Fibroblastic reticular cells regulate intestinal inflammation via IL-15-mediated control of group 1 ILCs

MyD88-independent FRC subset specification and homeostasis

FRCs generate the cellular scaffold required for the activation of immune cells in LNs^15,16, express multiple Toll-like receptors (TLRs) and respond to stimulation by the innate immune system¹⁹. We therefore reasoned that FRCs in PPs and mLNs might sense mucosal pathogens and affect immunity by directing those immune cells that respond first to the intrusion. To address this issue, we selectively abolished MyD88-dependent innate immunological sensing in FRCs through the use of mice in which loxP-flanked alleles encoding MyD88 undergo FRC-specific deletion by a transgene expressing Cre recombinase under the control of the Ccl19 promoter (Ccl19-Cre), called 'Myd88-conditional-knockout' (Myd88-cKO) mice here. FRCs in PPs can be highlighted in vivo through expression of enhanced yellow fluorescent protein (EYFP) from the ubiquitous Rosa26 locus (R26R) in Ccl19-CreR26R-EYFP mice (called 'Ccl19^EYFP mice' here)¹⁵ (Fig. 1a). Ablation of MyD88 in FRCs did not affect PP formation (Supplementary Fig. 1a), nor did it alter the size or structural organization of PPs (Fig. 1b) or their immune-cell content (Supplementary Fig. 1b–e). Likewise, the structural organization of mLNs (data not shown) and their immune-cell composition (Supplementary Fig. 1b–e) were not affected by this FRC-specific ablation of MyD88. High-resolution in situ analysis by confocal laser-scanning microscopy revealed that the PP FRCs of both Ccl19^EYFP mice and Ccl19^EYFPMyd88-cKO mice formed the typical network structure and expressed the FRC marker PDPN (Fig. 1a,b), CCL21, smooth muscle actin-α and the intercellular adhesion molecule ICAM1 (Supplementary Fig. 1f). Moreover, we found that the lack of innate immunological sensing in Myd88-cKO mice did not affect expression of the Ccl19-Cre transgene in CD31⁻PDPN⁺ stromal cells of PPs or mLNs (Fig. 1c,d). The appearance of FRC subsets such as CD35⁺ follicular dendritic cells (Fig. 1e) or CD157⁺MAdCAM-1⁺ marginal reticular cells was independent of MyD88 in PPs and mLNs (Fig. 1f and Supplementary Fig. 1g). Likewise, the expression of other canonical FRC markers was not altered by the absence of MyD88 in EYFP⁺ cells of PPs or mLNs (Fig. 1g). These data showed that the FRCs in PPs phenotypically resembled LN FRCs and that under homeostatic conditions, FRC-subset-differentiation patterns were independent of MyD88 when the intestinal surface was unperturbed.

MyD88 signaling in FRCs controls antiviral ILC1 responses

To assess whether an invasive enteric pathogen would substantially alter the activity of FRCs, we infected MyD88-sufficient mice and mice with FRC-specific MyD88 deficiency with mouse hepatitis virus (MHV). This cytopathic coronavirus is recognized via the TLR7-MyD88 pathway²⁰, 'preferentially' targets macrophages in SLOs²¹ and causes severe inflammatory disease in the intestine following uptake via the oral route²². Here, we used a dose of 5 × 10⁴ infectious particles, which led to substantial viral replication on days 3 and 6 after infection in the PPs and mLNs of MyD88-sufficient mice (Fig. 2a,b) but spared other regions of their intestine (data not shown). Viral titers were significantly lower in Myd88-cKO mice than in Myd88-sufficient mice on day 3 after infection, and infectious particles were almost completely eliminated from the Myd88-cKO mice on day 6 (Fig. 2a,b). This finding was unexpected, because global deficiency in MyD88 or TLR7 resulted in uncontrolled viral spread, with viral titers more than 10,000-fold higher in mice with such global deficiency than in Myd88-cKO mice (Fig. 2a,b). Moreover, the virus was purged from the spleen and liver of Myd88-cKO mice at day 6 (Supplementary Fig. 2a), which indicated that potent immunological effector mechanisms had prevented systemic spread of the pathogen.

