Fig 2: Senescence-induced canonical NF-?B signaling in endothelial cells regulates downstream signaling and lymphocyte recruitment. (A) Experimental setup: direct coculture of growing or RIS ER:HRASG12V IMR90 cells (asterisks) with HUVECs (arrowheads) expressing the I?Ba superrepressor (SR) or vector control. (B) Representative immunofluorescence of coculture with senescence-dependent IL8 expression in both CD31- IMR90s and CD31+ HUVECs. n = 5 biological replicates. Scale bar, 30 µm. (C) Separate quantification of IL8 positivity from the two cell types. Dots are individual replicates, and bars are means. Data were analyzed by one-way ANOVA with Sidak's multiple comparisons test; (****) P = 0.0001. (D) Experimental setup: HUVECs expressing the SR or vector control were incubated in CM from growing or RIS ER:HRASG12V IMR90 cells for 16 h before harvesting for flow cytometry for ICAM1 (middle panel) and ICOSLG (right panel). n = 3 biological replicates. (E) Experimental setup: HUVECs were incubated in CM from growing or RIS ER:HRASG12V IMR90 cells with vehicle or the indicated NF-?B inhibitors for 16 h before harvesting for flow cytometry for ICAM1. n = 3 biological replicates. (F) Experimental setup: LSECs or HUVECs with the indicated genetic or pharmacological NF-?B inhibitors were incubated in CM from growing or RIS ER:HRASG12V IMR90 cells for 16 h before performing a flow adhesion assay and analyses of lymphocyte adherence and trans-endothelial migration. (G) Representative photomicrographs of HUVECs with the indicated constructs and CM, showing adherent lymphocytes (black arrows). (H) Trans-endothelial migration of lymphocytes in the indicated cell lines and conditions. Dots are individual replicates, and bars are means. Data were analyzed by one-way ANOVA with Sidak's multiple comparisons test; (**) P = 0.01, (****) P = 0.0001.

Fig 3: Icos–Icosl signaling regulates senescence surveillance. (A) Experimental setup: Wild-type mice were injected with NRASG12V-containing transposons and then received intermittent intraperitoneal injection of anti-Icosl or matched IgG control on the indicated days before harvest on day 9 or 12 after HDTV injection. (B,C) Representative photomicrographs of RAS immunochemistry from the indicated conditions (B; scale bar, 100 µm) with quantification (C). Dots are individual mice, and bars are means. Data were analyzed by unpaired Student's t-test; (*) P = 0.05. (D) Experimental setup: Icos-DTR mice injected with NRASG12V-containing transposons then received intermittent intraperitoneal injection of diphtheria toxin (DTX) or vehicle control on the indicated days before harvest on day 12 after HDTV injection. (E,F) Representative photomicrographs of RAS immunochemistry from the indicated conditions (E; scale bar, 100 µm) with quantification (F). Dots are individual mice, and bars are means. Data were analyzed by one-way ANOVA with Sidak's multiple comparisons test; (****) P = 0.0001. (G) qRT-PCR analysis of ICOSLG expression relative to ACTB in samples from normal human livers (n = 5) or chronic liver disease of the indicated etiology (n = 10). Dots are individual patients, and bars are means. Data were analyzed by unpaired Student's t-test; (**) P = 0.01. (H) Immunoblotting analysis of ICOSLG expression in liver samples from normal human livers (n = 5) or chronic liver disease of the indicated etiology (n = 12). Dots are individual patients, and bars are means. Data were analyzed by unpaired Student's t-test; (**) P = 0.01. Example immunoblot of ICOSLG in samples from normal and CLD livers. (I) Kaplan–Meier analysis of overall survival for patients with HCC in the TCGA data set dichotomized by ICOSLG mRNA expression higher or lower than the median. Data were analyzed by log-rank test.

Fig 4: Identification of the m138 domains involved in the downmodulation of ICOSL and CD80.(A) Schematic representation (not drawn to scale) of m138-GFP protein mutants. (B) NIH3T3 ICOSL-HA cells were untransfected or transfected with m138-GFP, or m138-GFP mutant constructs and analyzed by flow cytometry for ICOSL and CD80 surface levels. Panels on the right represent the MFI of each sample minus the MFI of the control isotypes. Gray histograms represent the expression of untransfected cells and colored histograms represent the expression of cells transfected with m138-GFP (green), m138?Ig1-GFP (brown), m138?Ig2-GFP (blue), or m138?Ig2/3-GFP (red). The isotype for each antibody was used as a negative control (dotted lines). A representative experiment out of two performed is shown.

Fig 5: m138 leads to a reduction of the expression of ICOSL on the cell surface of infected cells but not of its overall intracellular levels.(A) NIH3T3 cells stably transfected with HA-ICOSL were mock-infected (gray histograms) or infected with MCMVwt or MCMV?m138 (brown or pink histograms, respectively) and analyzed by flow cytometry for surface expression of ICOSL and MHC I. (B) NIH3T3 HA-ICOSL cells non-transfected (gray histograms) or transfected with the m138-GFP construct (green histograms) were analyzed by flow cytometry for surface expression of ICOSL and CD44. In A and B, the isotype for each antibody was used as a negative control (dotted lines). (C) NIH3T3 HA-ICOSL cells were mock-infected (lane 2), treated with trypsin (lane 3), or infected with MCMVwt (lane 4) or MCMV?m138 (lane 5). Untransfected NIH3T3 cells were used as a control (lane 1). Samples were analyzed by western blot using an anti-HA mAb, followed by anti-rabbit IgG-HRP. An anti-IE1 and an anti-actin mAbs followed by anti-mouse IgG-HRP, were used as controls of MCMV infection and loading, respectively. (D) Flow cytometry analysis of NIH3T3 HA-ICOSL treated with trypsin (open histogram) or mock-treated (gray histogram) and stained with an anti-HA mAb or an isotype control (dotted line), followed by anti-mouse IgG PE. (E) NIH3T3 cells (lane 1) or NIH3T3 HA-ICOSL (lane 2) were surface labeled with biotin, immunoprecipitated with an anti-HA mAb, and analyzed by western blot analysis using streptavidin-HRP conjugate. (F) NIH3T3 HA-ICOSL cells were mock-infected (lane 2), or infected with MCMVwt (lane 3) or MCMVwt UV-inactivated (lane 4). Samples were lysed and processed as indicated in C. A lysate of untransfected NIH3T3 cells (lane 1) was also subjected to western blot and employed as a control. All infections were performed at an moi of 5 for 24 hr. Data are representative of at least two independent experiments. In C, E, and F, the size of the bands corresponding to ICOSL are indicated on the right margin in kilodaltons (kDa). In E, the sizes of the molecular weight markers are shown on the left side.