IFN-γ-mediated suppression of ANGPT2-Tie2 in endothelial cells facilitates tumor vascular normalization during immunotherapy

Figures

• 

  Figure 1. Anti-PD-L1 therapy normalizes tumor vasculature by decreasing the expression of Tie2 and ANGPT2 in a JAK1/2-dependent manner. (A) Diagrammatic representation of the research design used to explore the influences of anti-PD-L1 therapy on tumor blood vessels. The red arrows indicate PD-L1 blockade therapy. The blue arrows indicate a cotreatment with anti-PD-L1 antibody in combination with ruxolitinib (JAKi, a JAK1/2 inhibitor). The orange arrows indicate the time point of sacrifice. The volumes of the three different groups of tumors are presented. This p values were analyzed via two-way ANOVA. (B, C) Representative images showing CD31+ vessels (B) and CD105+ vessels (C) in LLC tumor tissues on day 5 as described in (A). The CD31+ or CD105+ MVD was analyzed separately by one-way ANOVA. Scale bar: 100 μm. (D) LLC tumor tissues on day 5 were stained for CD31 (red) and αSMA (green) to determine the percentage of pericyte coverage. Scale bar: 100 μm. The p values were evaluated by one-way ANOVA. (E) LLC tumor tissue lysates on day 5 were subjected to immunoblotting to assess the expression of specific proteins associated with angiogenesis. And the gray-scale value comparison of the…

• 

  Figure 2. IFN-γ primarily inhibits the expression of Tie2 and ANGPT2 in ECs. (A–C) The expression levels of TEK (A), ANGPT2 (B), and VEGF-A (C) were evaluated in HUVECs, HPMECs, LLC, HCC827, and A549 cells with or without the treatment of IFN-γ (1000 IU/mL) via RT-qPCR. (D–F) HUVECs and HPMECs were stimulated with IFN-γ (1000 IU/mL) for 0, 6, 12, or 24 h. (D) Protein expression levels of ANGPT2, total FOXO1, phosphorylated FOXO1 (S256), total AKT, and phosphorylated AKT (S473) were measured using immunoblotting techniques. mRNA levels of TEK (E) and ANGPT2 (F) were measured via RT–qPCR. The p values were analyzed via one-way ANOVA.

• 

  Figure 3. IFN-γ-mediated activation of STAT1 is required for the reduction in Tie2-ANGPT2 expression. (A–C) HUVECs and HPMECs were pretreated with a JAKi (ruxolitinib) for 24 h before being stimulated with IFN-γ (1000 IU/mL) for an additional 24 hours. mRNA levels of TEK (A) and ANGPT2 (B) were assessed by RT-qPCR. (C) Protein expression levels of Tie2, ANGPT2, and JAK1 were assessed using immunoblotting techniques. (D–F) HUVECs and HPMECs were transfected with siSTAT1 or siNC and subsequently stimulated with IFN-γ for 24 hours. mRNA expression levels of TEK (D) and ANGPT2 (E) were measured via RT-qPCR. The p values were analyzed via one-way ANOVA. (F) Protein levels of Tie2, ANGPT2, phosphorylated STAT1 (Y701), and total STAT1 were assessed using immunoblotting techniques. (G, H) The putative binding sites of the STAT1 binding motif in the −2 kb TEK promoter (G) or ANGPT2 promoter (H) were predicted via the JASPAR database. ChIP-qPCR confirmed the binding sites of STAT1 in the promoters of the TEK and ANGPT2 genes.

• 

  Figure 4. IFN-γ-mediated activation of the AKT–FOXO1 signaling pathway contributes to the regulation of ANGPT2 but not Tie2 expression. (A) HUVECs and HPMECs were stimulated with IFN-γ (1000 IU/mL) for 0, 15, 30, or 60 minutes. Protein levels as shown in Figure 2D were measured by immunoblotting. (B) HUVECs were stimulated with IFN-γ for 24 h or with LY294002 (a PI3K inhibitor) alone or in combination. HUVECs were dual-stained with DAPI (nucleus, blue) and FOXO1 (red). Scale bar: 50 μm. The right chart presents the quantified results of fluorescence digital imaging analysis, showing the percentage expression levels of FOXO1 in both the cytosol and the nucleus. (n = 3). (C–E) HUVECs and HPMECs were processed as detailed in (B). mRNA expression levels of ANGPT2 (C) and TEK (D) were assessed by RT-qPCR. The p values were analyzed by one-way ANOVA. (E) The specified proteins were assessed using immunoblotting.

• 

  Figure 5. IFN-γ inhibits the angiogenic properties of HUVECs. (A) The proliferation of HUVECs and HPMECs treated with IFN-γ (1000 IU/mL) was assessed employing a CCK-8 assay at 24, 48, and 72 hours post-culture. (B) Migration assays were conducted on HUVECs and HPMECs after stimulation with IFN-γ (1000 IU/mL) for 24 h. Scale bar: 50 μm. (C–F) Tube formation assays of HUVECs and HPMECs. Primary HUVECs were seeded in ECM and matrixgel. Photographs taken 10 h post-seeding were analyzed. HUVECs exhibited a tubular morphology after treatment with IFN-γ (1000 IU/mL) or F-ara-A (fludarabine, a STAT1 inhibitor) alone (C) or LY294002 alone (E) or in combination. Scale bar: 400 μm. (D, F) Quantification of tube formation, encompassing total length, number of branches, number of junctions, and number of meshes, is presented. The p values were determined via one-way ANOVA.

• 

  Figure 6. IFN-γ-mediated suppression of ANGPT2-Tie2 in endothelial cells facilitates tumor vascular normalization during immunotherapy. Diagrammatic representation of the mechanism by which PD-L1 blockade therapy modulates IFN-γ to influence the tumor vasculature in ECs through the ANGPT2-Tie2 signaling pathway. IFN-γ motivates the JAK1/2-STAT1 signaling pathway, thereby inhibiting the expression of ANGPT2 and Tie2. Following this process, phosphorylated STAT1 binds to the promoter regions of the TEK and ANGPT2 genes, thereby suppressing their gene expression. Furthermore, IFN-γ phosphorylates FOXO1 through activating the PI3K-AKT pathway, thereby reducing the nuclear activation of FOXO1 on the ANGPT2 gene. Consequently, the synthesis of the Tie2 and ANGPT2 proteins is diminished, contributing to the suppression of the ANGPT2-Tie2 signaling pathway. This effect subsequently suppresses the migration, proliferation, and tube formation of ECs, ultimately contributing to the suppression of tumor angiogenesis and vascular destabilization.