The accelerated viral control in Myd88-cKO mice as early as on day 3 after infection suggested that innate antiviral immune cells had been activated by the FRC-specific MyD88 deficiency. Indeed, we found significantly more cells expressing the activating natural killer cell (NK cell) receptors NK1.1 and NKp46 in the PPs and mLNs of Myd88-cKO mice than in those of Myd88-sufficient early after infection (Fig. 2c,d). Moreover, production of the antiviral effector cytokine IFN-γ was much greater in the NK1.1⁺ cells of Myd88-cKO mice than in their counterparts from Myd88-sufficient mice (Fig. 2e). Histological analysis of PPs from infected mice at day 3 after infection with MHV revealed that NK1.1⁺ cells were located mainly in the interfollicular regions (Fig. 2f). Moreover, these cells were in close contact with FRCs expressing the Ccl19-Cre transgene, both in the presence of MyD88 (Fig. 2g) and in its absence (Fig. 2h) in FRCs. The frequency and number of both non-cytotoxic group 1 ILCs expressing the cytokine receptor IL-7Rα and IL-7Rα-negative NK cells were increased when immunological sensing was altered in the FRCs of PPs (Fig. 2i,j and Supplementary Fig. 2b,c) and mLNs (Supplementary Fig. 2b,c). Control staining revealed that NK1.1⁺IL-7Rα⁻ group 1 ILCs, but not IL-7Rα⁺ group 1 ILCs, expressed the signature transcription factor Eomes (Supplementary Fig. 2d). FRC-specific MyD88 deficiency had a positive effect on the population expansion and activation of group 1 ILCs, whereas the size of the ILC2 and ILC3 compartments in PPs (Fig. 2i,j) and mLNs (Supplementary Fig. 2b,c) was slightly reduced relative to their size in Myd88-sufficient mice. To assess whether this enhanced antiviral immunity was exclusively dependent on group 1 ILCs, we ablated NK1.1-expressing cells using a well-established depletion protocol²³ (Supplementary Fig. 2e,f). Antibody-mediated depletion of group 1 ILCs completely restored viral replication in all organs of Myd88-cKO mice but had little effect on viral replication in Myd88-sufficient mice (Fig. 2k). These data indicated that the FRCs in PPs and mLNs were able to function as potent regulators of antiviral immunity and that this function was exerted via control of group 1 ILCs.

FRCs restrain ILC1 activity through IL-15 deprivation

The differentiation and activation of cytotoxic and non-cytotoxic ILCs is stringently regulated by cytokines²⁴. In addition, NK cell activity is controlled by inhibitory receptors that bind to major histocompatibility complex class I molecules²⁵. Since MyD88-deficient FRCs in PPs and mLNs did not show altered expression of major histocompatibility complex class I during intestinal infection with MHV (data not shown), we established an in vitro cell-culture system (Fig. 3a) to probe FRC cytokine responses following exposure to R848, a synthetic agonist of TLR7 and TLR8. We found that MyD88-sufficient FRCs responded to stimulation of TLR7 with considerable production of the inflammatory mediators IL-6 and CCL2, whereas FRCs lacking MyD88 failed to respond to stimulation with R848 (Fig. 3b). Exposure to this TLR7 ligand led to a substantial reduction in the production of IL-15 by MyD88-sufficient FRCs, whereas MyD88-deficient FRCs continued to produce large amounts of this NK cell– and ILC1-activating cytokine (Fig. 3b). Since IL-15 acts mainly in a cell-contact-dependent manner via trans-presentation by IL-15 receptor α-chain (IL-15Rα)²⁶, we used expression of IL-15Rα as an additional marker of differential FRC activation. Expression of IL-15Rα in MyD88-sufficient FRCs was significantly reduced after stimulation with the TLR7 ligands R848 or single-stranded RNA or with IL-1β (Fig. 3c and Supplementary Fig. 3). MyD88-dependent regulation of the production of IL-15 in vivo was confirmed by RT-PCR analysis showing significantly higher expression of Il15 mRNA in CD45⁻ PP stromal cells from MyD88-cKO mice than in those from Myd88-sufficient mice (Fig. 3d) and in EYFP⁺ cells sorted from the PPs of Ccl19^EYFPMyd88-cKO mice than in those sorted from the PPs of Ccl19^EYFP mice (Fig. 3e), on day 3 after infection. Moreover, flow cytometry revealed that IL-15 was detectable on the surface of EYFP⁺ cells expressing the cell-adhesion molecule CD44 (Fig. 3f). The lack of MyD88 substantially increased not only the amount of surface-bound IL-15 but also expression of the IL-15-trans-presenting molecule IL-15Rα (Fig. 3f). Back-gating showed that the enhanced IL-15 production under conditions of MyD88 deficiency could be attributed to PDPN^dimCD44^hi FRCs (Fig. 3g), which suggested that a distinct FRC subpopulation activated group 1 ILCs via trans-presentation of IL-15.

