ILC2-derived LIF licences progress from tissue to systemic immunity

Mice

All mice were maintained in the Medical Research Council ARES animal facility under specific-pathogen-free conditions at 19–23 °C and 45–65% humidity with a 12/12 h light/dark cycle. In individual experiments, mice were matched for age, sex and background strain. All experiments undertaken in this study were performed with the approval of the LMB Animal Welfare and Ethical Review Body and the UK Home Office. C57BL/6 JOla controls were bred in house. Mouse strains Il7r^Cre (ref. ⁵⁶), Rora^flox/flox (ref. ⁵⁷), Il1rl1^{−/−} (ref. ⁵⁸), Rag2^{−/−}, Rag2^{−/−}Il2rgc^{−/−} (Rag2^{−/−}gc^{−/−}), Lif ^flox/flox, BIC³³, Siglech^Cre (ref. ⁵⁹) and Lifr^–/flox were either on the C57BL/6J Ola background or back-crossed for at least six generations.

Generation of Lif
^flox/flox mice

To produce a Lif allele that could be conditionally deleted by Cre recombinase, we generated a homology-directed repair-template construct for use in combination with CRISPR–Cas9 to insert LoxP sites 5′ and 3′ of the final exon of both protein-coding annotated Lif transcripts (ENSMUSE00000656154). In addition, the construct included a neomycin selection cassette flanked by Frt sites to permit Flp-mediated excision, and both LoxP sites were followed by a BglII site to facilitate screening and verification of appropriately targeted embryonic stem cell clones (Extended Data Fig. 1r). Embryonic stem cells were transfected with this repair-template construct along with expression constructs for WT Cas9 and four single-guide RNAs, two targeting sequences 5′ and two targeting 3′ of the final Lif exon. Neomycin-resistant clones were screened for correct targeting initially by PCR and digested with BglII using 5′ primer pairs P1 and P2 such that a product cleaved by BglII indicated correct targeting. Clones were further verified by Southern blot analysis using 5′ and 3′ probes that both detected a 16.4 kb fragment in the WT allele and 4.9 and 7.6 kb fragments, respectively, in the targeted allele (Extended Data Fig. 1s). Guide RNA target sequences were: G1 (F) TAATGATTCTAGTTGCCTACAGG; G2 (F) TGGAGTCCCCATGTCACAGGTGG; G3 (F) TTCCTCCATCGGTCCAGGAGGGG; G4 (R) TACCCCTCCTGGACCGATGGAGG. Screening primers were: P1 TAGGAAGCCAGAGTCTAGTGGCAGTTTTAAGAGATGG; P2 AAGGCTTCTTTGTCAGAGTGGTCGG. Primers for generation of probes were: 5′ probe 1 fwd CCTGCCACCCCCTTAACCTCCATAAGTGAAAAGCAAGTGG; 5′ probe 1 rev. ACTGGGCCTGCTAGGGGTTTGACAG; 3′ probe 1 fwd TGATGGAGCTGTGGGATGGG; 3′ probe 1 rev. ACACACTCGGGCTCCATTATGC.

Generation of LIFR conditional mice

For generation of pDC-specific LIFR knockout (SiglecH^Cre/Cre Lifr^−/flox) mice, pDC^Cre (SiglecH^Cre/Cre) mice were crossed with Lifr^−/flox to delete exon 5 of Lifr (transcript variant 1). Both Tg(Siglech-Cre,-mCherry)⁵⁹ and Lifr^{tm1a(EUCOMM)Hmgu} (ref. ²⁵) mice were obtained from the European mouse mutant archive. To delete the lacZ–neomycin-resistance cassette and generate mice with a loxP-flanked Lifr allele (EuComm Lifr^tm1c is denoted as Lifr^flox in this paper), chimaeras were bred to FLPe C57BL/6 mice. However, we also detected inefficient Lifr^flox allele recombination and consequently analysed pDC^Cre × Lifr^−/flox mice in which two Cre-mediated recombination events occurred to produce LIFR deficiency, with pDC^Cre × Lifr^−/+ mice as controls. Despite two Cre alleles and one functional flox allele, pDC^Cre × Lifr^−/flox mice showed a reduction in LIFR expression of only 50% in pDCs (Extended Data Fig. 1t).

