Depletion of endogenous tumor-associated regulatory T cells improves the efficacy of adoptive cytotoxic T-cell immunotherapy in murine acute myeloid leukemia

Abstract

Tumor-induced immune suppression can permit tumor cells to escape host immune resistance. To elucidate host factors contributing to the poor response of adoptively transferred tumor-reactive cytotoxic T lymphocytes (CTLs), we used a systemic model of murine acute myeloid leukemia (AML). AML progression resulted in a progressive regulatory T-cell (Treg) accumulation in disease sites. The adoptive transfer of in vitro-generated, potently lytic anti-AML-reactive CTLs failed to reduce disease burden or extend survival. Compared with non-AML-bearing hosts, transferred CTLs had reduced proliferation in AML sites of metastases. Treg depletion by a brief course of interleukin-2 diphtheria toxin (IL-2DT) transiently reduced AML disease burden but did not permit long-term survival. In contrast, IL-2DT prevented anti-AML CTL hypoproliferation, increased the number of transferred CTLs at AML disease sites, reduced AML tumor burden, and resulted in long-term survivors that sustained an anti-AML memory response. These data demonstrated that Tregs present at AML disease sites suppress adoptively transferred CTL proliferation, limiting their in vivo expansion, and Treg depletion before CTL transfer can result in therapeutic efficacy in settings of substantial pre-existing tumor burden in which antitumor reactive CTL infusion alone has proven ineffective.

Figures

Figure 1

Phenotype of in vitro cultured and activated CTLs. (A) CTLs were harvested on day 16 of in vitro cell culture. Flow cytometric analysis of CTLs was performed. Cells were gated on CD8 expression. CTLs displayed an activated and effector memory phenotype. (B) Lysis of C1498FFDsR target or irrelevant tumor target, EL-4, was performed with CTLs 16 days after in vitro culture for 5 hours. CTLs were capable of killing the AML target but not the EL-4 cells in vitro. Error bars represent SD.

Figure 2

CTLs fail to control advanced AML disease in vivo. B6 mice (10 mice/group) were injected with 106 C1498FFDsR cells through an intravenous route 14 days before being infused with 30 × 106 CTLs. Mice were then monitored for survival. Infusion of CTLs 14 days after tumor injection showed no reduction of AML burden (A, ■ vs ▴) or survival advantage (B, ■ vs ▴). Results from 1 of 5 representative experiments are shown.

Figure 3

AML progression results in increased Tregs present at the primary sites of AML metastases. B6 mice were injected intravenously with 106 C1498FFDsR cells and killed 6, 14, or 25 days after tumor injection. Flow cytometric analysis was performed on liver leukocytes and splenocytes. (A) Increased percentage of DsR+ tumor cells was found in the livers and spleens of mice. (B-C) An increased percentage of Foxp3+ Tregs in the CD4+ T-cell population was found in the livers of mice, shown in flow dot plot (B) and bar graph (C, P < .05). (D-E) An increased percentage of Foxp3+ Tregs in the CD4+ T-cell population was found in the spleens of mice, shown in flow dot plot (D) and bar graph (E, P < .05). (F) An increased number of tumor cells was correlated with an increased number of Treg cells in the livers and spleens of mice (P < .05). (G) An increased percentage of Tregs was found in the LNs (P < .05) and blood (P < .05) but not in the BM. Results from 1 of 3 representative experiments are shown. Error bars represent SD.

