Natural Killer Cell Subpopulations and Inhibitory Receptor Dynamics in Myelodysplastic Syndromes and Acute Myeloid Leukemia

Abstract

Natural killer (NK) cells are key innate immunity effectors that play a major role in malignant cell destruction. Based on expression patterns of CD16, CD56, CD57, and CD94, three distinct NK cell maturation stages have been described, which differ in terms of cytokine secretion, tissue migration, and the ability to kill target cells. Our study addressed NK cell maturation in bone marrow under three conditions: a normal developmental environment, during pre-leukemic state (myelodysplastic syndrome, MDS), and during leukemic transformation (acute myeloblastic leukemia, AML). In this study, we used a new tool to perform multicolor flow cytometry data analysis, based on principal component analysis, which allowed the unsupervised, accurate discrimination of immature, mature, and hypermature NK subpopulations. An impaired NK/T cell distribution was observed in the MDS bone marrow microenvironment compared with the normal and AML settings, and a phenotypic shift from the mature to the immature state was observed in NK cells under both the MDS and AML conditions. Furthermore, an impaired NK cell antitumor response, resulting in changes in NK cell receptor expression (CD159a, CD158a, CD158b, and CD158e1), was observed under MDS and AML conditions compared with the normal condition. The results of this study provide evidence for the failure of this arm of the immune response during the pathogenesis of myeloid malignancies. NK cell subpopulations display a heterogeneous and discordant dynamic on the spectrum between normal and pathological conditions. MDS does not appear to be a simple, intermediate stage but rather serves as a decisive step for the mounting of an efficient or ineffective immune response, leading to either the removal of the tumor cells or to malignancy.

Keywords: acute myeloid leukemia; bone marrow; inhibitory receptors; killer immunoglobulin-like receptors; myelodysplastic syndrome; natural killer cell maturation; natural killer cells.

Figure 1

Representative example for NK analysis strategy and classification into three subsets: CD56^bright CD94^hi CD16⁻ CD57⁻ as the immature subset (yellow dots); CD56^dim CD94^med CD16⁺ CD57⁻ as the mature subset (green dots); and CD56^dim CD94^low CD16⁺ CD57⁺ as the hypermature subset (blue dots). (A) Lymphocytes gating on the side scatter (SSC) vs CD45 dot-plots. (B) NK cells were gated according to CD56 expression and the absence of CD3 and CD19. (C, D) Reliable separation of NK cells into subsets was obtained using the Principal Component Analysis (PCA) for all analyzed cases, regardless of the group. Dots and circles represent the median values of individual cases, and the solid line represents the 2 SD curve for an NK subset (yellow, CD56^bright CD94^hi CD16⁻ CD57⁻ represents the immature subset; green, CD56^dim CD94^med CD16⁺ CD57⁻ represents the mature subset; and blue, CD56^dim CD94^low CD16⁺ CD57⁺ represents the hypermature subset). The table shows the contribution of the most relevant parameters (those markers that received a weight over 10) to the first (PC1, x-axis) or second (PC5, y-axis) principal component reflected as percent values. (E) Distribution of CD159a and KIR receptors inside of the different NK cell subpopulations for the presented case.

Figure 2

Regression statistics describing the relationship between bone marrow NK, T, and B cell counts for the three case groups: NBM, MDS, and AML. Correlation coefficients (R) were computed for (A) NK and T cell counts in NBM cases (n = 30); (B) NK and T cell counts in MDS (n = 25); (C) NK and T cell counts in AML (n = 8); (D) NK and B cell counts in NBM cases (n = 30); (E) NK and B cell counts in MDS (n = 25); (F) NK and B cell counts in AML (n = 8). Data are presented as scatter plots. Spearman test was used to analyze the significance of the identified correlations (***P < 0.001, *P < 0.05, ns, not significant).

