Prdm1 promotes type 1 ILCs homeostasis and terminal maturation.

(A) Correlation between the average expression of NK cell-associated genes(NCR1, NCR3, PRF1, CD160, KLRB1) and PRDM1 in LIHC (Liver Hepatocellular Carcinoma; n=363) patients sourced from TCGA datasets. (B) Heatmap of the ordered, z-score normalized expression values for PRDM1 and NK cell-associated genes in liver cancer patients. The top and bottom quartiles of all patient samples are indicated. (C) Prognostic value of the NK-PRDM1 signature for overall survival of liver cancer patients comparing high and low quartiles. (D) Schematic representation of the Prdm1-conditional knockout mouse model. Targeted exon 6-8 of Prdm1 (top) is flanked with loxP sites (middle). Ncr1-expressed Cre recombinase was used to generate the Prdm1ΔNcr1 allele (bottom). (E) Real-time RT-PCR quantification of Prdm1 expression in NKp46+ cells to determine the presence of Prdm1 (n=5). (F) Representative flow cytometric plots (left) and quantification (right) of the proportion and absolute number of CD3-NKp46+ cells among lymphocytes in liver, lung, and bone marrow (n=8). (G) Representative flow cytometric plots (left) of the CD11b and CD27 expression within CD3-NK1.1+NKp46+ cells in liver, lung, and bone marrow (n=7). Right panel showed the percentage of distinct stages of NK cells. Data are presented as the mean±SEM and were analyzed by 2-tailed, paired t-test. Differences were evaluated between littermates. Each circle and square on graphs represents an individual mouse; P, P-value; r, pearson correlation coefficient; *, P<0.05; **, P<0.01, ns, not significant.

Prdm1 is required for type I ILCs to control tumor metastasis.

(A) Representative flow cytometric plots (left) of liver cNK cells (CD49a-CD49b+) and ILC1s (CD49a+CD49b-) from Prdm1+/+ and Prdm1ΔNcr1 mice. The two bar graghs (right) quantitate the percentages and absolute numbers of cells respectively (n=7). (B and C) Splenocytes from B2m-/- and B2m+/+ were labeled with CFDASE and eF670 respectively. Labeled cells were 1:1 mixed and injected i.v. into Prdm1+/+and Prdm1ΔNcr1 recipient mice to evaluate in vivo NK cell target-killing ability. Representative flow cytometric plots (left) of transferred cells recovered from recipient mice and percentage (right) of NK cell-specific rejection of donor cells in spleen (n=6) (B) and liver (n=7) (C) between Prdm1+/+ and Prdm1ΔNcr1 mice. (D and F) Image (left) and quantification (right) of tumor nodes on the livers (n=7) (D) and lungs (n=10) (F) of Prdm1+/+ and Prdm1ΔNcr1mice at day 14 or 21 after inoculation with B16F10 melanoma cells. (E and G) Histopathological images of liver (E) and lung (G) tissues stained by hematoxylin-eosin to detect tumor metastasis. Red bar indicates 500 μm distance under the microscope. (H) Liver cells and splenocytes were costimulated in the presence or absence of IL-12 and IL-18 for 12 hours. GolgiStop was added 4 hours before intracellular staining of IFN-γ. The graphs showed percentage of IFN-γ+ splenic cNK cells, liver cNK cells and ILC1s from Prdm1+/+ and Prdm1ΔNcr1mice (n=5). Data are presented as the mean±SEM and were analyzed by 2-tailed, paired t-test. Differences were evaluated between littermates. Each circle and square on graphs represents an individual mouse; P, P-value; *, P<0.05; **, P<0.01, ns, not significant.

Bulk RNA-seq depicts Prdm1-mediated functions in peripheral cNK cells.

(A) Splenic cNK cells and liver CD45+ cells were sorted from Prdm1+/+ and Prdm1ΔNcr1 mice using flow cytometry, and prepared for bulk RNA-seq and single-cell RNA-seq analysis. (B) Volcano plot of the bulk RNA-seq differentially expressed genes (log2(fold change)>0.5; P<0.05) in splenic cNK cells between Prdm1+/+ and Prdm1ΔNcr1mice. Upregulated and downregulated genes in Prdm1ΔNcr1 cells were highlighted in red and blue. (C) Enriched Gene Ontology (GO) terms of DEGs in Prdm1ΔNcr1 cells compared Prdm1+/+ cells. The Enrichment gene set in upregulated (red) and downregulated (blue) genes were indicated in different colours. Bar length represents statistical significance. (D and E) Gene Set Enrichment Analysis (GSEA) showing the enrichment of negative regulation of IL-6 production (D) and NF-kappa B signaling pathway (E) of DEGs in Prdm1ΔNcr1 cells compared Prdm1+/+ cells. NES, normalized enrichment score. (F) Heatmap of selected genes from DEGs. Shown is z-score transformed expression of DEGs. (G and I) Representative flow cytometric plots (left) and cumulative data (right) showing the relative mean fluorescence intensities (MFIs) of Granzyme B (G) and Perforin (I) in liver cNK cells and ILC1s from Prdm1+/+ and Prdm1ΔNcr1mice (n=5). (H and J) Relative MFIs of Granzyme B (H) and Perforin (J) in splenic NK cells at different stages of maturation was analyzed by flow cytometry (n=5). Data are presented as the mean±SEM and were analyzed by 2-tailed, paired t-test. Differences were evaluated between littermates. Each circle and square on graphs represents an individual mouse; P, P-value; *, P<0.05; **, P<0.01, ns, not significant.

scRNA-seq reveals unique properties of two clusters from liver type I ILCs following Prdm1 knockout.