Figure 4

AML-associated Tregs reduced proliferation and IFN-γ secretion of adoptive transferred CTLs. (A) B6 mice were injected with 106 C1498FFDsR cells. Congenic B6-ly5.2 (CD45.1+) CTLs (30 × 106) were infused through an intravenous route 14 days after tumor injection or to naive mice. BrdU was added to the drinking water to track proliferation. Eighteen days after tumor injection, 4 mice per group were killed. Flow Cytometric Analysis (FACs) was done with liver leukocytes and splenocytes. Tumor-bearing mice had significantly reduced proliferation of CTLs compared with naive mice in the livers and spleens (P < .05). (B-C) CTLs (106) and Tregs (106) isolated from AML-bearing or naive mice were adoptively transferred to AML-bearing Rag−/− mice. Thirteen days after transfer, FACs was performed on splenocytes (B) and liver leukocytes (C). Cells were gated on CD8 expression for CTLs. Intracellular IFN-γ expression was measured on gated CTLs. Tregs from AML-bearing mice significantly reduced the percentage of IFN-γ–producing CTLs in the spleen and liver. (D) Tregs isolated from AML-bearing or naive mice were cocultured with OT I CD8+ T cells stimulated with SIINFEKL peptide (Ovap) for 6 days. Cell supernatant was harvested and the IFN-γ level was determined. Tregs from AML-bearing mice inhibited IFN-γ production by OT I CD8+ T cells (P < .01), whereas naive Tregs were not suppressive. Results from 1 of 2 representative experiments are shown. Error bars represent SD.

Figure 5

IL-2DT treatment followed by anti–AML-reactive CTLs significantly prolonged the survival of mice with advanced AML. B6 mice (10 mice/group) were injected with 106 C1498FFDsR cells followed by IL-2DT and CTL treatment as described. (A) Dot plots showing depletion of Tregs 1 day after IL-2DT treatment in the liver and spleen of mice. (B-C) Kinetics of Treg depletion by IL-2DT. (D) Combination of IL-2DT and CTL treatment significantly decreased tumor burden 17 and 24 days after tumor injection (■ vs □, P < .01). (E) IL-2DT treatment alone significantly prolonged the survival mice compared with either control mice or mice treated with CTLs alone (▵ vs ■ or ▿, P < .01). Combined IL-2DT and CTLs had superior effect (by log-rank test, P < .005) compared with either control mice or mice receiving IL-2DT or CTLs alone. Representative data from 1 of 3 similar experiments are shown. (F) Combed IL-2DT and CTL treatment promotes anti-AML memory response. Naive C57BL/6 or surviving mice (7-8 mice/group) were challenged with either EL-4 (105/dose) or C1498FFDsR (106/dose) cells. Both naive and surviving mice succumbed to EL-4 challenge. Significant increases in survival were observed only in rechallenged recipients of C1498FFDsR (■ vs ▵, P < .001). Error bars represent SD.

Figure 6

Combined IL-2DT treatment restored the proliferation and number of CTLs at sites of AML metastases. Naive or tumor-bearing mice were treated with IL-2DT or control DT (1 μg/dose) 4 and 13 days after tumor injection. Congenic B6-Ly5.2 (CD45.1+) CTLs (30 × 106/dose) were then injected intravenously 14 days after tumor injection and mice were fed with BrdU water to track proliferation. Eighteen days after tumor injection, 4 mice per group were killed. Flow cytometric analysis was done with liver leukocytes and splenocytes. IL-2DT treatment significantly augmented the percentage of BrdU+ CTLs in the spleens compared with control DT treatment. IL-2DT treatment significantly increased the percentage (A) and number (B) of CTLs found in the livers and spleens compared with control DT treatment. Representative data from 1 of 2 similar experiments are shown. Error bars represent SD.

Figure 7

Treg depletion by IL-2DT treatment augments CTL infiltration into sites of AML metastases. AML-bearing mice were treated with IL-2DT or control DT (1 μg/dose) 4 and 13 days after tumor injection. Congenic B6-Ly5.2 (CD45.1+) CTLs (30 × 106/dose) were then injected intravenously 14 days after tumor injection and organs were harvested 6 days after CTL infusion. Confocal microscopy was performed. Naive B6 mice were used as normal tissue controls. (A) CD4 (green) and Foxp3 (red); (B) CD45.1 (green) and DsR (red); (C) CD45.1 (green) and Foxp3 (red); and (D) CD45.1 (green) and CD31 (red). Representative data from 1 of 2 similar experiments are shown.