Mouse challenge protocols

IL-33-induced type 2 lung inflammation

Mice were challenged intranasally with IL-33 (Biolegend; 0.25 µg in 40 µl of PBS) on 3 consecutive days. All tissues were harvested 24 h following the final dose.

Neutralizing LIF antibody treatment

Mice were intranasally challenged with IL-33 (0.25 µg in 40 µl of PBS) and intraperitoneally injected with 100 µg of either isotype antibody (R&D systems) or anti-LIF neutralizing antibody (R&D systems) on 3 consecutive days. All tissues were harvested 24 h following the final dose.

rLIF intranasal challenge

Mice were intranasally challenged with rLIF (R&D systems; 1 µg in 40 µl of PBS) on 3 consecutive days. All tissues were harvested 24 h following the final dose. For the pDC migration kinetics experiment, mice were intranasally challenged with one rLIF dose (1 µg in 40 µl of PBS) and tissues harvested at 6 or 24 h after challenge.

RWP-induced type 2 lung inflammation

Mice were intranasally challenged with RWP (300 μg of protein per dose, Ambrosia artemisiifolia, short form; Greer Laboratories) on 3 consecutive days. All tissues were harvested 24 h following the final dose. For the chronic lung inflammation model, mice were intranasally challenged with RWP thrice weekly over 5 weeks. All tissues were harvested on day 38.

FITC-dextran-labelled cell migration

Mice were challenged intranasally with IL-33 (Biolegend, 0.25 µg) and 40 kDa FITC-dextran (Sigma-Aldrich, 40 µg in 50 µl of PBS) on 3 consecutive days. All tissues were harvested 24 h following the final dosing.

Mouse infection models

PVM infection

Mice were infected intranasally with a single dose of PVM (50 PFU in PBS). PVM strain J3666 stock was a gift from A. J. Easton. All tissues were harvested at 8 days postinfection unless stated otherwise. For the CCL21 kinetics experiment, mice were intranasally challenged with PVM (50 PFU in PBS) and tissues harvested on 0, 4, 8 and 11 days postinfection.

For PVM rechallenge, mice were challenged with PVM (50 PFU in PBS) on days 0 and 30 and all tissues were harvested on day 38.

FITC-dextran-labelled cell migration

Mice were challenged intranasally with PVM (50 PFU) and 40 kDa FITC-dextran (Sigma-Aldrich, 40 μg in 50 µl of PBS). All tissues were harvested 3 days postinfection.

rLIF and PVM challenge

Mice were infected with a single dose of PVM (50 PFU in PBS) intranasally and treated with a daily dose of intranasal rLIF(1 μg) or PBS. All tissues were harvested on day 8 postinfection.

Mice were infected with a single dose of PVM (50 PFU in PBS) intranasally and, on day 8, postinfection intravenous CD45 labelling was performed by injection of 3 μg of anti-CD45 antibody (in 200 μl of PBS) via the tail vein. Mice were then culled 3 min after injection and tissues harvested.

FTY720 and PVM challenge

Mice were infected with a single dose of PVM (50 PFU in PBS) intranasally and injected intraperitoneally daily with either FTY720 (ref. ³⁴) (25 μg in 250 µl; Enzo Life Sciences) or PBS. All tissues were harvested on day 8 postinfection.

Tissue processing

BAL isolation

Mice were culled at the experimental endpoint, tracheae were exposed and BAL was performed by flushing the lungs three times with 0.5 ml of PBS. The fluid obtained was centrifuged at 350g for 5 min; supernatants were stored at −20 °C for cytokine detection.

Serum isolation

Mice were culled at the experimental endpoint and whole blood was collected. Blood samples were allowed to clot for 2 h at room temperature. Samples were centrifuged at 2,000g for 10 min and serum was collected and stored at −20 °C. For immunoglobulin enzyme-linked immunosorbent assay (ELISA), serum was diluted 1/50.

Viral load

Mice were culled at the experimental endpoint, and one lung lobe was snap-frozen in trizol (Invitrogen) and stored at −80 °C for RNA purification.

Tissue preparation

Lung tissue was predigested with 750 U ml^−1 collagenase I (Gibco) and 0.3 mg ml^−1 DNase I (Sigma-Aldrich) before obtaining a single-cell suspension at 37 °C for 30 min; tissue was then passed through a 70 μm cell strainer. For lymphocyte enrichment, a single-cell lung suspension was centrifuged through 30% Percoll (GE Healthcare) at 800g for 15 min. Spleen, thymus and mediastinal LN single-cell suspensions were prepared by passing tissue through a 70 μm cell strainer and lysing red blood cells. Single-bone marrow cell suspensions were prepared by flushing the femur and tibia with endotoxin-free PBS and lysing red blood cells.