Figure 3

Comparison of cell percentages of distinct NK maturation subsets within the bone marrow microenvironment of NBM, MDS, and AML cases. (A) Percentage of bone marrow CD56^bright CD94^hi CD16^- CD57^- immature NK cells in normal bone marrow (NBM, n = 30), myelodysplastic syndromes (MDS, n = 25), and acute myeloid leukemias (AML, n = 8) cases. Bars represent the mean ± SEM (*P < 0.05, ns, not significant; one-tailed unpaired t-test). (B) Percentage of bone marrow CD56^dim CD94^med CD16⁺ CD57^- mature NK cells in NBM (n = 30), MDS (n = 24), and AML (n = 8) cases. Bars represent the mean ± SEM (*P < 0.05, ns, not significant; one-tailed unpaired t-test). (C) Percentage of bone marrow CD56^dim CD94^low CD16⁺ CD57⁺ hypermature NK cells in NBM (n = 30), MDS (n = 25), and AML (n = 8) cases. Bars represent the mean ± SEM (ns, not significant; one-way ANOVA followed by Tukey’s Multiple Comparison test).

Figure 4

Ratio of cell percentages in distinct NK maturation subsets within the bone marrow microenvironment of NBM, MDS, and AML cases. (A) Ratio of cell percentages for the indicated NK maturation subsets (mature to immature, hypermature to immature, and hypermature to mature) in normal bone marrow (NBM, n = 30) cases. Data are presented as scatter dot plots, and the lines represent the mean ± SEM (****P < 0.0001, ns, not significant; Wilcoxon signed rank test). (B) Ratio of cell percentages for the indicated NK maturation subsets (mature to immature, hypermature to immature, and hypermature to mature) in myelodysplastic syndromes (MDS, n = 25) cases. Data are presented as scatter dot plots, and the lines represent the mean ± SEM (****P < 0.0001, ns, not significant; Wilcoxon signed rank test). (C) Ratio of cell percentages for the indicated NK maturation subsets (mature to immature, hypermature to immature, and hypermature to mature) in acute myeloid leukemia (AML, n = 8) cases. Data are presented as scatter dot plots, and the lines represent the mean ± SEM (*P < 0.05, ns, not significant; Wilcoxon signed rank test).

Figure 5

Comparison of the mean percentages of NK subsets expressing CD159a and KIR receptors within the bone marrow microenvironment of NBM, MDS, and AML cases. (A–D) Left panel: CD56^bright CD94^hi CD16^- CD57^- immature NK subset; middle panel: CD56^dim CD94^med CD16⁺ CD57^- mature NK subset; right panel: CD56^dim CD94^low CD16⁺ CD57⁺ hypermature NK subset; (A) Percentage of bone marrow CD159a-positive NK subsets in normal bone marrow (NBM, n = 30), myelodysplastic syndromes (MDS, n = 25), and acute myeloid leukemias (AML, n = 8) cases. Bars represent the mean ± SEM (*P < 0.05, ns, not significant; one-way ANOVA test followed by Tukey’s Multiple Comparison test; only for comparing immature NK subsets of NBM vs. MDS: Mann Whitney test). (B) Percentage of bone marrow CD158a-positive NK subsets in NBM (n = 30), MDS (n = 24), and AML (n = 8) cases. Bars represent the mean ± SEM (*P < 0.05, ns, not significant; one-tailed Mann Whitney and Kruskal-Wallis followed by Dunn’s Multiple Comparison tests for immature and mature NK subsets, one-way ANOVA followed by Tukey’s Multiple Comparison test for hypermature NK subsets). (C) Percentage of bone marrow CD158b-positive NK subsets in NBM (n = 30), MDS (n = 25), and AML (n = 8) cases. Bars represent the mean ± SEM (**P < 0.01, *P < 0.05, ns, not significant; Mann Whitney and Kruskal-Wallis followed by Dunn’s Multiple Comparison tests for immature and mature NK subsets, one-way ANOVA followed by Tukey’s Multiple Comparison test for hypermature NK subsets). (D) Percentage of bone marrow CD158e1-positive NK subsets in NBM (n = 30), MDS (n = 25), and AML (n = 8) cases. Bars represent the mean ± SEM (*P < 0.05, ns, not significant; Kruskal-Wallis followed by Dunn’s Multiple Comparison test).

