the duodenum and the lung during H. polygyrus infection (Supp Fig 3A&B) similar to type I IFN and ISG expression, and remained elevated 1 hour after RSV infection (Supp Fig 3C). It appeared that mCRAMP expression in the duodenum was perhaps preceding type I IFN and ISG expression and could therefore be driving these responses. Thus, type I IFN and ISG expression were investigated during H. polygyrus infection in CRAMP deficient mice. However, type I IFN and ISG expression was intact in H. polygyrus infected CAMP^-/- mice (Supp Fig 3D-F), indicating that mCRAMP is not the initiator of the protective antiviral immune response induced during H. polygyrus infection.

H. polygyrus adult excretory secretory products are not responsible for the effects on RSV infection, while larval stages alone confer protection

Much interest has been building around the prospect of helminth excretory secretory (ES) products as potential therapeutics (42). H. polygyrus ES (HES), secreted by adult worms collected from the intestinal lumen, has been shown to have systemic effects in models of disease, and to mimic the effects of live infection (43). HES was administered in various regimes, by the intranasal and intraperitoneal routes, the day before RSV infection, for a week prior to infection, prior and post-infection and also by continuous HES treatment via an intra-peritoneal osmotic mini-pump. None of these protocols resulted in significant reduction in viral titers when compared to RSV infected controls without HES treatment (Supp Fig 4).

The lack of protection afforded by adult worm products, together with the lack of requirement for an adaptive immune response caused us to question whether adult worms play any role in the interaction with RSV, or if larval stages of H. polygyrus and the damage associated with their initial invasion of submucosal tissue is key. Therefore, we irradiated stage 3 H. polygyrus larvae, as a non-lethal means of preventing their maturation to adulthood (44). The larvae are consequently able to penetrate the duodenal wall and enter into the submucosa, causing the initial trauma associated with infection, but do not re-emerge into the lumen as adults. With any of the 3 irradiation doses given, the larvae reduced RSV titers to the same level as the non-irradiated larvae, providing the same degree of protection (Fig 4A). No adults were found in the lumen in the 300Gy treated group, and numbers were severely reduced following 100Gy irradiation of larvae, but granulomas were observed in all groups on the duodenal serosa (data not shown), confirming that the irradiated larvae were still able to invade the intestinal mucosal epithelium (45).

Fig. 4. H. polygyrus Larval stages are sufficient and microbiota are required to protect against RSV infection.

(A) 200 L3 H. polygyrus were irradiated at 100, 200 or 300 Gy and used to infect mice following the standard co-infection protocol. On day 0 4x10⁵ PFU RSV was administered intranasally. Lungs were harvested on day 4 of RSV infection and plaque assays performed.

(B-D) The standard co-infection protocol was followed in BALB/C germ-free and SPF mice, using 400 germ free H. polygyrus L3 larvae and 3x10⁷ sterile RSV in 100μl. On day 4 after RSV infection, (B) the right lung lobes were removed and placed in Trizol for RT-PCR for the RSV L gene using quantifast SYBR green RT-PCR kit (C) the left lung lobe was removed and plaque assays performed (D) the first centimetre of the duodenum was removed and placed in Trizol and RT-PCR was performed for expression of IFNβ. Results in B are normalised to beta-actin expression and represented as fold change in expression over SPF RSV infected controls. Results in D are normalised to 18S and represented as fold change in expression over C57BL/6 RSV infected controls. All data are depicted as mean ± SEM. A is representative of two individual experiments, total n=3-4 per group. Statistical significance of differences between groups was determined in A by one-way ANOVA with Tukey’s post hoc test, and in B-D by unpaired t-test. *P<0.05, **P<0.01, ***P<0.001, NS= non-significant.

The presence of the gut microbiota is essential for H. polygyrus induced protection against RSV infection

Larval stages of H. polygyrus protected against RSV infection, and this effect could be attributed to either the direct damage caused upon larval penetration of the submucosa, and/or the consequent translocation of intestinal bacteria into the mucosal tissues. To ascertain whether the microbiota play an important role in protection, we studied RSV infection in germ-free mice in the presence or absence of H. polygyrus infection.

In contrast to fully-colonised SPF mice, RSV titers and RSV L gene expression were not suppressed in germ-free mice by H. polygyrus co-infection (Fig 4 B&C). Furthermore, the upregulation of type I IFN expression seen in the duodenum of H. polygyrus infected SPF mice was absent in H. polygyrus infected germ free mice (Fig 4D). These data support a model in which the microbiota play a critical role in the induction of type I IFNs and ISGs during H. polygyrus infection, which in turn leads to functional antiviral protection in the lung.

