key: cord-0024469-p98bdq59
authors: Henden, Andrea S.; Koyama, Motoko; Robb, Renee J.; Forero, Adriana; Kuns, Rachel D.; Chang, Karshing; Ensbey, Kathleen S.; Varelias, Antiopi; Kazakoff, Stephen H.; Waddell, Nicole; Clouston, Andrew D.; Giri, Rabina; Begun, Jakob; Blazar, Bruce R.; Degli-Esposti, Mariapia A.; Kotenko, Sergei V.; Lane, Steven W.; Bowerman, Kate L.; Savan, Ram; Hugenholtz, Philip; Gartlan, Kate H.; Hill, Geoffrey R.
title: IFN-λ therapy prevents severe gastrointestinal graft-versus-host disease
date: 2021-08-26
journal: Blood
DOI: 10.1182/blood.2020006375
sha: 484206e760671d75336d8c2d159cc0b515f141b2
doc_id: 24469
cord_uid: p98bdq59

Immunopathology and intestinal stem cell (ISC) loss in the gastrointestinal (GI) tract is the prima facie manifestation of graft-versus-host disease (GVHD) and is responsible for significant mortality after allogeneic bone marrow transplantation (BMT). Approaches to prevent GVHD to date focus on immune suppression. Here, we identify interferon-λ (IFN-λ; interleukin-28 [IL-28]/IL-29) as a key protector of GI GVHD immunopathology, notably within the ISC compartment. Ifnlr1(−/−) mice displayed exaggerated GI GVHD and mortality independent of Paneth cells and alterations to the microbiome. Ifnlr1(−/−) intestinal organoid growth was significantly impaired, and targeted Ifnlr1 deficiency exhibited effects intrinsic to recipient Lgr5(+) ISCs and natural killer cells. PEGylated recombinant IL-29 (PEG-rIL-29) treatment of naive mice enhanced Lgr5(+) ISC numbers and organoid growth independent of both IL-22 and type I IFN and modulated proliferative and apoptosis gene sets in Lgr5(+) ISCs. PEG-rIL-29 treatment improved survival, reduced GVHD severity, and enhanced epithelial proliferation and ISC-derived organoid growth after BMT. The preservation of ISC numbers in response to PEG-rIL-29 after BMT occurred both in the presence and absence of IFN-λ–signaling in recipient natural killer cells. IFN-λ is therefore an attractive and rapidly testable approach to prevent ISC loss and immunopathology during GVHD.

entire tissue was selected for processing using Phenochart and the images were spectrally unmixed using inForm software and exported as multi-image TIF files, which were analyzed with HALO image analysis software (Indica Labs, Cooales, NM) . Cellular analysis of the images was performed by first identifying cells based on nuclear recognition (DAPI stain), then measuring fluorescence intensity of the estimated cytoplasmic areas of each cell. A mean intensity threshold above background was used to determine positivity for each fluorochrome within the cytoplasm, thereby, defining cells as either positive or negative for each marker. The positive cell data was then used to define colocalized populations. Intestinal stem cells (ISC) and Paneth cells were defined as Lgr5-GFP+ EpCAM+ DAPI+ and Lysozyme+ EpCAM+ DAPI+, respectively. The average distance between every ISC to the nearest Paneth cell was calculated by Nearest Neighbor Analysis.

Expansion of luciferase expressing T cells was quantitated through measurement of luciferinluciferase signal intensity using the Xenogen imaging system (Xenogen IVIS 100; Caliper Life Sciences, CA, USA). Fur on the ventral surfaces was shaved and mice were injected with 500g of luciferin subcutaneously and imaged 5 minutes later under continuous isoflurane-based anesthesia.

After total body imaging, mice were again injected with luciferin and then euthanized and single organs were isolated and imaged. For assessment of donor T cell expansion after transplant, BALB/c Luc T cells were given at time of BMT. For assessment of APC function, expansion of TEa Luc T cells in response to selected antigen presenting cells was performed at day 15 posttransplantation, with 1-2x10 6 flow cytometrically sorted T cells given on day 12 post-transplant.

Seven days post-transplantation, mice were fasted of food and water for 4 hours prior to oral gavage with 8mg of FITC labelled Dextran (MW 4kDa, Sigma-Aldrich) in 200L of PBS. Peripheral blood was collected 4 hours later and serum separated. FITC-Dextran concentration in serum was determined using a Synergy H4 Fluorometer (Biotek) at excitation 485nm and emission 535nm.

Serum IL-6, IL-17A, TNF and IFN were measured via murine Flex Array™ sets (BD Biosciences Pharmingen, San Diego, CA, USA) according to the manufacturer's instructions. Samples were acquired on a BD LSR Fortessa and analyzed using FCAP Array™ Software (BD Biosciences).