To assess whether IL-15 is the dominant FRC-derived factor that mediates the population expansion of group 1 ILCs and controls viral replication, we applied a neutralizing antibody to IL-15 (anti-IL-15) during the course of the infection starting 1 d before infection with MHV. We found that treatment with anti-IL-15 efficiently blunted the exaggerated population expansion of group 1 ILCs in Myd88-cKO PPs (Fig. 4a,b). Notably, neutralization of IL-15 led to a reduced frequency (Fig. 4a) and absolute number (Fig. 4b) of group 1 ILCs in Myd88-sufficient mice and diminished the ILC3 population both in Myd88-sufficient PPs and in Myd88-cKO PPs (Fig. 4c). These data confirmed the importance of IL-15 for ILC homeostasis in PPs²⁴ and indicated that innate immunological signaling in FRCs functioned as an important regulatory switch for this IL-15-driven ILC1 proliferation. Assessment of the expression of CD122, a component of the IL-15 receptor, on Eomes1⁺ and Eomes1⁻ ILC1 subsets revealed no major change in expression of the IL-15 receptor β-chain (CD122) under conditions of FRC-specific MyD88 deficiency (Fig. 4d), indicative of an ILC1-extrinsic regulation circuit. Notably, in vivo neutralization of IL-15 disinhibited viral replication in Myd88-cKO mice, while Myd88-sufficient mice showed only minor changes in viral load (Fig. 4e). Together these data revealed that the ILC1 activity in PPs and mLNs was controlled almost exclusively via a single cytokine derived from a particular FRC subset. Moreover, it appeared that innate immunological activation of FRCs via direct recognition of viral RNA or through IL-1β-mediated stimulation was critical for the adjustment of ILC1 reactivity.

FRCs regulate homeostatic ILC1 and NK cell maintenance

To assess whether FRC-derived IL-15 not only controls the proliferation and activity of group 1 ILCs and/or NK cells during an infection but also contributes to their homeostasis, we selectively abolished IL-15 expression in FRCs by generating Ccl19-CreIl15^fl/fl mice (called 'Il15-cKO' mice here). We found that FRC-specific ablation of IL-15 affected neither PP formation (Supplementary Fig. 4a) nor the composition of T lymphocytes and B lymphocytes in this SLO (Supplementary Fig. 4b). NK1.1⁺NKp46⁺ cells were almost completely absent under conditions of FRC-specific loss of IL-15, whereas neither the ILC2 compartment nor the ILC3 compartment was significantly affected (Fig. 5a,b and Supplementary Fig. 4c,d). Since the Ccl19-Cre transgene was active in less than 0.08% of non-endothelial bone marrow stromal cells (Supplementary Fig. 4e), we concluded that the IL-15-expressing FRCs in gut-associated SLOs generated an essential niche for the maintenance of group 1 ILCs and NK cells under homeostatic conditions.

Next we assessed to what extent such specific ablation of Il15 in FRCs affected the activation of ILC1- and/or NK cell–mediated antiviral immunity. Even under inflammatory conditions, we found an almost complete absence of NK1.1⁺NKp46⁺ cells in the PPs (Fig. 5c) and mLNs (Fig. 5d) of Il15-cKO mice. Moreover, other CD45⁺Lin⁻ cells (i.e., group 2 and group 3 ILCs) did not produce IFN-γ during the early phase of the viral infection in Il15-cKO mice (Fig. 5c,d), which indicated that other sources of IL-15, such as dendritic cells or macrophages^27,28, failed to compensate for the lack of this growth factor in the FRC niche. Consistent with the results of the anti-NK1.1 ablation experiment (Fig. 2k), we found only a moderate effect of the selective loss of group 1 ILCs and/or NK cells on the control of viral replication (Fig. 5e). Thus, we concluded that the FRCs built and maintained an exclusive IL-15-dependent niche for the maintenance of group 1 ILCs and/or NK cells in the PPs and mLNs and controlled antiviral immunity in gut-associated SLOs through cessation of IL-15 production.