Figure 6

Ratio of cell percentages in distinct CD159a-positive NK maturation subsets within the bone marrow microenvironment of NBM, MDS, and AML cases. (A) Ratio of cell percentages for the indicated CD159a-positive NK maturation subsets (mature to immature, hypermature to immature, and hypermature to mature) in normal bone marrow (NBM, n = 30) cases. Data are presented as scatter dot plots, and the lines represent the mean ± SEM (****P < 0.0001, *P < 0.05, ns, not significant; Wilcoxon signed rank test). (B) Ratio of cell percentages for the indicated NK maturation subsets (mature to immature, hypermature to immature, and hypermature to mature) in myelodysplastic syndromes (MDS, n = 25) cases. Data are presented as scatter dot plots, and the lines represent the mean ± SEM (****P < 0.0001, **P < 0.01, ns, not significant; Wilcoxon signed rank test). (C) Ratio of cell percentages for the indicated NK maturation subsets (mature to immature, hypermature to immature, and hypermature to mature) in acute myeloid leukemia (AML, n = 8) cases. Data are presented as scatter dot plots, and the lines represent the mean ± SEM (*P < 0.05, ns, not significant; Wilcoxon signed rank test).

Figure 7

Comparison of CD159a and KIR receptors mean fluorescence intensities (MFI) of different NK maturation subsets within the bone marrow microenvironment of NBM, MDS, and AML cases. (A–D) Left panel: CD56^bright CD94^hi CD16^- CD57^- immature NK subset; middle panel: CD56^dim CD94^med CD16⁺ CD57^- mature NK subset; right panel: CD56^dim CD94^low CD16⁺ CD57⁺ hypermature NK subset; (A) CD159a MFI of different bone marrow NK maturation subsets in normal bone marrow (NBM, n = 30), myelodysplastic syndromes (MDS, n = 25), and acute myeloid leukemias (AML, n = 8) cases. Bars represent the mean ± SEM (*P < 0.05, ns, not significant; one-tailed unpaired t-test, and Mann Whitney test only for comparing the MFI means of mature NK subsets of NBM vs. MDS: Mann Whitney test). (B) CD158a MFI, (C) CD158b MFI, (D) CD158e1 MFI of different bone marrow NK maturation subsets in NBM (n = 30), MDS (n = 24), and AML (n = 8) cases. Bars represent the mean ± SEM (*P < 0.05, ns, not significant; one-way ANOVA followed by Tukey’s Multiple Comparison test).

Figure 8

Differences in the repartition of CD159a and KIR receptors in distinct NK maturation subsets within the bone marrow microenvironment of NBM, MDS, and AML cases. (A) Mean fluorescence intensity (MFI) of CD159a and KIR receptors on the surface of bone marrow CD56^bright CD94^hi CD16^- CD57^- immature NK cells in normal bone marrow (NBM, n = 30), myelodysplastic syndromes (MDS, n = 25), and acute myeloid leukemia (AML, n = 8) cases. Bars represent the mean ± SEM (*P < 0.05; two-way ANOVA followed by Bonferroni post-test). (B) Mean fluorescence intensity (MFI) of CD159a and KIR receptors on the surface of marrow CD56^dim CD94^med CD16⁺ CD57^- mature NK cells in NBM (n = 30), MDS (n = 25), and AML (n = 8) cases. Bars represent the mean ± SEM (ns, not significant; two-way ANOVA followed by Bonferroni post-test). (C) Mean fluorescence intensity (MFI) of CD159a and KIR receptors on the surface of bone marrow CD56^dim CD94^low CD16⁺ CD57⁺ hypermature NK cells in NBM (n = 30), MDS (n = 25), and AML (n = 8) cases. Bars represent the mean ± SEM (*P < 0.05; two-way ANOVA followed by Bonferroni post-test).