Serum, SI and colon IL-28A/B were measured using the R&D Systems Mouse IL-28A/B (IFN-lambda 2/3) DuoSet ELISA on samples obtained from either serum or from mucosal intestinal homogenate as per a protocol provided by Invitrogen. The mucosa was scraped from the underlying muscle layer with a glass slide. The cells were lysed with Tris EDTA (10 mM Tris-HCl, and 1 mM EDTA, pH 7.4) containing 0.05% sodium azide, 1% Tween-80, 2 mM Phenylmethylsulfonyl fluoride (PMSF), and 1 microgram per milliliter of each of the following protease inhibitors: aprotinin, leupeptin, and pepstatin A prior to homogenization. The homogenate was then centrifuged (11,000 x g, 10 minutes at 4°C) and supernatant collected and filtered (4.5 micron filter).

RT-qPCR was performed on RNA isolated from tissues obtained from naïve and post-transplant mice. Tissues were frozen in 500L Trizol and then mechanically homogenized. RNA was then extracted using the QIAGEN RNeasy micro kit, converted to cDNA, and PCR performed using Taqman reagents. For Ifnlr1 the Taqman Gene Expression Assay SM Mm00558035_m1 was used, and for the housekeeping gene SM Mm03024075_m1 (Hprt) was used. For Reg3b, Reg3g, LysP PCR GAPDH was used as the housekeeping gene and primers were used with Sybr Green Supermix using standard PCR conditions on an ABI ViiA7 PCR machine.

BALB/c T cells were isolated from spleen and purified by magnetic bead selection. WT or Ifnlr1 -/-DC were isolated from spleen via density gradient and further purified by magnetic bead selection. DC were irradiated with 2100cGy. Serial dilutions (20,000, 10,000, 5,000 and 0) of stimulator DC were plated with either CD4 + or CD8 + T cells at 200,000 T cells per well, in triplicate. After 4 days of culture, 100L per well of 1:1000 3 H-thymidine was added. 18 hours later proportionate inclusion of 3 H-thymidine was measured scintigraphically.

Congenically marked recipient B6 (CD45.2 + ) mice were conditioned with 1000cGy radiation on day -1 and then on day 0 co-injected with allogeneic 12x10 6 Balb/c (CD45.1 + ) and syngeneic 12x10 6 PTPxC57 (CD45.1 + CD45.2 + ) bone marrow. 48 hours after transplantation mice were culled, spleens harvested, mashed in 2% FCS containing RPMI, and then filtered for single-cell suspensions. For in vivo cytotoxicity assays proportions of remaining syngeneic, allogeneic and recipient derived haematopoietic cells were enumerated by flow cytometry. The index of cytotoxicity was calculated as the ratio of CD45.1+2 syngeneic derived cells to CD45.1 allogeneic derived cells. For recipient NK transcriptional profiling, splenocytes were isolated 24 hours after BMT (48 hours after irradiation) and stained with 7-AAD, CD45.1, CD45.2, CD3, NK1.1 and NKp46 for subsequent cell sorting.

DNA was extracted from 50-100 mg of fecal material using an initial bead beating step followed by extraction using the Maxwell 16 Research Instrument (Promega, USA), according to the manufacturer's protocol, with the Maxwell 16 Tissue DNA Kit (Promega, USA). DNA concentration was measured using a Qubit assay (Life Technologies, USA) and was adjusted to a concentration of 5 ng/l. The 16S rRNA gene encompassing the V6 to V8 regions was targeted using the 803F and 1392R primers 30 modified to contain Illumina specific adapter sequence. Preparation of the 16S library was performed as described, using the workflow outlined by Illumina (#15044223 Rev.B). In the first stage, PCR products of ~466 bp were amplified according to the specified workflow, with an alteration in polymerase used to substitute Q5 Hot Start High-Fidelity 2X Master Mix (New England Biolabs, USA) in standard PCR conditions. Resulting PCR amplicons were purified using Agencourt AMPure XP beads (Beckman Coulter, USA). Purified DNA was indexed with unique 8 bp barcodes Ecogenomics according to manufacturer's protocol. Heat map includes OTUs identified as significantly different (p<0.001) between separately housed WT and Ifnlr1 -/-at week 4, where OTU relative abundance exceeds 2% in at least one sample. Each column includes scaled read counts for one mouse. Read counts normalized using metagenomeSeq.

