PD-L1 mediates Foxp3⁺ iT reg cell development. (A and B) Development of Foxp3⁺ iT reg cells was assessed by flow cytometric analysis of Foxp3-GFP expression after co-culture of naive CD4⁺CD62L⁺Foxp3.GFP⁻ T cells with anti-CD3 and irradiated WT or PD-L1^−/− APCs plus the indicated range of TGF-β concentrations for 3 d (A) or PD-L1–Ig or control Ig (human IgG1)–coupled beads (B). One representative experiment of at least three similar experiments is shown. (C) Analysis of Foxp3-GFP expression after culture of naive CD4⁺CD62L⁺Foxp3.GFP⁻ T cells with PD-L1–Ig beads and over the indicated range of TGF-β concentrations. *, P < 0.001 for PD-L1 bead comparing 0 ng/ml TGF-β versus 0.5–8 ng/ml; **, P < 0.001 comparing PD-L1 bead versus control bead at 0.5 ng/ml TGF-β. Data represent the mean ± SD and are representative of at least four independent experiments. (D) Naive CD4 T cells were cultured in the presence of control Ig or PD-L1–Ig beads (with increasing amounts of PD-L1) in the presence of low levels of TGF-β. *, P = 0.05; **, P = 0.0006; ***, P < 0.0001. Data are representative of three similar experiments. (E) Analysis of CFSE dye dilution of naive CD4⁺CD62L⁺CD25⁻ CD44^low 2D2 Tg T cells cultured with increasing quantities of PD-L1–Ig (PD-L1 bead is coated with anti-CD3, anti-CD28, and 20, 40, or 60% PD-L1–Ig, respectively) or control Ig in the presence of TGF-β. After 3 d of co-culture, cells were stained for Foxp3 expression. 20 U/ml of exogenous IL-2 was added to cultures shown on the bottom (green). Data are representative of three similar experiments. (F) Quantitative analysis of Foxp3⁺ T cell conversion in the presence or absence of IL-2 and increasing quantities of immobilized PD-L1. *, P < 0.0001 for all concentrations of PD-L1–Ig versus control Ig in the absence of IL-2; *, P = 0.005 and **, P = 0.0021 for 40 and 60% PD-L1–Ig, respectively, versus control Ig, in the presence of IL-2. Data are representative of three similar experiments. (G) Analysis of Foxp3 expression on a per cell basis, quantified via MFI. *, P = 0.0003 and **, P < 0.001, PD-L1–Ig 20% and PD-L1–Ig 40% versus control Ig, in the absence of IL-2; *, P = 0.03, PD-L1–Ig 60 versus control Ig, in presence of IL-2. Data represent the mean ± SD and are representative of more than five independent experiments with n = 3 mice per treatment condition.

PD-L1 maintains Foxp3 expression by iT reg cells during suppression of effector cell function. (A) Schematic depiction of experiment. Naive CD4⁺CD62L^hiFoxp3.GFP⁻ T cells were induced toward T reg cell differentiation for 3 d in the presence of TGF-β plus IL-2 and either control or PD-L1 beads. Foxp3.GFP⁺CD45.1⁻ cells were then sorted and co-cultured with sorted CD4⁺CD25⁻CD45.1⁺ in the presence of either control or PD-L1 beads during the 3-d suppression assay. (B) 72 h after co-culture, CD4⁺CD45.1⁻ cells were gated and analyzed for GFP expression. (C) Quantification of experiment depicted in B. Data represent the mean ± SD and are representative of at least two independent experiments.

Attenuated iT reg cell development in PD-L1^−/−PD-L2^−/− mice in vivo. (A) CD4⁺CD62L^hiFoxp3.GFP⁻ cells were adoptively transferred i.v. into the tail veins of WT Rag^−/− or PD-L1^−/−PD-L2^−/−Rag^−/− mice. Spleens and lymph nodes were analyzed for Foxp3.GFP expression 14–17 d after transfer. (B) Quantitation of Foxp3.GFP expression from independent mice depicted in A. Data represent the mean ± SE of five independent mice. (C and D) Analysis of IL-17⁺ and IFN-γ⁺ T eff cells by intracellular cytokine staining (C) and ratios of IL-17–producing T eff/T reg cells and IFN-γ–producing T eff/T reg cells from WT Rag^−/− or PD-L1^−/−PD-L2^−/−Rag^−/− mice 14–17 d after transfer (D). Data represent the mean ± SD of n = 5 mice per group and are representative of two independent experiments.

