Fig 2: Monocytes, neutrophils, and macrophages increased in chronic CS-treatment emphysema model miceFlow cytometry was used to measure the percentage of CD11b+ Ly6C+ cells (‘inflammatory’ monocytes), CD11b+ Ly6G+ cells (neutrophils), and CD11b+ F4/80+ cells (macrophages) in the small intestine (A) and colon (B) of air-exposed control and emphysema model mice. (C) Immunohistochemical evaluation of F4/80 expression in the small intestine (upper) and colon (lower) from air-exposed control or emphysema model mice (n=5 per group). Scale bar = 100 µm. (D) The ratio of M1 (CD11b+F4/80+CD86+ cells) to M2 (CD11b+F4/80+CD206+ cells) were assessed in the small intestine and colon from three groups (n=5).

Fig 3: Autologous tumor cells/BCG/formalin vaccine reduces TAMs and MDSCs recruitmentTumor slides obtained from each mice group were processed by immunohistochemistry to determine CD68+ (TAMs) and Gr-1+/CD11b+ cells (MDSCs). (A) Significant decreased numbers of CD68+ cells in G2 and G4, compared to G1. Representative immunohistochemical staining of CD68+ cells in all four groups studied are presented as CD68+ cells in red and DAPI nuclear staining in blue. Scale bar, 10 µm. (B) Significant decreased numbers of Gr-1+/CD11b+ cells in G3 and G4, compared to G1. Representative immunohistochemical staining of Gr-1+/CD11b+ cells in all four groups studied are presented as Gr-1+/CD11b+ cells in yellow and DAPI nuclear staining in blue. Scale bar, 10 µm. (C) Scatter diagram analysis showing a positive correlation between CD68+ and CD4+ T cells in G1. (D) Scatter diagram analysis showing a negative correlation between CD68+ cells and CD49b+ cells in G1. The data is shown as the mean percentage ± SEM of five mice per group. Tukey’s post hoc test results are shown (*p = 0.05; **p = 0.01; ***p = 0.001). Scale bar, 10 µm.

Fig 4: Enpp1 levels modulate tumor–immune landscape and mediate LRF. LRF was assessed in tumors derived from cells expressing different levels of Enpp1. A, Western blot analysis of Enpp1 protein levels in control (ANV5-mock) and cells overexpressing Enpp1 (OE-Enpp1). B, Kaplan–Meier curve of the LRF rate in mice inoculated with control and OE-Enpp1 cells (n = 18 mice per group). No differences in tumor margins between groups were detected. C, Tumor volume at day of surgery for each mouse. D, Western blot analysis of lentivirally silenced 720Cy2Rx cells with a shGFP control and silenced with shEnpp1 (shEnpp1a). E, LRF-free survival curves with 10 mice per group. No differences in tumor margins between groups were detected. F, Tumor volume at day of surgery is represented for each mouse. G, Tumor volume at the day of resection from orthotopic injected tumor cells (left) and flow-cytometric quantification by CytoFLEX of tumor-infiltrating immune subpopulations derived from orthotopic tumors of control (720Cy2Rx) and shEnpp1a cells. H, Similar quantification as in G in orthotopic tumors derived from control (ANV5-mock) and OE-Enpp1 cells. I, Time-course analysis by flow cytometry of PMN-MDSC infiltration after orthotopic injection of OE-Enpp1 cells and controls at the indicated times post-injection (n = 8 mice per group). Student t test was used for comparison. J, Similar experiment as in I with shEnpp1a cells and controls (shGFP). PMN-MDSC infiltration in orthotopic tumors (n = 8 mice per group) injected with control or shEnpp1a cells. Student t test was used for comparison. K, Orthotopic tumor volume of tumors derived from OE-Enpp1 and control ANV5 cells in animals (n = 8 mice/group) treated with vehicle or anti-Gr1 antibody (200 µg/mouse/ injection) at the indicated frequency (arrows). Kruskal–Wallis was used for comparison. Median and interquartile range is represented. L, Blood neutrophil counts on day of sacrifice. M, Representative images of anti-Gr1+ immunostaining of the indicated tumors. Quantification of Gr1+ cells was performed by ImageJ. One-way ANOVA was used for comparison. Scale bar, 100 µm. N, LRF-free survival after tumor resection of mice orthotopically inoculated with OE-Enpp1 cells or control cells and treated with vehicle or anti-Gr1+ antibody three times per week (arrows). control groups: 15 mice/group and n = 30/group for OE-ENPP1 groups. Tumor volume at surgery is represented in the right-hand plot. No differences in tumor margins between groups were detected (?2 test). O, Representative images of relapsed tumors (left) and quantification (right) of Gr1+ immunostaining of the indicated tumors. Student t test; scale bar, 100 µm. Log-rank test was used in Kaplan–Meier plots. *, P < 0.05; **, P < 0.01; ***, P < 0.0001.

Fig 5: The C5a receptor inhibitor suppresses MDSC infiltration in TME against cancer metastasis and tumorigenesis. A. Schematic diagram of subcutaneous ID8-luciferase tumors in C57BL/6 mice with CCX168 treatment. The ID8-luciferase cells (5x105) were inoculated into 6-week C57BL/6 mice subcutaneously. After inoculation, the CCX168 (10 mg/kg) or normal saline as control was given by oral gavage every day. B-D. Tumor growth curves, post-dissection measured tumor weights, and volumes of C57BL/6 mice treated with CCX168 or normal saline as control. ID8-luciferase cells (5x106) were inoculated s.c. into 6-week C57BL/6 mice (n=6 per group). The C57BL/6 mice were treated with CCX168 (C5aR inhibitor, 10 mg/kg) or by oral gavage every day. E. The flow cytometry analysis showed fraction change of MDSCs (CD11b+, Gr-1+) in ID8 mice tumor microenvironment. F. Bioluminescence imaging of intraperitoneal ID8-luciferase tumors in C57BL/6 mice with CCX168 treatment. The ID8-luciferase cells (5x105) were inoculated i.p. into 6-week C57BL/6 mice. After 9-day inoculation, the CCX168 (10 mg/kg) or normal saline as control was given by oral gavage every day. Statistic plot of total flux in bioluminescence imaging. G. The graphic illustration of LncOVM-mediated PPIP5K2 regulating complement C5 secretion through the Golgi complex and MDSC recruitment in tumor microenvironment.