Flow cytometry

Single-cell suspensions were incubated with fluorochrome- or biotin-conjugated antibodies in the presence of anti-CD16/CD32 antibody (Fc block, clone 2.4G2), followed by fluorochrome-conjugated streptavidin where necessary. All samples were costained with a cell viability dye (Fixable dye eFluor780, Invitrogen) and analysed on either a 5-5-laser LSRFortessa system (BD Biosciences, BD FACSDiva software v.6.2) or spectral cytometer ID7000 (Sony Biotechnology). Either FACSAria Fusion systems or iCyt Synergy (70 μm nozzle, Sony Biotechnology) was used for cell sorting. Precision Count Beads (BioLegend) were used to calculate cell numbers. Intracellular transcription factor staining was performed using the Foxp3 staining kit (eBioscience) according to the manufacturer’s instructions. For lymphocyte intracellular cytokine staining, cells were cultured with complete RPMI supplemented with Cell Stimulation Cocktail or protein transport inhibitors (eBioscience) for 4 h at 37 °C. Intracellular cytokine staining was performed using BD Cytofix/Cytoperm Plus reagents (BD Biosciences) following the manufacturer’s instructions. The expression of LEC CCL21 was detected by additional staining with goat anti-mouse CCL21 (R&D systems) and anti-goat-Alexa 488 (Invitrogen), with no stimulation, and using BD Cytofix/Cytoperm Plus reagents (BD Biosciences) following the manufacturer’s instructions.

For intracellular phospho-STAT3 staining, cells were fixed with 2% paraformaldehyde (PFA) for 15 min and overnight permeabilization with 90% methanol at −20 °C, followed by incubation with fluorochrome antibodies diluted in 2% bovine serum albumin PBS.

Flow cytometric analysis, including unsupervised dimensionality reduction and clustering, was performed using FlowJo, LLC v.10 (BD) and associated plug-ins. Unless otherwise stated, pDCs are defined as LiveCD45⁺CD11b^−F4/80^−CD317⁺SiglecH⁺. Myeloid cells include cDCs and CD11b^− cDCs as LiveCD45⁺CD11b^−CD11c^highSiglecF^−MHCII^high; CD11b⁺ cDCs as LiveCD45⁺CD11c^−/intm SiglecF^−Ly6G^−CD19^−TCRb^−MHCII^high; monocytes as LiveCD45⁺CD11b⁺ SiglecF^−Ly6G^−F4/80⁺MHCII^−; and alveolar macrophages as CD45⁺CD11c⁺F4/80⁺SiglecF⁺. Eosinophils are defined as CD45⁺CD11c^−F4/80^−CD11b⁺Gr1^int SiglecF⁺; and neutrophils as CD45⁺CD11c^−F4/80^−CD11b⁺Ly6G^highSiglecF^−. T cells are defined as LiveCD45⁺TCRb⁺ and B cells as LiveCD45⁺CD19⁺TCRb^−; CD4⁺ T cells as CD45⁺CD3⁺CD4⁺ and ILC2s as CD45⁺Lin^−(CD3,CD4,CD8,CD19,CD11b,CD11c,FcεR1) CD127⁺ICOS⁺. Endothelial cells are defined as LiveCD45^−CD31⁺, LECs as LiveCD45^−CD31⁺PDPN⁺ and BECs as LiveCD45^−CD31⁺ PDPN^−.

All flow cytometry data were processed and analysed using FlowJo v.10, RRID: https://scicrunch.org/resolver/SCR_008520.

In vitro cultured cells

Lung immune cell sorting

Mouse ILC2s were purified from IL-33-treated lungs (see Methods for IL-33-induced type 2 lung inflammation) as LiveCD45⁺Lineage^−IL-7Rα⁺ST2⁺KLRG1⁺; and pDCs were purified from IL-33-treated lungs (see Methods for IL-33-induced type 2 lung inflammation) as LiveCD45⁺F4/80^−CD11b^−CD317⁺SiglecH. Cells were snap-frozen in trizol for RNA purification, and conditioned medium was collected and stored at −20 °C.