Innate immunological sensing in FRCs restricts gut inflammation

Early IFN-γ production by NK cells has been shown to be important for the differentiation of helper T cell subsets²⁹. Accordingly, the greater abundance of IFN-γ-secreting group 1 ILCs in Myd88-cKO mice was associated with an elevated frequency of antiviral CD4⁺ T cells that secreted IFN-γ in the PPs and mLNs on day 10 after infection with MHV (Fig. 6a,b and Supplementary Fig. 5a,b) but only moderate proliferation of IL-17-producing helper T cells (Supplementary Fig. 5c). Notably, MyD88 deficiency in FRCs precipitated a significantly lower abundance of regulatory T cells expressing the transcription factor Foxp3 in the PPs of Myd88-cKO mice than that in the PPs of Myd88-sufficient mice (Fig. 6c), which suggested that immunological regulation in the small intestine might have been compromised. Indeed, the lack of innate immunological sensing in FRCs led to substantial weight loss in MHV-infected Myd88-cKO mice (Fig. 6d), despite their accelerated viral clearance (Fig. 2a). As expected, global deficiency in MyD88, which permitted almost unrestricted replication of the cytopathic virus (Fig. 2b and Supplementary Fig. 5d), was associated with a worsened clinical appearance (Fig. 6d). Gross pathological analysis revealed that the intestinal wall of Myd88-cKO mice was considerably inflamed on day 10 after MHV infection, with a significantly shorter colon length than that of Myd88-sufficient mice (Fig. 6e). Histopathological examination confirmed that the enhanced immunopathology in the small intestine of Myd88-cKO mice (Fig. 6f) included an edematous muscular layer and blunting of villi (Fig. 6g). Those pathological changes were associated with a greater antibody response to intestinal microbiota such as Escherichia coli than that of Myd88-sufficient mice (Fig. 6h), which indicated that the integrity of the epithelial barrier was compromised in infected Myd88-cKO mice. Such weakening of the epithelial barriers in MHV-infected Myd88-cKO mice was accompanied by pronounced changes in the composition of the microbiome (Fig. 6i and Supplementary Fig. 5e), whereas the composition of the commensal flora in naive mice was not affected by FRC-specific MyD88 deficiency (Supplementary Fig. 5e). The finding that Myd88-cKO mice showed significantly lower resistance to colonization by other intestinal pathogens such as Citrobacter rodentium on day 12 after infection with MHV, relative to that of Myd88-sufficient mice (Fig. 6j), further emphasized the importance of immunoregulatory functions of FRCs and their ability to adjust global immune responsiveness in the intestine.

Mice.

C57BL/6/N (B6) mice were purchased from Charles River Laboratories (Germany). Myd88^−/− and Tlr7^−/− mice were obtained from the Institute for Laboratory Animal Sciences at the University of Zürich. BAC-transgenic C57BL/6N-Tg(Ccl19-Cre)489Biat (Ccl19-Cre) mice have been previously described¹⁵ and were crossed with R26R-EYFP mice⁴³. To specifically ablate MyD88 or IL-15 in FRCs, Ccl19-Cre and Ccl19^EYFP mice were crossed with mice with loxP-flanked Myd88 alleles (Jackson Laboratory) or with mice with loxP-flanked Il15 alleles. Constitutive ablation of Il15 was achieved by flanking exon 5 with two loxP sequences by standard germline recombination in embryonic stem cells. All mice were on the C57BL/6N genetic background and were maintained in individually ventilated cages and were used between 8 and 10 weeks of age. As control animals, co-housed and Cre-negative littermate mice have been used in all experiments. Experiments were performed in accordance with federal and cantonal guidelines (Tierschutzgesetz) under permission numbers SG11/05, SG09/14 and SG10/14 following review and approval by the Cantonal Veterinary Office (St. Gallen, Switzerland).