For analysis of 16S microbial sequencing, reads were cleaned of adapter sequences using Cutadapt 35 and trimmed using Trimmomatic 36 employing a sliding window of 4 bases with an average base quality above 15, followed by hard-trimming to 250 bases with exclusion of reads less than this length. Remaining forward reads were processed following the QIIME2 workflow 37 using DADA2 38 to de-noise sequences. Taxonomy assignment was performed on amplicon sequence variants using BLAST 39 against the SILVA 40 reference database version 132. Differential abundance analysis was performed on raw read counts using DESeq2 41 . Counts were normalized prior to principal component analysis (PCA) and heat map visualization using cumulative sum scaling implemented within metagenomeSeq 42 . PCA was performed using the rda function within the vegan R package 43 . Heat maps were generated using pheatmap 44 .

For intestinal epithelial analyses single cells were isolated using the Lamina Propria dissociation kit (Miltenyi Biotec), stained with 7AAD, CD45.2 and EpCAM and sorted based on GFP expression into Lgr5 + and Lgr5 -populations (see Supplemental Figure 5 ). For NK transcriptome analyses recipient NK cells were gated as 7-AAD -CD45.2 + CD3 -NK1.1 + NKp46 + . All subsequent cell sorting was performed on a BD FACSAria III cytometer. Sorted Lgr5 +/-cells were treated with TRIzol and cryopreserved at -80C. RNA was subsequently extracted after a second chloroform extraction step using the QIAGEN RNeasy Micro Kit. Sorted NK cells were stored at -80C in Arcturus® PicoPure® RNA Isolation Buffer and RNA isolated as per manufacturer's instructions. RNA libraries were prepared using the NEBnext Ultra RNA Library Prep Kit for Illumina (New England Biolabs), assessed for size, and quantified using the 2100 Bioanalyzer (Agilent Technologies) and Qubit fluorometer (Thermofischer Scientific). Libraries were sequenced using high output single-end 75 cycle sequencing kits (version 2) on the Illumina Nextseq 550 platform. Sequence reads in each fastq file were trimmed for adapter sequences using Cutadapt 32 (version 1.11) and aligned using STAR 33 (version 2.5.2a) to the mm19 assembly with the gene, transcript, and exon features of Ensembl (release 70) gene model. Expression was estimated using RSEM 34 (version 1.2.30) and was used as input to assess differential gene expression between groups.

Differential gene expression was determined using the edgeR package 45 within R v3.3.4 and significance defined as p < 0.05 after Benjamini-Hochberg false discovery rate correction. Pathway analysis was performed by single sample gene set variation analysis via the GSVA package 46 using KEGG, BioCarta, Reactome and Gene Ontology (GO) pathway databases and only gene sets between 25-500 genes considered. Heat maps were generated using heatmap.2 function in gplots v3.0.1 R package. Canonical Pathway enrichment analysis for differentially expressed genes (log2 Fold-change >|0.58| and adj. p-value <0.05) across PEG-rIL-29-treated Lgr5 + and Lgr5 -samples relative to genotype-matched PBS-treated samples was done using Ingenuity Pathway Analysis (IPA) 49 . IPA function enrichment was calculated using a right-tailed Fisher exact test with a threshold of significance set at P value of 0.05. Inferences in the significant activation state (z-score >|2|) of canonical pathways, upstream regulatory transcriptional regulators, cytokines and kinases were done using IPA. Positive z-scores reflect a predicted activation state, while negative z-scores reflect the inhibition of upstream regulatory activity. (104 

A-B) RNAseq from sort purified NKp46 Cre+ .Ifnlr1 fl.fl and NKp46 Cre-.Ifnlr1 fl.fl splenic NK cells isolated 24 hours after allogeneic BMT (48hrs after lethal irradiation). A) Heat map showing the pattern of differential gene expression between NKp46 Cre+ .Ifnlr1 fl.fl and NKp46 Cre-.Ifnlr1 fl.fl splenic NK. B) Heat map showing canonical gene sets associated with functional pathways identified in NKp46 Cre+ .Ifnlr1 fl.fl versus NKp46 Cre-.Ifnlr1 fl.fl (n = 6 per group) by GSVA analysis

C) Day 7 posttransplant qPCR enumeration of Cryptidins (pan-cryptidin), Lysozyme P (Lysp) and Regenerating islet-derived protein III gamma (Reg3g) defensin expression from WT and Ifnlr1 -/-recipient mice lethally irradiated (1000cGy), and transplanted with BALB/c derived BM and T-cells (n = 4). Data are presented as mean ± SEM. P values calculated using two tailed Mann-Whitney T test, **p < .01. Supplemental Figure 5. Sort strategy for isolation of Lgr5 + cells from GI tissues. A) spleen as for A) and B). E) Numbers of GFP + Lgr5 + cells isolated from small intestine and ileum from digested gut preparations at day 7 post-BMT (BALB/C  B6) from PBS or PEG-rIL-29 K, with secondary staining for GFP and Ki-67. Data are presented as mean ± SEM. p values calculated using two tailed Mann-Whitney T test