PD-L1 regulates T reg cell development by antagonizing the Akt–mTOR signaling cascade. (A, C, F, and G) Phospho-Akt, phospho-mTOR, PTEN, and phospho-S6 analysis at 18 h after culture with control Ig bead (hIgG = 60% of bead surface, with remaining surface coated with anti-CD3 and anti-CD28) or various titers of PD-L1–Ig bead (PD-L1–Ig 20, 40, 60 = 20, 40, and 60% of bead surface coated with PD-L1–Ig, with remaining surface coated with anti-CD3 and anti-CD28 plus control Ig). (B, D, E, and G) MFI analysis of phospho-Akt (B; *, P = 0.001; **, P = 0.003; and ***, P = 0.0008, at 20, 40, and 60% PD-L1, respectively, compared with control Ig), phospho-mTOR (D; *, P = 0.0064; and **, P = 0.0001, at 20 and 60% PD-L1, respectively, compared with control Ig), phospho-S6 (E; P = 0.0012, P = 0.0007, and P = 0.0002, at 20, 40, and 60% PD-L1, respectively, compared with control Ig), and PTEN (H; *, P = 0.0378, PD-L1–Ig 40 compared with control Ig) at 18 h. n = 3 mice per experiment, representative of three experiments. Data are representative of the MFI ± SD and are representative of three experiments.

PD-L1–induced CD4⁺ Foxp3⁺ T reg cells suppress CD4⁺ T eff cells in vitro. (A) PD-L1 iT reg cell function was assessed by [³H]thymidine incorporation of naive CD4⁺CD25⁻ T eff cells after 3 d of co-culture at a 1:1 T reg/T eff cell ratio plus PD-L1 beads (5:1 bead/T eff cell ratio). Data represent the mean ± SD and are representative of at least two independent experiments. (B) PD-L1 iT reg cell function was assessed by CFSE dilution of naive CD4⁺CD25⁻ T eff cells after 3 d of incubation with 1:1 T reg/T eff cell ratio and PD-L1 beads (5:1 bead/T eff cell ratio). Data represent the mean ± SD and are representative of three similar experiments. (C) Quantification of T eff cell proliferation in B, analyzing the division index of gated CD4⁺CD45.1⁺ (the number of divisions a single cell has divided) by FlowJo software. Data represent the mean ± SD.

PD-L1 enhances the efficiency of iT reg cell–mediated suppression of T eff cells. (A) Foxp3.GFP⁺ iT reg cells were sorted and co-cultured with naive CD4⁺CD25⁻CD45.1⁺ T eff cells plus either PD-L1–Ig beads or control Ig beads (at various T reg/T eff cell ratios). 72 h later, cultures were pulsed with [³H]thymidine for 12–14 h. P < 0.0009 at a 1:4 ratio cultured with PD-L1 beads (comparing T eff + iT reg vs. T eff cells). Data represent the mean proliferation ± SD and are representative of at least four independent experiments. (B) Quantification of suppression at 1:4 ratio of T reg/T eff cells. P = 0.0149. Data represent the mean ± SD and are representative of at least four independent experiments. (C) PD-L1 iT reg cells suppress T eff cells more effectively than control iT reg cells. CD4⁺CD62L⁺ FoxP3.GFP⁻ naive T cells were induced with PD-L1 or control beads in the presence of TGF-β. GFP⁺ iT reg cells were sorted and co-cultured at the indicated T reg/T eff cell ratios with CFSE-labeled CD4⁺CD25⁻ Thy1.1 T eff cells and beads coated with anti-CD3 and anti-CD28 (in the absence of PD-L1) for 3 d. Graphs are representative of three experimental replicates and data are representative of three independent experiments.

Dramatic weight loss, severe pulmonary inflammation, and fatal inflammatory disorder develop in PD-L1^−/−PD-L2^−/− Rag^−/− mice after adoptive transfer of naive CD4⁺CD62L^hiFoxp3.GFP⁻ T cells. Sorted CD4⁺CD62L^hiFoxp3.GFP⁻ cells were adoptively transferred i.v. into WT Rag^−/− or PD-L1^−/−PD-L2^−/− Rag^−/− mice. (A and B) Clinical manifestations shown are the following: percentage of weight loss of mice after adoptive transfer of CD4⁺CD62L^hiFoxp3.GFP⁻ cells into WT Rag^−/− or PD-L1^−/−PD-L2^−/−Rag^−/− (P < 0.001; n = 5 mice per group; A) and quantified lymph node (axillary, brachial, and inguinal) cellularity (B). Data represent the mean ± SD and represent two independent experiments. (C) Survival of mice after adoptive transfer of naive CD4⁺CD62L^hiFoxp3.GFP⁻ T cells was monitored for 30 d. PD-L1^−/−PD-L2^−/− Rag^−/−, n = 12; WT Rag^−/−, n = 8. (D) Hematoxylin and eosin–stained paraffin sections of lung tissue obtained on days 14–17 after transfer of naive CD4⁺CD62L^hiFoxp3⁻ T cells. Bars: (top, 40×) 500 µm; (bottom, 400×) 50 µm. PD-L1^−/−PD-L2^−/− Rag^−/−, n = 9; WT Rag^−/−, n = 10. Data represent more than four independent experiments.