(A) Uniform manifold approximation and projection (UMAP) visualization of liver CD45+ cells from Prdm1+/+ and Prdm1ΔNcr1 mice. Twelve clusters were defined and indicated by distinct colours. Each dot represents a single cell. (B) UMAP visualization of liver cNK and ILC1 clusters. Cells were colored by origins (Prdm1+/+-blue; Prdm1ΔNcr1-red). (C) Percentages of cNK cells and ILC1s in type I ILCs (left), and their distribution in each cluster (right). (D-I) UMAP visualization of three different liver cNK (D) and ILC1 (G) clusters from two mouse strains (E and H). Proportions of cNK cells (F) and ILC1s (I) among total cells (cNK cells or ILC1s) (left) and within clusters (right) were calculated respectively. (J) Violin plots showing the normalized expression of select genes in different cNK clusters. (K) Enriched GO term of marker genes in three cNK clusters. Dot size represents enriched gene number, and color intensity represents significance.

Junbhi signature is associated with the activation of multiple signaling pathways.

(A) Ridge plots showing the normalized expression of Gzmb, Prf1, and Junb in cNK and ILC1 clusters between Prdm1+/+ and Prdm1ΔNcr1 cells. (B) Violin plots showing the normalized expression of select genes in different ILC1 clusters. (C) Violin plot showing the Junbhi signature score for cNK cell and ILC1 clusters, calculated using the signature genes of Junbhi cNK cluster. (D) Enriched GO term of marker genes in three ILC1 clusters. Dot size represents enriched gene number, and color intensity represents significance. (E-J) GSEA plots (left) depicting the enrichment of NF-kappa B (E), TNF (F), IL-17 (G), and MAPK (H) signaling pathway in Junbhi cNK cluster compared with clusters of Prf1hi and Cxcr3hi cNK cells, and the enrichment of IL-17 signaling pathway (I) and T cell differentiation (J) in Il7rhi ILC1 cluster compared with clusters of Klrahi and Gzmahi ILC1s. Right panel showed dynamic relative expression of the given gene sets from cluster1 to cluster3 between Prdm1+/+ and Prdm1ΔNcr1. Dots represent the average expression of given gene set in each cell, which was calculated through the sum of normalized expression of each individual gene within the designated gene set in every single cell. NES, normalized enrichment score.

Prdm1 facilitates the intercellular communication between liver type I ILCs and macrophages.

(A) UMAP visualization of macrophages (Mac) cluster. Four cell subsets were defined. (B and C) Circle plots (B) and summary data (C) illustrating the significant enriched ligand–receptor pairs among cluster of liver cNK cells, ILC1s, and macrophages from Prdm1+/+ (left) and Prdm1ΔNcr1 (right) cells. The thickness of the line indicates the number of enrich pairs, and the arrow reflects the direction of the interaction. (D) Heatmap of overall signaling pattern recognized from ligand-receptor pairs, which contained the sum of signaling from the sender and target cells. (E) Bar graphs showing the information flow in selected active signaling patterns between Prdm1+/+ and Prdm1ΔNcr1 cells. Relative information flow was calculated as the sum of the communication probability in given signaling patterns. (F) Chord plot of the CXCL signaling interaction network among cluster of liver cNK cells, ILC1s, and macrophages in Prdm1+/+ cells.

Prdm1 safeguards type I ILCs from exhaustion-like phenotypes in the tumor microenvironment.

(A-C) The Mitochondrial mass (MitoTracker Green staining; n=8) (A), Mitochondrial ROS (MitoSOX staining; n=5) (B), and Mitochondrial membrane potential (TMRM staining; n=5) (C) of splenic cNK cells, liver cNK cells and ILC1s were analyzed by flow cytometry. Representative flow cytometric plots (left) and cumulative data (right) showing the relative mean fluorescence intensities (MFIs) of each group. (D) Percentages of IFN-γ+ liver cNK cells and ILC1s from Prdm1+/+ and Prdm1ΔNcr1tumor-bearing mice at day 14 after inoculation with B16F10 melanoma cells via intrasplenic injection (n=5). Data are presented as the mean±SEM and were analyzed by 2-tailed, paired t-test. Differences were evaluated between littermates. Each circle and square on graphs represents an individual mouse; P, P-value; *, P<0.05; **, P<0.01, ns, not significant.