Viral and bacterial infection.

Mice were infected with MHV A59, orally by gavage with 5 × 10⁴ plaque-forming units as previously described²¹. Mice were sacrificed and organs were stored at −70 °C until further analysis. MHV titers were determined by standard plaque assay using L929 cells⁴⁴. L929 cells were purchased from the European Collection of Cell Cultures. A nalidixic-acid-resistant strain of Citrobacter rodentium (ICC169) was grown overnight at 37 °C in Luria broth supplemented with nalidixic acid (50 μg/ml, Sigma Aldrich). Mice were fed, on average, 2 × 10⁹ bacteria by oral gavage at day 12 after MHV infection (resolution phase). Organs were harvested at day 6 after infection, homogenized and cultured on selective plates⁴⁵.

Preparation of stromal cells.

PPs or mLNs were dissected into small pieces and transferred into a 24-well dish filled with RPMI 1640 medium containing 2% FCS, 20 mM HEPES (all from Lonza), 1 mg/ml collagenase D (Sigma) and 25 μg/ml DNaseI (Applichem). Dissociated PPs or LNs were incubated at 37 °C for 30 min. After enzymatic digestion, cell suspensions were washed with PBS containing 0.5% FCS and 10 mM EDTA. To enrich for fibroblastic stromal cells, hematopoietic cells were depleted by incubating the cell suspension with MACS anti-CD45 and anti-Ter119 microbeads and passing over a MACS LS column (Miltenyi Biotec). Cell sorting was performed using a FACSAria cell sorter (BD Biosciences) or a S3 cell sorter (Biorad). For in vitro assays, PP or mLN fibroblastic stromal cells were cultured for 7 d in RPMI with 10% FCS, and 5 × 10⁴ cells were stimulated with 100 ng of R848 or left untreated. Supernatants were collected after 24 h, and IL-6 and CCL2 concentrations were determined by cytomeric bead array (CBA, BD biosciences) and IL-15 concentration was determined by ELISA (Abcam).

Flow cytometry.

Single-cell suspensions were incubated for 20 min at 4 °C in PBS containing 1% FCS and 10 mM EDTA with fluorochrome-labeled antibodies (Supplementary Table 1). Cells were acquired with a FACSCanto (BD Biosciences) and analyzed using FlowJo software (Treestar Inc.). Analysis of EOMES, GATA-3, TBET, Foxp3 and RORγt expression was performed using the Foxp3/transcription factor staining buffer set from eBioscience, according the manufacturer's instructions.

Analysis of NK1.1⁺ and CD4⁺ T cells responses was performed using cytokine production after stimulation with phorbolmyristateacetate (PMA, 50 ng/ml) and ionomycin (500 ng/ml; both purchased from Sigma) in the presence of brefeldin A (5 μg/ml) for 3 h at 37 °C. For peptide-specific cytokine production, cells were restimulated with M133 peptide (TVYVRPIIEDYHTLT; GenScript) in the presence of brefeldin A (5 μg/ml) for 4 h at 37 °C. For intracellular staining, restimulated cells were surface-stained and fixed with cytofix-cytoperm (BD Biosciences) for 20 min. Fixed cells were incubated at 4 °C for 40 min with permeabilization buffer (2% FCS, 0.5% Saponin in PBS) containing antibodies to IFN-γ and to IL-17A (Supplementary Table 1). Samples were analyzed by flow cytometry using a FACSCanto (Becton Dickinson), data were analyzed using FlowJo software (Tree Star, Inc.).

Histology.

PPs and mLNs were fixed overnight in freshly prepared 4% paraformaldehyde (Merck) at 4 °C under agitation. Fixed organs were embedded in 4% low melting agarose (Invitrogen) in PBS and sectioned with a vibratome (Leica VT-1200). 20- to 30-μm-thick sections were blocked in PBS containing 10% FCS, anti-Fcγ receptor (Supplementary Table 1) and 0.1% Triton X-100 (Sigma). Sections were incubated over night at 4 °C with the following antibodies: anti-PDPN, anti-B220, anti-CD31, anti-CD4, anti- smooth muscle actin-α, anti-EYFP and anti-CCL21 (Supplementary Table 1). Unconjugated antibodies were detected with the following secondary antibodies: Dylight649-conjugated anti-rat-IgG, Alexa488-conjugated anti-rabbit-IgG, Dylight549-conjugated anti-syrian hamster-IgG and Dylight549-conjugated Streptavidin (Supplementary Table 1). Microscopy was performed using a confocal microscope (Zeiss LSM-710) and images were processed with ZEN 2010 software (Carl Zeiss, Inc.) and Imaris (Bitplane).

