Activated Allogeneic NK Cells Preferentially Kill Poor Prognosis B-Cell Chronic Lymphocytic Leukemia Cells

Abstract

Mutational status of TP53 together with expression of wild-type (wt) IGHV represents the most widely accepted biomarkers, establishing a very poor prognosis in B-cell chronic lymphocytic leukemia (B-CLL) patients. Adoptive cell therapy using allogeneic HLA-mismatched Natural killer (NK) cells has emerged as an effective and safe alternative in the treatment of acute myeloid and lymphoid leukemias that do not respond to traditional therapies. We have described that allogeneic activated NK cells eliminate hematological cancer cell lines with multidrug resistance acquired by mutations in the apoptotic machinery. This effect depends on the activation protocol, being B-lymphoblastoid cell lines (LCLs) the most effective stimulus to activate NK cells. Here, we have further analyzed the molecular determinants involved in allogeneic NK cell recognition and elimination of B-CLL cells, including the expression of ligands of the main NK cell-activating receptors (NKG2D and NCRs) and HLA mismatch. We present preliminary data suggesting that B-CLL susceptibility significantly correlates with HLA mismatch between NK cell donor and B-CLL patient. Moreover, we show that the sensitivity of B-CLL cells to NK cells depends on the prognosis based on TP53 and IGHV mutational status. Cells from patients with worse prognosis (mutated TP53 and wt IGHV) are the most susceptible to activated NK cells. Hence, B-CLL prognosis may predict the efficacy of allogenic activated NK cells, and, thus, NK cell transfer represents a good alternative to treat poor prognosis B-CLL patients who present a very short life expectancy due to lack of effective treatments.

Keywords: allogeneic NK cells; bad prognosis leukemia; chronic lymphocytic leukemia; leukemia resistance; mismatch.

Figure 1

Allogeneic NK cells require activation to kill B-CLL cells. (A) Jurkat and R69 cells or cells from B-CLL patients were incubated with naïve (left) or R69-LCL activated (right) NK cells (MACS enriched, >90% CD56⁺CD3⁻ cells; CTGreen labeled) for 4 h at 9:1 effector:target ratio as described in Section “Materials and Methods.” Subsequently, PS traslocation (annexin V DY634) and membrane permeabilization (7AAD uptake) were analyzed by flow cytometry in the cell population negative for CTGreen. (B) Cells from B-CLL were incubated with R69- or 721-LCL-activated NK cells (MACS enriched, >90% CD56⁺CD3⁻ cells; CTGreen labeled) for 4 h at 9:1 effector:target ratio as described in Section “Materials and Methods.” Subsequently, PS traslocation (annexin V DY634) and membrane permeabilization (7AAD uptake) were analyzed by flow cytometry in the cell population negative for CTGreen. Data in the graphics are represented as the mean ± SEM from three independent NK cell donors (A) or from 14 independent B-CLL patients (B). Annexin V⁺ cells represent the % of AnnexinV⁺7AAD⁻ plus AnnexinV⁺7AAD⁺ cells. Statistical analysis was performed by comparing the means of naive versus activated NK cells within each group using 2-way ANOVA with Tukey HSD post hoc test; ns, not significant, *p < 0.05, **p < 0.01, ***p < 0.001. (C) Representative dot plot of Annexin V/7AAD staining in a B-CLL sample incubated in medium alone (control) or together with NK cells (NK). Numbers correspond to the % of cells in each quadrant.

Figure 2

Analyses of the cytotoxic potential of activated allogeneic NK cells employing six CLL patients and seven NK cell donors. Cells from B-CLL patients were incubated with R69-activated NK cells (MACS enriched, >90% CD56⁺CD3⁻ cells; CTGreen labeled) from seven independent donors for 4 h at 9:1 effector:target ratio in two independent experiments as described in Section “Materials and Methods.” Subsequently, PS traslocation (annexin V DY634) and membrane permeabilization (7AAD uptake) were analyzed by flow cytometry in the cell population negative for CTGreen. (A) The graph represents the % of annexin V positive cells in the 42 combinations (six B-CLL × seven NK). Annexin V⁺ cells represent the % of Annexin V⁺7AAD⁻ plus AnnexinV⁺7AAD⁺ cells. Each symbol represents a NK cell donor. All NK cell donors were incubated with every B-CLL sample. (B) Cell death (Annexin V⁺) in every B-CLL sample was represented in a boxplot where median ± SD is indicated. Statistical analysis was performed employing a regression tree in which patient samples are clustered according to their similarity of sensitivity to NK cell cytotoxicity.

Figure 3

Analyses of the expression of activating and inhibitory ligands of NK cell receptors in B-CLL cells. Expression of NKp30, NKp46, and NKG2D activating ligands using Fc quimeras (A), HLA-ABC and HLA-E inhibitory ligands (B), and the adhesion molecule ICAM-1 (C) using specific antibodies were analyzed in B-CLL cells by flow cytometry as described in Section “Materials and Methods.” The Mean fluorescence intensity (MFI) of every B-CLL sample is represented in the graphs. MFI = (MFI specific Ab) − (MFI isotype control).

Figure 4

Correlation between matched B-CLL/NK cell combinations and cell death. The data in Figure 2 were now represented as matched and mismatched combinations. The graph represents the % of annexin V⁺ cells in the 42 combinations (six B-CLL × seven NK) as described in legend to Figure 2 separated in two independent experiments (NK1-4 and NK5-7). White and solid symbols correspond to mismatched and matched combinations, respectively. Statistical analyses and results are described in Tables 4 and 5.

Figure 5

Poor prognosis B-CLL cells are more susceptible to allogeneic R69-LCL-activated NK cells than good prognosis samples. R69-LCL-activated allogeneic NK cells (MACS enriched, >90% CD56⁺CD3⁻ cells; CTGreen labeled) were incubated for 4 h at 9:1 effector:target ratio with B-CLL cells from three groups of patients classified according to prognosis based on the mutational status of TP53 and IGHV (good: TP53^wtIGHV^mut, n = 9; intermediate: TP53^wtIGHV^wt, n = 8; bad: TP53^mutIGHV^wt, n = 5). Subsequently, PS traslocation (annexin V DY634) and membrane permeabilization (7-AAD uptake) were analyzed by flow cytometry in the cell population negative for CTGreen. Annexin V⁺ cells represent the % of AnnexinV⁺7AAD⁻ plus AnnexinV⁺7AAD⁺ cells. Every B-CLL sample was incubated with NK cells from the several healthy donors and the % of Annexin V⁺ cells for every B-CLL sample were represented in a boxplot where median ± SD is indicated. Statistical analysis was performed by comparing the means of every group using 1-way ANOVA with Tukey HSD post hoc test; ns, not significant, *p < 0.05, ***p < 0.001.