Discussion

Here we demonstrate that a strictly enteric helminth can have protective effects against RSV infection in the lung, through a mechanism mediated by microbiota-dependent type I IFN production. Firstly, we established that co-infection with H. polygyrus ameliorated RSV-induced disease (manifesting as weight loss and deteriorated lung function) as well as reducing the production of pro-inflammatory cytokines and infiltration of immune cells (NK cells, cDCs, CD8⁺ T cells and B cells) into the lungs. Unexpectedly, this was associated with, and presumably a consequence of, a reduction in RSV load following H. polygyrus co-infection. These protective effects were found to be independent of the type 2 and adaptive immune responses as demonstrated in IL-4Rα^-/- and RAG^-/- mice respectively. In addition, these protective effects could not be replicated with HES treatment instead of live infection. Finally, enteric helminth infection upregulated antiviral IFN-β, ISG, and CRAMP gene expression in both the duodenum and the lung, and the protective effects of H. polygyrus on RSV infection were dependent on type I IFN receptor signaling and the presence of microbiota, as demonstrated in IFNAR1^-/- and germ-free mice, which were not protected against RSV infection by H. polygyrus.

The role of helminths in co-infections is not well understood (21). In particular, respiratory viral infection in the context of co-infection with helminths has not been investigated in epidemiological studies, nor in any great detail in animal models. H. polygyrus co-infection has previously been shown to reduce influenza virus titers and antibodies against the virus regardless of the lifecycle stage of helminth used (24). In addition, Trichinella spiralis was found to have protective effects against influenza infection that were dependent on the intestinal phase of infection, enhancing weight gain following influenza-induced weight loss and reducing cellular infiltration into the lung (25). These observations are similar to the reduced weight loss observed in the H. polygyrus and RSV co-infection model reported here, and the reduced cellular infiltrate into the lung. However, the mechanisms involved in this protection were not elucidated in previous studies. More recently, chronic infection with Schistosoma mansoni provided significant protection against lethal influenza infection and infection with pneumovirus of mice (PVM) (46). This was found to be dependent upon the presence of eggs, which are known to cause significant damage to the gut wall. S. mansoni induced TNF-α dependent induction of Muc5ac led to goblet cell hyperplasia in the lung, indicating increased epithelial barrier function. However, this was independent of type I IFN production, without any increase in type I IFN in the lung of S. mansoni infected mice over controls.

Helminths induce a strong Th2 immune response, which is characterised by high levels of IL-4, IL-5 and IL-13, infiltration of eosinophils, basophils and alternatively activated macrophages, as well as high production of IgE (29–31). In recently reported murine models, helminth induced type 2 immune responses and associated alternative macrophage activation aggravated γ-herpesvirus and norovirus infection (22, 23). However, our data show clearly that the Th2 response is not involved in protection against RSV which was maintained in IL-4Rα^-/- mice. In fact the helminth-induced adaptive immune response was all together dispensable for protection, indicating an important role for the innate antiviral immune response.

Type I IFNs are an important part of the innate antiviral immune response that can be triggered through activation of pathogen recognition receptors (PRR) by viral components. They not only have direct antiviral activity, but they also have the ability to upregulate the expression of ISGs which have further antiviral potential, further limiting viral infection and spread. Type I IFNs and ISGs are rapidly upregulated following RSV infection and decline by 24 hours post infection (47, 48). The ISGs Viperin and OAS, have previously been found to play a role in inhibiting RSV infection and have potent antiviral activity (35, 37). However, there is very little evidence linking helminths and type I IFNs in the literature. Aksoy et al, found that double stranded structures found in S. mansoni egg RNA triggered TLR3 activation which in turn lead to the activation of the type I IFN response (49). In H. polygyrus infection, the type I IFN response has previously been reported to inhibit granuloma formation around larval parasites, but expression of the cytokines in direct response to infection was not measured (50).

Irradiation of stage 3 H. polygyrus larvae has been previously shown to inhibit their maturation, but allows larval migration into the intestinal submucosa, after which point the larvae do not develop further into adults (44). By taking this approach, we demonstrated that larval stages are sufficient to induce protection against RSV infection. Further investigation in germ free mice revealed a requirement for microbiota in helminth-induced protection. It is therefore plausible to speculate that the damage caused by initial penetration of larvae into the submucosa may result in bacterial translocation from the gut and activation or priming of the innate immune response. Indeed, upregulation of type I IFNs at epithelial barrier surfaces can reduce bacterial translocation by upregulating tight junctions (51). Thus bacterial translocation in the intestine during H. polygyrus infection may induce upregulation of type I IFNs systemically to limit such translocation. In addition, commensal, but not pathogenic, bacteria have been shown to induce type I IFN production and can also provide protection against influenza infection (52–54).

