Program death-1 signaling and regulatory T cells collaborate to resist the function of adoptively transferred cytotoxic T lymphocytes in advanced acute myeloid leukemia

Abstract

Tumor-induced immune defects can weaken host immune response and permit tumor cell growth. In a systemic model of murine acute myeloid leukemia (AML), tumor progression resulted in increased regulatory T cells (Treg) and elevation of program death-1 (PD-1) expression on CD8(+) cytotoxic T cells (CTLs) at the tumor site. PD-1 knockout mice were more resistant to AML despite the presence of similar percentage of Tregs compared with wild type. In vitro, intact Treg suppression of CD8(+) T-cell responses was dependent on PD-1 expression by T cells and Tregs and PD-L1 expression by antigen-presenting cells. In vivo, the function of adoptively transferred AML-reactive CTLs was reduced by AML-associated Tregs. Anti-PD-L1 monoclonal antibody treatment increased the proliferation and function of CTLs at tumor sites, reduced AML tumor burden, and resulted in long-term survivors. Treg depletion followed by PD-1/PD-L1 blockade showed superior efficacy for eradication of established AML. These data demonstrated that interaction between PD-1 and PD-L1 can facilitate Treg-induced suppression of T-effector cells and dampen the antitumor immune response. PD-1/PD-L1 blockade coupled with Treg depletion represents an important new approach that can be readily translated into the clinic to improve the therapeutic efficacy of adoptive AML-reactive CTLs in advanced AML disease.

Figures

Figure 1

PD-1 expressing CD8+ T cells in the liver of AML-bearing mice displayed impaired function. B6 mice were injected intravenously with 106 C1498FFDsR cells and killed 14, 20, or 25 days after tumor injection. Flow cytometric analysis was performed on liver leukocytes and splenocytes (A-C). (A) Flow dot plot of PD-1 expression on liver CD8+ T cells 25 days after AML injection. (B) PD-1 was up-regulated on liver CD8+ T cells at late phase of AML progression. (C) PD-1 expression was not found on the spleen CD8+ T cells of AML-bearing mice. (D) PD-L1 expression on C1498FFDsR left untreated or treated with IFN-γ for 48 hours was assessed by flow cytometric analysis. PD-L1 expression was found on C1498FFDsR, and mean fluorescent intensity was increased by IFN-γ treatment. (E) Percentage of IFN-γ–producing PD-1+/PD-1− CD8+ T cells was determined by flow cytometry analysis. Percentage of IFN-γ–secreting cells. PD-1+ fraction was much lower than the PD-1− fraction. (F) Liver samples from naive mice (left side) or AML-bearing mice 25 days after C1498FFDsR injection (right side) were evaluated by immunofluorescence staining. Slides were mounted with VECTASHIELD (Vector Laboratories) and images were taken at 40×/1.30 oil objective through Olympus UPlanApo oil lens and an Olympus FV500 camera, compiled with Fluoview software (Version 4.3), then cropped in Adobe Photoshop CS3. (Top panels) Colocalization of tumor cells (DsR+, blue), CD8+ T cells (CD8+, green), and Tregs (Foxp3+, red) is designated by white arrows. (Bottom panels) Colocalization of tumor cells, CD8+ T cells, and PD-1 (red) is designated by white arrows. Results from one of 5 representative experiments are shown. Bar graphs represent mean ± SD.

Figure 2

PD-1 KO mice were more resistant to AML. B6 PD-1 KO and B6 WT mice were injected intravenously with 106 C1498FFDsR cells. Mice (8-10 mice/group) were monitored using BLI, analyzed for survival, or killed (3 or 4 mice/group) at 25 days after AML injection for immune parameter determination. (A) B6 PD-1 KO versus WT mice had much lower tumor burdens (A) and significantly prolonged survival time (B) compared with WT mice (C). The number of Foxp3+ Tregs per liver in WT naive mice was significantly lower than in mice with AML. The total number of Tregs in naive PD-1 KO mice was significantly higher than WT mice and did not further increase by AML challenge. (D) The ratio of CD8+ T cells to Tregs was significantly reduced in AML-bearing WT versus naive WT mice, whereas the CD8+ T cell-to-Treg ratio was significantly higher in naive PD-1 KO versus WT mice because of a proportionately higher number of CD8+ T cells in PD-1 KO versus WT mice (data not shown). Challenging PD-1 KO with AML did not significantly alter this ratio. Results from one of 3 representative experiments are shown. Bar graphs represent mean ± SD.

Figure 3

Blocking PD-1/PD-L1 interaction reduced Treg-mediated suppression of CD8+ T cells in vitro. Treg suppression assay was performed as described. Proliferating CD8+ T cells were quantified by the dilution of CFSE resulting in a CFSElow population and IFN-γ secretion was measured by enzyme-linked immunosorbent assay. CFSE-labeled CD8+ T cells were cocultured with syngeneic, anti-CD3ϵ mAb-loaded APCs with or without Tregs and/or anti–PDL1 mAb as indicated. CD8+ T cells, Tregs, and APCs were obtained from either WT or B6 PD-1 KO mice, as indicated. PD-1 KO CD8+ T cells are more resistant to WT Treg-mediated suppression of proliferation (A) and IFN-γ secretion (B) than WT CD8+ T cells. Compared with WT Tregs, PD-1 KO Tregs have a defective capacity to suppress the proliferation (C) and IFN-γ secretion (D) of CD8+ T cells. Anti–PD-L1 mAb prevents WT Treg suppression of proliferation (E) and IFN-γ secretion (F) of WT CD8+ T cells. Neither PD-1 KO nor WT Tregs are able to suppress proliferation of CD8+ T cells in the presence of PD-L1 KO APCs (G). Results from one of 3 representative experiments are shown. Bar graphs represent mean ± SD.