Assessment of intestinal inflammation.

Intestinal tissues were removed and cleaned and the distal ileum was immediately fixed in neutral buffered 10% formalin solution. Two to four 5-μm-thick sections of the ileum from each mouse were stained with hematoxylin and eosin. Each sample was graded for the following four criteria on a scale of 0–3 (0, none; 1, mild; 2, intermediate; 3 severe): leukocyte infiltration in the lamina propria; iedema; villus distortion; and presence of markers of severe inflammation such as crypt abscesses, submucosal inflammation, and ulcers. Scores for each criterion were added to generate an overall inflammation score for each sample. Sections were scored in a blinded fashion by two independent observers.

NK cell depletion and IL-15 neutralization.

Mice were administrated 400 μl/mouse of a NK1.1-depleting monoclonal antibody (clone PK136; produced in-house) at days −3, −1 and 2 of infection (day 0)²³. Efficacy of NK1.1⁺ cell depletion was assessed by flow cytometry. For neutralization of IL-15, mice were given daily intraperitoneal injection of 25 μg/mouse of an IL-15-neutralizing antibody (eBiosciences, clone AIO.3), from day −1 to day 2 of the infection.

Detection of bacteria specific IgG.

Serum samples from mice at day 10 after infection with MHV were heat-inactivated at 56 °C for 30 min. Heat-inactivated sera were incubated with cultivable ileal microbiota or E. coli K12 for 30 min at 4 °C. A647-labeled anti-mouse IgG was used for detection in flow cytometry (Supplementary Table 1). Samples were analyzed by flow cytometry using a FACSCanto (BD Biosciences). Data was analyzed using FlowJo software (Tree Star, Inc.).

Quantitative real-time PCR.

Total cellular RNA was extracted from homogenized tissues and sorted cells using TRIZOL reagent (Invitrogen) following the manufacturer's protocol. cDNA was prepared using cDNA archive kit (Applied Biosystems), and quantitative RT-PCR was performed using the Light Cycler-FastStart DNA Master SYBR Green I kit (Roche Diagnostics) on a LightCycler machine (Roche Diagnostics). mRNA expression was measured using the QuantiTect SYBR Green PCR primers (Qiagen).

Microbiota analysis.

For microbial composition analysis, the ileal content of naive or MHV-infected Myd88-cKO and their Cre-negative littermates was used. Microbial composition was assessed by a 16S high-throughput amplicon analysis. The 16S rRNA gene segments spanning the variable V5 and V6 regions were amplified from DNA from ileal content samples, using a multiplex approach with the barcoded forward fusion primer 5′-CCATCTCATCCCTGCGTGTCTCCGACTCAG BARCODE ATTAGATACCCYGGTAGTCC-3′ in combination with the reverse fusion primer 5′-CCTCTCTATGGGCAGTCGGTGATACGAGCTGACGACARCCATG-3′. The PCR-amplified 16S V5-V6 amplicons were purified and prepared for sequencing on the Ion torrent PGM system according to the manufacturer's instructions (Life Technologies). Samples with over 500 reads were accepted for analysis. Data analysis was performed using the QIIME pipeline version 1.8.0. Operational taxonomic units were picked using UCLUST with a 97% sequence identity threshold, followed by taxonomy assignment using either the latest Greengenes database.

Statistical analysis.

Statistical analyses were performed with Graphpad Prism 5.0 using an unpaired two-tailed Student's t-test or Mann-Whitney test. Longitudinal comparison between different groups was performed by one-way ANOVA with Tukey's post-test or two-way ANOVA with Benferroni's post-test. Statistical significance was defined as P < 0.05.

Integrated supplementary information

Supplementary information