Fig 2: Localization of Cyp26a1+NK cells in the uterus on gd6. Immunohistofluorescence images of uterus sections were co-stained for Cyp26a1 and NKp46; co-localization of Cyp26a1 and NKp46 were shown by the white arrows, scale bar: 20 µm

Fig 3: Warfarin decreases pancreatic cancer metastasis and increase NK cell numbers and activation in lymph nodes and at the metastatic site. (A) KPCluc/zsGreen (zsGreen) -derived pancreatic tumor cells (FC1242luc/zsGreen) were orthotopically implanted into the pancreas of syngeneic C57BL/6 recipient mice. At day 14 the mice were treated with either control drinking water or warfarin sodium in drinking water (0.5 mg/L). Warfarin water was replenished every 3–4 days. Primary tumors, livers, lungs and lymph nodes were harvested at day 29/30. (B) Tumor weights from control (n = 7) or warfarin (n = 6) treated mice. (C) Immunohistochemical staining of CK19+ in mice with lung metastases. (D) Quantification of the total area of lung metastasis per mouse as a percentage of the total lung area for control (n = 4) or warfarin (n = 3) treated mice. *p = 0.05 using unpaired student T-test. Values shown are mean and SEM. (E) Immunohistochemical staining of NKp46+ NK cells in the lungs from pancreatic tumor bearing mice. (F) Quantification of the number of NKp46+ NK cells per cm2 in the lungs of control (n = 7) or warfarin (n = 6) treated mice. (G) Immunoflourescence staining of NK cells in the mesenteric lymph nodes from pancreatic tumor bearing mice. The NKp46 marker is shown in red, granzyme B is shown in green and the nuclei were stained with DAPI (in blue). Scale bar 50 µm. (H) Quantification of the number of NKp46+ and granzyme b+ cells in tumor draining lymph nodes for control (n = 7) and warfarin (n = 6) treated mice. Values shown are the mean and SEM and 3-10 fields/mouse tissue were quantified.

Fig 4: BXCL701 increases tumor levels of chemoattractants CXCL9/10 and CXCR3 + cells. (A) Heatmap showing average fold change of cytokine concentration in mT3-2D tumors treated with BXCL701. Each value is the average from two timepoints: week 4 and week 5 of treatment. Each timepoint consisted of three to four pooled tumors. (B) Representative IHC images and quantification of CXCR3+ staining in tumors treated with PBS or BXCL701 for 4 weeks. Each point represents the average value from one tumor. (*p<0.05 as determined by unpaired two-tailed t-test). (C) Representative IHC images of CD4+, CD8+ and NKp46+ staining in tumors treated with PBS or BXCL701. (D) Quantification of CD4+, CD8+ and NKp46+ staining in tumors treated with PBS or BXCL701 for 7 days and 28 days. Each point represents one tumor/mouse. (ns=nonsignificant, *p<0.05, **p<0.01, ***p<0.001, ****p<0.0001 as determined by unpaired two-tailed t-test). IHC, immunohistochemistry; PBS, phosphate-buffered saline.

Fig 5: Disruption of NK cell homeostasis and maturation of ILC1 and NK cells in Ly5.1C14R mice. (A) Dot plots showing the frequency of ILC1 and NK cells in the spleen. Data show representative plots gated on live NK1.1+ lymphocytes excluding T and B cells in C57BL/6, Ly5.1C14R and C57BL/6 × Ly5.1C14R mice. (B) Total number of NK cells in spleen, thymus, liver, small intestine lamina propria (LP) and within the intestinal intraepithelial compartment (IE). Data show the mean ± SEM pooled from three to six independent experiments (n = 6–15 mice/genotype); thymus are pooled from three to five independent experiments (n = 6–12 mice/genotype). P values were calculated using an unpaired two-tailed Student’s t test. (C) FACS plots showing the frequency of immature (Imm, KLRG1-CD11b-), mature 1 (M1, KLRG1-CD11b+) and mature 2 (M2, KLRG1+CD11b+) NK cells in splenic NK1.1+CD3-CD19- NK cells. (D) Total number of Imm, M1 and M2 NK cells in the spleen and liver of C57BL/6, Ly5.1C14R and C57BL/6 × Ly5.1C14R mice showing the mean ± SEM pooled from three to six independent experiments (n = 6–15 mice/genotype). P values were calculated using an unpaired two-tailed Student’s t test. (E) Total number of ILC1s in spleen, thymus, liver, small intestine lamina propria (LP) and within the intestinal intraepithelial compartment (IE). Data show the mean ± SEM pooled from three to six independent experiments (n = 6–15 mice/genotype); thymus data are pooled from three to five independent experiments (n = 6–12 mice/genotype). P values were calculated using an unpaired two-tailed Student’s t test. (F) Histograms showing the mean fluorescence intensity of NKp46 in various tissues for both NK cells and ILC1 for wild-type (black solid line), Ly5.1C14R (solid blue) and C57BL/6 × Ly5.1C14R (red solid line). CD3?+ cells were used as a control for NKp46 expression (black dashed line). Data are representative of tissues analyzed in (A–E). (G) Expression of TRAIL on NK1.1+CD49a+CD3-CD19- hepatic ILC1 in C57BL/6, Ly5.1C14R and C57BL/6 × Ly5.1C14R (results shown in the upper panels) and in C57BL/6 controls compared to Ncr1gfp/gfp (results from experiment shown in the lower panels). Data show representative plots from three to five independent experiments and indicate the the frequency of expression (n = 6–12 mice/genotype).