Viral LRTI with RSV and rhinoviruses in the first years of life has been linked to the development of asthma (55–57), which helminth infections have been shown to protect against in mouse model systems (58, 59). In parallel, intestinal helminth infections in humans have been reported to increase bacterial translocation (60). Thus, we speculate that helminth infection may protect against severe respiratory viral infections in early life, and that this effect in turn contributes to a reduced potential for asthma development.

In conclusion, we show that intestinal helminth infection can be beneficial in respiratory viral infection. Based on our findings we hypothesise that helminth infection in the gut triggers type I IFN production through bacterial interactions, which leads to systemic type I IFN induction thus raising preparedness of remote sites such as the lung to mount an effective innate response against incoming unrelated viral pathogens. These findings suggest a potential for new helminth-based approaches to prevention and treatment of respiratory viral disease.

Materials and Methods

Animals

BALB/c, C57BL/6, IL-4Rα^-/-(61), RAG1^-/-(62) (BALB/c background), IFNAR1^-/- (63) and CAMP-/- (64) (C57BL/6 background) mice were bred in-house at the University of Edinburgh. Germ-free BALB/c mice were obtained from the Clean Mouse Facility (CMF), University of Bern, Bern, Switzerland, and were compared to SPF BALB/c mice from Charles River Breeding Laboratories (l'Arbresle Cedex, France). 6-12 week old female mice were infected by oral gavage with 200 stage 3 H. polygyrus larvae. Ten days later, mice were intranasally infected with RSV or mock infected with UV-inactivated RSV (UV-RSV) (standard coinfection protocol).

Parasites, parasite products and virus stocks

Parasites were maintained as previously described (65). Stage 3 H. polygyrus larvae were irradiated with 100, 200 or 300 Gy using a GSR-C1 irradiator at a rate of 6.2 Gy/min prior to administration by oral gavage. Axenic H. polygyrus larvae were produced as previously described (66). Plaque purified human RSV (Strain A2, ATCC, United States) was grown in Hep-2 cells as previously described (38).

Whole body plethysmography

Baseline lung function was assessed in individual mice, using whole body plethysmography (Buxco Europe, UK). Mice were placed into individual chambers, and baseline measurements were recorded for 5 minutes. Enhanced pause (Penh) values were recorded, averaged, and expressed as absolute values as previously described (67).

RSV immunoplaque assay

RSV titers were assessed as previously described (27), in lung homogenate by titration on HEp-2 cell monolayers in 96-well, flat-bottom plates. Twenty-four hours after infection, monolayers were washed, fixed with methanol, and incubated with peroxidase-conjugated goat anti-RSV antibody (Biogenesis, United Kingdom). Infected cells were detected using 3-amino-9-ethylcarbazole and infectious units enumerated by light microscopy.

Lung cell isolation and flow cytometry

Right lung lobes were excised, cut into small pieces, incubated on a shaker with collagenase A (Sigma; 0.23 mg/ml PBS) at 37 °C for 45 minutes and sheared through a 19 gauge needle. After red blood cell lysis (Sigma), the single cell suspension was passed through a 40 μm cell strainer and stained using viability dye eFluor 780 (eBioscience, Hatfield, UK). The following anti-mouse antibodies were used to phenotype lung immune cells: PDCA-1 (EBIO-927), Ly6G (RB6-C5), NKp46 (29A1.4), B220 (RA3-6B2) eFluor 450 conjugated (eBioscience), Ly6C (AL-21), CD8 (Ly-2) Fluorescein isothiocyanate (FITC) conjugated (BD Bioscience), CD11b (M1/70), CD4 (RM4-5) Phycoerythrin (PE) conjugated (eBioscience), CD45 (30-F11) eFluor605 Nanocrystal (NC605) conjugated (eBioscience), CD49B (DX5), CD19 (6D5) (Biolegend), F480 (Cl:A3-1) AlexaFluor 647 conjugated (AbD Serotech), MHCII (M5/114.15.2), CD3 (145-2C11) PerCPCy5.5 conjugated (Biolegend), CD19 (EBIO1D3), CD3 (17A2) (eBioscience), Ly6G (1AB) AlexaFluor 700 conjugated (BD Bioscience) CD11c (N418) PE-Cy7 conjugated (eBioscience). Isotype control antibodies were used on pooled samples. Cells were gated as viable and CD45⁺ and subsequently phenotyped based on their markers as follows: Ly6G^- CD19^- CD3^- CD49B⁺ NKp46⁺ NK cells, Ly6G^- CD19^- CD3^- MHCII⁺ CD11B⁺ CD11C⁺ conventional dendritic cells, Ly6G^- CD19^- CD3⁺ CD4⁺ or CD8⁺ T cells, Ly6G^- CD3^- CD19⁺ CD19⁺ B220⁺ B cells. Samples were collected using LSR Fortessa II. Post-acquisition analysis performed using FlowJo version 7.6.5 software (treestar.inc, Oregon, USA).

