Fig 2: Carcinogen-induced immunogenicity is dependent on reduced M-CSF and CD155 expression by cancer cells.(A) Cytokine array on supernatant from DMBA3-4 and DMSO3-1 cells. Red and blue boxes indicate the upregulated and downregulated proteins secreted by DMBA3-4 compared with DMSO3-1 cells, respectively. (B) Relative levels of the select upregulated and downregulated proteins from the DMBA3-4/DMSO3-1 cytokine array (n=2 per group). (C) M-CSF protein levels in DMBA3-4 compared with DMSO3-1 cell lysates (n = 9 per group). (D) Csf1 mRNA expression levels in DMBA3-4 compared with DMSO3-1 cells (n = 7 per group). (E) BMDM migration toward DMBA3-4 versus DMSO3-1 cells in the presence of anti-CSF1R or IgG control antibody. Fold change is determined as the ratio of BMDM migration in the absence of tumor cells at 96 hours after coculture (n=7 per group). (F) CD155, CD112, PD-L1 and PD-L2 expression on DMBA3-4 and DMSO3-1 cells. Numbers on the flow histograms represent the ligands’ MFI. (G) DMBA3-4 plus DMSO3-1 mixed tumor growth in WT mice treated with anti-TIGIT and/or anti-CSF1R antibody compared with IgG-treated controls (n = 10 per group). Mice received 450,000 DMBA3-4 plus 50,000 DMSO3-1 cells per injection site. (H) Survival rate of WT mice that received DMBA3-4 plus DMSO3-1 cells and treated with anti-TIGIT and/or anti-CSF1R antibody compared with IgG-treated controls (n = 10 per group). Unpaired t test (C–E), 2-way ANOVA (G) and log-rank test (H), bar graphs show mean ± SD.

Fig 3: CSF-1/TLR4 induced differentiation promotes endothelial progenitors with predominant lymphatic phenotype.(A) Schematic of the protocol used to differentiate BM cells by sequential activation of CSF-1 and TLR4 pathways. (B) Representative histograms demonstrating expression of lymphatic specific targets Vegfr-3, Colec12, Itga9, Lyve-1, and Pdpn in freshly-isolated BM cells (upper panel, Ex-vivo, blue line) and cells differentiated with CSF-1 and LPS (lower panel, red line). Numbers pointing to blue and red lines indicate the percentage of marker-positive cells. Secondary controls are shown by the grey area outlined by black line in each panel. (C) The mean percentage of positive cells and (D) MFI (x103) of BM cells expressing LEC markers. (E) Representative histograms demonstrating the percentage of Ex-vivo (blue line) and differentiated (red line) cells positive for blood vascular endothelial markers Vegfr-1, CD105, and Vegfr-2. Secondary controls are shown by the grey area outlined by black line in each panel. (F) The mean percentage of positive cells and (G) MFI (x103) of various endothelial markers in ex-vivo and differentiated cells. Black bars indicate statistical significance determined by Student’s t-test with P-values shown by *<0.05, **<0.01, and ***<0.001. All experiments were performed in duplicate and reproduced three times.

Fig 4: M-LECP recruited to ZR-75 human tumors retain myeloid-lymphatic identity but downregulate T-cell and erythroid markers.Human ZR-75 breast tumor cells were orthotopically implanted into SCID mice (N = 5). When tumors reached ~300mm3, mice were infused i.v. with 1.5x106 of GFP+ M-LECP differentiated in vitro using standard CSF-1/LPS protocol. Four weeks after transfer, tumors were excised and triple-stained for GFP, Lyve-1, and (A) CD11b, (B) Ter-119 or (C) CD3e. Nuclei (blue) shown in merged images were visualized by Hoechst stain. Yellow arrowheads indicate GFP+/Lyve-1+ cells that lack markers of other lineages. Cells with rare co-expression of erythroid or lymphoid markers in GFP+/Lyve-1+ cells are indicted by yellow arrows. All images were acquired at 400X magnification. (D) The mean percentage of GFP+/Lyve-1+ cells expressing CD11b, Ter-119 or CD3e was determined on 3–4 tumor sections each derived from individual mice. A minimum of 100 GFP+/Lyve-1+ cells identified in multiple sections were analyzed to calculate the percentage cells positive for each marker. (E) Identical analysis was applied to native Lyve-1+ cells lacking GFP expression. Results are presented as the means ± S.D. “Exogenous” and “Endogenous” labels refer to adoptively transferred (GFP+) or tumor-induced host M-LECP (GFP-), respectively. Statistically significant differences between Lyve-1+ cells with myeloid and other markers were determined by Student’s t-test with P-values of <0.001 indicated by triple stars.

Fig 5: IL-10 pathway regulates transcription of specific markers of multiple lineages in BM myeloid precursors.Marker expression levels in CSF-1/LPS differentiated BM cells from WT mice in the presence of anti-IL-10R antibodies and from IL-10R KO mice were compared with the levels in WT cells treated with a control antibody. Cells harvested on the sixth day of differentiation were analyzed by RT-qPCR for mRNA expression of markers of (A) lymphatic-specific, (B) endothelial, (C) myeloid, (D) erythroid, (E) T-cell, and (F) B-cell lineages. All analyses were performed in triplicate for each target, normalized to β-actin, and independently reproduced twice. Data for each target are reported as the mean fold-change in treated cells compared with the WT cells treated with control rat IgG ± S.D. Statistical significance of expression changes compared with control group was determined by Student’s t-test with P-values indicated by *<0.05, **<0.01, and ***<0.001.