Mouse lung ILC2s and T cells for qPCR analysis were purified from PVM-challenged mice using the same gating strategy as for ILC2s: LiveCD45⁺Lineage^−IL-7Rα⁺ST2⁺KLRG1⁺; CD4⁺ T cells as LiveCD45⁺TCRb⁺CD4⁺; CD8⁺ T cells as LiveCD45⁺TCRb⁺CD8⁺; BECs as LiveCD45^−CD31⁺PDPN^−; and LECs as LiveCD45^−CD31⁺PDPN⁺. Purified cells were snap-frozen in trizol for RNA purification.

ILC2 in vitro stimulation

Purified ILC2s were cultured for 24 h with IL-7 (10 ng ml^−1) and IL-2 (50 ng ml^−1) with or without IL-33 (10 ng ml^−1). Cells for RNA purification and conditioned media were collected and stored at −20 °C.

pDC culture, purification and activation

Bone marrow cells were obtained by flushing femurs and tibias with RPMI, followed by incubation with red blood cell lysis buffer for 5 min. Following washing, cells were cultured with RPMI containing 10% fetal calf serum, 1% penicillin/streptomycin, sodium pyruvate, non-essential amino acids, l-glutamine, β-mercaptoethanol and Flt3L (10 ng ml^−1) for 7–10 days. Medium was refreshed on days 3 and 6. pDCs were sorted from bone marrow-derived cultures by fluorescent activated cell sorting as LiveCD45⁺CD11c^intSiglecH⁺CD317⁺ cells. Purified pDCs were activated with CpG (6 μg ml^−1, Invivogen) and treated with or without rLIF (500 ng ml^−1) for 24 h.

Chemotaxis assay

Migration assays were performed using Millicell cell culture inserts (Merck Millipore) with 2–3 × 10⁵ cells per well. Purified pDCs were activated with CpG (6 μg ml^−1) and treated with or without rLIF (10 ng ml^−1) for 24 h. Activated pDCs were placed in inserts with 5 μm pores for 3 h in the presence or absence of cytokine rLIF (500 or 1000 ng ml^−1) or chemokine rCCL21 (R&D systems) at 150 ng ml^−1. The number of migrating cells was then evaluated using a flow cytometer. The results are expressed as migration index (number of migrating cells in chemokine/number of migrating cells in medium).

Lung LEC purification and culture

The preparation of single-cell suspensions from lung tissues is described in ‘Tissue preparation’. CD31⁺ lung cells were isolated from lung cell suspension using magnetic beads (CD31 biotin, Streptavidin dynabeads)⁶⁰. Isolated CD31⁺ lung cells were seeded onto 0.2% gelatin-coated, six-well plates and cultured on complete growth medium. consisting of ECGS (Corning), 20% fetal bovine serum, 1% penicillin/streptomycin, sodium pyruvate, non-essential amino acids and 25 mM HEPES, in a humidified incubator with a gas mixture of 21% O₂ and 5% CO₂ at 37 °C until 70–80% confluence was achieved (usually reached in 4–7 days). Endothelial cells were detached with Accutase (Stemcell Technologies) and purified using magnetic beads (podoplanin biotin, Streptavidin dynabeads). Purified LECs were seeded onto 0.2% gelatin-coated, six-well plates, cultured in complete growth medium and used for experiments.

In vitro LEC treatment

Isolated LECs were treated with or without rLIF for 6 or 24 h, detached using Accutase and stained for flow cytometry analysis.

ELISA and MAGPIX Luminex Array

Culture supernatant was collected and stored at −20 °C until analysis. Serum IgE was measured by ELISA (Invitrogen). LIF, IL-5, type I IFN, CCL19, CCL21, CCL25, CXCL9 and CXCL10 were measured using ProcartaPlex kits (Invitrogen).