Real Time PCR

Lung and duodenum was harvested and homogenised in 1 ml of TRIzol (Invitrogen) using a TissueLyser. Complementary DNA (cDNA) was made from the extracted RNA using the Qiagen QuantiTect Reverse Transcription Kit (Qiagen) following the manufacturer’s instructions. 1 μg RNA was used for the reverse transcription. Primers were diluted in TE buffer to a final concentration of 0.025 nM/μl and probes to 0.005 nM/μl. Custom primers and probes were purchased from Jena Bioscience or Applied Biosystems. PCR amplification was carried out in a 25 μl volume made up of custom 7 μl primer probe mix (300nM primers and 200nM probe), 12.5 μl TaqMan mastermix (Applied Biosystems); 1.75 μl H₂0; 1.25 μl 18S (Applied Biosystems); 2.5 μl DNA template. 1.25 μl of pre-made primer probe mix was used in the following mixture: 12.5 μl mastermix; 5 μl H₂0; 1.25 μl 18S; 2.5 μl DNA template. IFN-β (Mm00439552_s1) and CAMP (Mm00438285) primers and probes were bought premade from Life Technologies. Custom primers used are shown in Table 1.

Table 1. Primers used for Real-time PCR.

Gene	Forward Primer (5’-3’)	Reverse Primer (5’-3’)	Probe (FAM-TAMRA 5’-3’)
OAS-1A	TCCTGGGTCATGTTAATACTTCCA	GAGAGGGCTGTGGTGGAGAA	CAAGCCTGATCCCAGAATCTATGCC
Viperin	CGAAGACATGAATGAACACATCAA	AATTAGGAGGCACTGGAAAACCT	CCAGCGCACAGGGCTCAGGG
RSV-L	GAACTCAGTGTAGGTAGAATGTTTGCA	TTTCAGCTATCATTTTCTCTGCCAAT	TTTGAACCTGTCTGAACATTCCCGGTT
Beta-Actin	GATCAAGATCATTGCTCCTCCTGA	CAGCTCAGTAACAGTCCGCC	N/A
RSV-L	GAACTCAGTGTAGGTAGAATGTTTGCA	TTCAGCTATCATTTTCTCTGCCAA	N/A

Statistical Analysis

All data were analysed using Prism 6 (Graphpad, La Jolla, CA, USA). Analysis of 2 groups used an unpaired t-test. Analysis of 3 or more groups was either using One-way ANOVA with Tukey’s post test or Two-way ANOVA with Bonferonni’s post test. Unless otherwise stated, the differences are non-significant. *** p <0.001; ** p <0.01; * p <0.05.

Study Approval

All procedures were carried out with institutional ethical approval and under Home Office licences. Germ-free animal experiments were performed according to institutional guidelines and to Swiss Federal and Cantonal laws on animal protection.

Significance Statement.

RSV is a major cause of lower respiratory tract infection in infants and is associated with an increased risk of asthma development. Helminth infections have been shown to have beneficial effects in models of asthma, but their role in respiratory viral infection is not well understood. Here we demonstrate that an intestinal helminth can protect against RSV infection and disease in the lung. Analysis of the lung and duodenum revealed helminth induced increases in antiviral genes. We established through knockout and germ free mice that protection against RSV by intestinal helminths is dependent upon both type I interferon signalling and the presence of microbiota.

Abbreviations

cDC

Conventional DC

ES

excretory secretory

HES

H. polygyrus ES

ISG

IFN stimulated gene

LRTI

lower respiratory tract infection

PVM

pneumovirus of mice

PRR

pathogen recognition receptor

OAS

2’ 5’ oligoadenylate synthetase

RSV

respiratory syncytial virus