Peripheral blood T cells in acute myeloid leukemia (AML) patients at diagnosis have abnormal phenotype and genotype and form defective immune synapses with AML blasts

Abstract

Understanding how the immune system in patients with cancer interacts with malignant cells is critical for the development of successful immunotherapeutic strategies. We studied peripheral blood from newly diagnosed patients with acute myeloid leukemia (AML) to assess the impact of this disease on the patients' T cells. The absolute number of peripheral blood T cells is increased in AML compared with healthy controls. An increase in the absolute number of CD3+56+ cells was also noted. Gene expression profiling on T cells from AML patients compared with healthy donors demonstrated global differences in transcription suggesting aberrant T-cell activation patterns. These gene expression changes differ from those observed in chronic lymphocytic leukemia (CLL), indicating the heterogeneous means by which different tumors evade the host immune response. However, in common with CLL, differentially regulated genes involved in actin cytoskeletal formation were identified, and therefore the ability of T cells from AML patients to form immunologic synapses was assessed. Although AML T cells could form conjugates with autologous blasts, their ability to form immune synapses and recruit phosphotyrosine signaling molecules to the synapse was significantly impaired. These findings identify T-cell dysfunction in AML that may contribute to the failure of a host immune response against leukemic blasts.

Figure 1

CD3⁺/CD56⁺ PB T cells in AML compared with healthy donors. (A) Dot plot of CD3⁺/CD56⁺ T cells in PB of acute myeloid leukamia (AML) patients (n = 36) compared with healthy persons (n = 17). P value obtained using a Mann-Whitney unpaired 2-tailed t test. (B) Mean percentage ± SD of CD3⁺56⁺ and CD3⁺56⁻ cells expressing CD25 and CD69 in AML patients and healthy controls. CD25 and CD69 expression is significantly higher in CD3⁺56⁺ cells compared with CD3⁺56⁻ cells in healthy controls (■) and AML patients (□) and significantly higher in the CD3⁺56⁺ cells of AML patients (□) compared with healthy controls. (C) Size and granularity characteristics of CD3⁺56⁺ cells. The scatterplot shows CD3⁺56⁺ cells in blue backgated onto the FSC/SSC plot. Two populations of cells with different scatter characteristics are apparent. (D) Expression of perforin and granzyme B in the CD3⁺SSC^High and total CD3⁺56⁺ populations in these AML patients was compared with healthy persons. Although a higher mean percentage of CD3⁺56⁺ cells express cytotoxic granules compared with CD3⁺56⁻ cells, there is no difference in the percentage of cells that express these granules in AML (□) and healthy persons (■) until the LGL population is examined (CD3⁺SSC^High). Within this population, significantly fewer CD3⁺56⁺ cells express perforin and granzyme B compared with healthy controls.

Figure 2

Supervised analysis: AML versus healthy T cells. Heatmap created using the top 50 most significantly up- and down-regulated genes for CD4 and for CD8 in the comparison of AML versus healthy cells. Up-regulated genes are in red and down-regulated, in blue. The yellow bars on the gray vertical bar highlight genes that were also found to be differentially expressed in the CLL versus healthy cell experiment. Labeled genes are those involved in T-cell activation or signaling or with actin polarization.

Figure 3

AML T cells are able to conjugate effectively with autologous AML blasts but show impaired immune synapse formation and recruitment of T-cell signaling molecules. T cells from AML patients or age-matched healthy donors (healthy) were allowed to conjugate with autologous (auto) AML or healthy autologous B cells, respectively, ± sAg acting as APCs (CMAC dyed, blue). Conjugates were then fixed, stained, and scored by visual counting using confocal microscopy. (A) Percentage of T-cell–forming conjugates. (B) Percentage of T-cell conjugates resulting in F-actin (rhodamine phalloidin, red) polarization. (C) Percentage of T-cell conjugates resulting in phosphotyrosine polarization (P-Tyr, green) at the T-cell immune synapse. Data are the mean ± SD from 10 independent experiments with 50 conjugates analyzed per experiment. Arrows indicate protein localization at the T-cell–APC synapse site. Original magnification × 63. Statistical differences between experimental groups were evaluated by 2-tailed Student t test. P < .05 was considered statistically significant.

Figure 4

AML blasts exhibit defective APC function with healthy T cells in mixed allogeneic experiments. Healthy T cells were allowed to conjugate with allogeneic (allo) healthy CD34⁺ cells, healthy B cells, or AML blasts ± sAg acting as APCs (CMAC dyed, blue). Conjugates were then fixed and stained with rhodamine phalloidin to detect F-actin (red). Conjugates were selected at random for imaging and were scored for F-actin polarization at the immune synapse. Each dataset is the mean ± SD from 10 independent patient experiments with at least 50 conjugates analyzed per experiment. Arrows indicate F-actin accumulation at the T-cell–APC synapse site. Note the equivalent APC function of healthy B cells and healthy CD34⁺ cells acting as control cells. Original magnification × 63. Statistical differences between experimental groups were evaluated by 2-tailed Student t test. P < .05 was considered statistically significant.

Figure 5

AML T cells exhibit impaired immunologic synapse formation with healthy APCs in mixed allogeneic experiments. T cells from AML patients or age-matched healthy donors (healthy) were allowed to conjugate with allogeneic (allo) healthy B cells ± sAg acting as APCs (CMAC dyed, blue). Conjugates were then fixed, stained, and scored for F-actin (rhodamine phalloidin, red) polarization at the T-cell immune synapse. Data are the mean ± SD from 10 independent experiments with 50 conjugates analyzed per experiment. Arrows indicate protein localization at the T-cell–APC synapse site. Colocalization of F-actin (red) and phosphotyrosine (P-Tyr, green) in the upper right merge images is shown in yellow. Original magnification × 63. Statistical differences between experimental groups were evaluated by 2-tailed Student t test. P < .05 was considered statistically significant