Fig. 3: Differential PD-L1 signaling on LEC regulates Treg and Teff TEM.

a–d TEM assays. LEC treated with 2 μg/ml intact (a), F(ab)’2 or Fab (b) anti-mouse PD-1 (Rmp1-14), or with 5 μg/ml anti-mouse PD-L1 (10 F.9G2), anti-mouse PD-L1 (10 F.2H11), or rat IgG (2A3), loaded with iTregs or Teffs (as in Fig. 2). Human LECs (c) treated with 5 μg/mL anti-human PD-1 mAb (EH12.2H7), anti-human PD-L1 mAb (29E.2A3), or mouse IgG1, then loaded with activated human iTregs or Teffs (d) Flow cytometry for PD-L1 on iTreg or Teff migration across wild-type (WT) or PD-L1^−/−LEC. e Immunohistochemistry for ZO-1, F-actin, and VCAM-1 expression in resting WT or PD-L1^−/− LECs. f Flow cytometry for PD-L1, VCAM-1, and Lyve-1 in resting WT or PD-L1^−/− LECs. g Whole-mount ear staining for VE-cadherin, ZO-1, and Lyve-1 in WT and PD-L1^−/− mice. LV lymphatic vessel, BV blood vessel. Arrow heads indicate button junctions; arrows indicate zipper junctions. h Immunohistochemistry for ZO-1 and VE-cadherin in WT or PD-L1^−/− LECs after iTreg TEM (LECs from same experiment as panel d). Quantification of zipper junctions in LVs or LECs. Each dot represents one LV or every 10 LECs (g, h). i Flow cytometry analysis of PD-L1 expression on LEC stimulated with 100 ng/mL mouse IFNγ or 20 ng/mL mouse TNFα for 16 h at 370 °C. Representative histograms shown. j iTregs loaded to Boyden chamber with LECs treated as in i. k Immunoblots for Akt, NFκB-p65, and ERK phosphorylation in LECs stimulated with 1 μg/mL mouse PD-1 Ig, mouse CD80 Ig, or human IgG1 for the indicated times. Relative band intensities shown. Data representative of 3 independent experiments (a–k). Magnification ×60 (e, g, h); scale bar 42 μm (e, h). 14 μm (g), MFI shown (e–i). Mean ± SEM (a–k). *p < 0.01 by one-way ANOVA with Sidak’s multiple comparisons test (a–c, i, j); *p < 0.01 versus WT (d–h) or hIgG1 (k) by unpaired, two-tailed t-test with Welch’s correction. Source data are provided as a Source Data file.