Figure 4

Anti–PD-L1-blocking mAb restored endogenous T-cell function and prolonged surviving time of AML-bearing mice. B6 mice (10 mice/group) were injected with 106 C1498FFDsR cells followed by anti–PD-L1 mAb (□) or rat IgG (■) treatment as described. In some studies, mice (3 or 4 mice /group) were killed 20 days after AML injection for the determination of proliferation by BrdU incorporation or 25 days after AML injection for determination of function by flow cytometric analysis. (A) Anti–PD-L1 mAb-treated mice had significantly prolonged survival compared with rat IgG control (□ vs ■, P = .029). (B) Anti–PD-L1 mAb treatment significantly increased the proliferation of CD8+ T cells in the liver of AML-bearing mice. (C) Anti–PD-L1 mAb treatment significantly increased the percentage of IFN-γ–secreting cells in the liver. (D) Anti–PD-L1 mAb treatment enhanced IFN-γ production by PD-1+ (left graph) but not PD-1− (right graph) cells in the liver. Results from one of 3 representative experiments are shown. Bar graphs represent mean ± SD.

Figure 5

Anti–PD-L1 mAb treatment enhanced the resistance of adoptively transferred anti-AML CTLs to Treg-induced suppression, resulting in increased long-term survival. B6 mice (10 mice/group) were injected with 106 C1498FFDsR cells followed by anti–PD-L1 mAb (every other day from days 10-20) and CTL treatment (day 14) as described. (A) Anti–PD-L1 mAb treatment alone significantly prolonged the survival of mice compared with either control mice or mice treated with CTLs alone, although all mice died of AML (▵ vs ■ or ▿, P < .01). Combined anti–PD-L1 mAb and anti-AML CTLs had a significantly greater survival benefit (P < .005) compared with either control mice (■) or mice receiving anti–PD-L1 mAb (▵) or CTL alone (▿). Combined therapy permitted 30% of mice with advanced AML to survive long-term. (B-C) B6 mice were injected with 106 C1498FFDsR cells. A total of 30 × 106 congenic B6-ly5.2 (CD45.1+) CTLs and anti–PD-L1 mAb treatment were given as described. BrdU was added to the drinking water to track proliferation. On days 20 or 25 after tumor injection, 4 mice per group were killed. Flow cytometry was done with liver leukocytes (B-C). (B) Anti–PD-L1 mAb treatment significantly augmented the percentage of BrdU+ adoptively transferred CTLs in the liver of mice compared with control mice. (C) Intracellular IFN-γ was determined for adoptively transferred CTLs in the liver of mice with advanced AML. Anti–PD-L1 mAb treatment significantly increased the percentage of IFN-γ–secreting CD8+ T cells. (D) Anti–PD-L1 mAb treatment did not alter the percentage of Foxp3+ Tregs found in the liver of AML-bearing mice. (E) A total of 106 CTLs and 106 Tregs isolated from AML-bearing mice were adoptively transferred to AML-bearing Rag−/− mice. Anti–PD-L1 mAb was given as described. Thirteen days after CTL transfer, flow cytometric analysis was performed on adoptively transferred CTLs in the liver. Intracellular IFN-γ expression was measured on gated, transferred CTLs. Tregs obtained from AML-bearing primary recipients significantly reduced the percentage of IFN-γ, producing adoptively transferred CTLs in AML-bearing secondary recipients. Anti–PD-L1 mAb treatment significantly increased the percentage of IFN-γ–secreting transferred CTLs, similar to a level of transferred CTLs without Treg cotransfer. Results from one of 3 representative experiments are shown. Bar graphs represent mean ± SD.

Figure 6

Combined IL-2DT and anti–PD-L1 mAb had superior antitumor effect to either treatment alone and, when coupled with anti-AML CTLs, more rapidly and uniformly reduced AML tumor burden. B6 mice (10 mice/group) were injected with 106 C1498FFDsR cells followed by IL-2DT, anti–PD-L1 mAb, and/or anti-AML CTL treatment as described. Whole body imaging was performed to determine tumor burden 19, 26, and 33 days after AML injection. (A) Tumor burden was decreased by combination therapy. IL-2DT, but not anti–PD-L1 mAb alone, significantly reduced tumor burden at an early time point (day 19). Combined IL-2DT and anti–PD-L1 had superior effect compared with either treatment alone. Although CTL therapy alone did not significantly decrease AML tumor burden at any time point, adding CTL therapy to IL-2DT and anti–PD-L1 mAb treatment resulted in the greatest and most consistent decrease in AML tumor burden throughout the 33-day post-AML challenge observation period. *P < .05 compared with nontreated AML controls. Error bars represent SEM. (B) BLI studies of 5 mice per group receiving AML with or without IL-2DT, anti–PD-L1 mAb, and CTL therapy are shown. On day 19, 2 of 5 mice receiving IL-2DT and anti–PD-L1 mAb had detectable tumor versus 0 of 5 mice also receiving CTLs. (C) IL-2DT or anti–PD-L1 mAb treatment alone significantly prolonged the survival mice compared with either control mice or mice treated with CTLs alone (♦ and ▴ vs ■ or ▾, P < .05). Combined IL-2DT with CTLs or anti–PD-L1 mAb with CTLs had superior effect compared with either control mice (■, P < .05) or mice receiving IL-2DT (♦, P < .05), anti–PD-L1 mAb (▴, P < .05), or CTL single treatment (▾, P < .05). Combined IL-2DT and anti–PD-L1 mAb with (●) or without CTLs (▿) had the best survival outcome compared with single treatment or in combination with CTLs.