Virus neutralization assay

Sera from PVM-rechallenged mice were diluted in a 1:10 ratio and heat-inactivated at 55 °C for 30 min. An equal volume of PVM at 500 PFU per well concentration (1:20 final serum dilution) was incubated with serum for 1 h at 37 °C and 5% CO₂. BHK-21 cell monolayers (1 × 10⁵ per well) were infected with the virus mixture and incubated for 72 h at 37 °C and 5% CO₂. Before harvesting the cells were washed three times with PBS, snap-frozen in Trizol (Invitrogen) and stored at −80 °C for RNA purification.

qPCR with reverse transcription

RNA was purified using Direct-zol RNA Purification Kits. For assessment of viral load, frozen tissue samples were homogenized before RNA purification. Complementary DNA synthesis was performed using SuperScript IV Reverse Transcriptase and oligo d(T)₂₀ (Invitrogen). PVM viral load was tested with forward primer 5′-GCCTGCATCAACACAGTGTGT and reverse primer 5′-GCCTGATGTGGCAGTGCTT³⁸ in a SYBR green qPCR assay. For lung samples, the mouse HPRT gene was used as an internal control. For the PVM neutralization assay, the dC_T for viral amplification was measured with respect to the hamster GAPDH gene. For other qPCR analyses, commercially available Taqman gene expression assays (Applied Biosystems; Extended Data Table 1) were used. Samples were run on the ViiA7 real-time PCR system (Applied Biosystems).

RNA-seq

Cells were sorted by flow cytometry into PBS and 50% fetal calf serum. and RNA was extracted using the RNeasy Plus Micro kit (Qiagen). Following assessment using a Bioanalyser (Agilent), RNA was processed for RNA-seq using Ovation RNA-seq System v.2 (Nugen), fragmented by a Covaris M220 ultrasonicator and bar-coded using Ovation Ultralow Library Systems (Nugen). Samples were sequenced using an Illumina HiSeq 4000 by running a single-read 50-base-pair protocol (Cancer Research UK, Cambridge Institute). Sequence data were trimmed to remove adaptors and sequences with a quality score below 30 using Trim Galore (v.0.50, Babraham Bioinformatics) and then aligned to the mouse genome (GRCm38) using STAR (v.2.6.0a); differential expression was calculated using DESeq2 (v.1.18.1).

Bioinformatic identification of candidate ligands and receptor pairs

Gene lists for cytokines and cytokine receptors were obtained by downloading Gene Ontology gene lists for ‘Cytokine Activity’ and ‘Cytokine Receptor Activity’ from the Mouse Genome Informatics website. The curated mouse CellTalkDB database of ligand–receptor pairs was used to identify interacting gene pairs between these gene lists⁶¹. Using R programming language, the dplyr package was utilized to filter the CellTalkDB database by the cytokine and cytokine receptor gene lists to remove non-cytokine-related ligand–receptor pairs. This filtered list of ligand–receptor pairs was then used to interrogate bulk RNA-seq data of pDCs and ILC2s isolated from mouse lung. Expression of cytokine ligand–receptor pairs in which expression of the receptor by pDCs was greater than 10 RPMK and expression of the ligand by ILC2s was greater than 10 RPKM was then extracted. Ligand–receptor pairs involving Cd44 or Cd74 were excluded from the analysis due to the high expression levels of these transcripts.

Histology

Tissue was fixed in 10% formalin overnight and paraffin embedded; sections were stained with haematoxylin and eosin. Lung histology sections were assessed by a researcher blinded to groupings and given a score between 0 and 5 based on the presence or absence of large cellular aggregates.

Microscopy

Mice were euthanized and received intracardiac perfusion with PBS, followed by 4% PFA (Invitrogen). Lung and LN were collected and fixed with 4% PFA overnight. Fixed tissues were washed with PBS and placed in 30% sucrose for 24 h. Subsequently these were embedded in Optimum Cutting Temperature compound (VWR, catalogue no. 25608-930), frozen in Isopentane and sectioned on a Leica CM1860 cryostat. Sections were incubated in a blocking solution (2% goat serum and 0.5% Triton X-100 in PBS) for 1 h at room temperature. Tissue sections were then incubated overnight at 4 °C with primary antibodies against CD3e, B220 and VEGFR3, then with secondary antibodies for 1 h at room temperature. Images were acquired with an Olympus VS200 slide scanner and processed and analysed using ImageJ2 v.2.14.0/1.5 f.

Lung histology sections were assessed for TLS by a researcher blinded to groupings and given a score between 0 and 5 based on the pathology.

Data and statistical analyses

Statistical analysis was performed using GraphPad Prism v.10.0b software.

Bulk RNA-seq data generated in this study have been deposited at the Gene Expression Omnibus (GEO) under accession number GSE243691.

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.