Natural killer cell killing of acute myelogenous leukemia and acute lymphoblastic leukemia blasts by killer cell immunoglobulin-like receptor-negative natural killer cells after NKG2A and LIR-1 blockade

Abstract

Although the study of natural killer (NK) cell alloreactivity has been dominated by studies of killer cell immunoglobulin-like receptors (KIRs), we hypothesized that NKG2A and LIR-1, present on 53% +/- 13% and 36% +/- 18% of normal NK cells, respectively, play roles in the NK cell killing of primary leukemia targets. KIR(-) cells, which compose nearly half of the circulating NK cell population, exhibit tolerance to primary leukemia targets, suggesting signaling through other inhibitory receptors. Both acute myelogenous leukemia and acute lymphoblastic leukemia targets were rendered susceptible to lysis by fresh resting KIR(-) NK cells when inhibitory receptor-major histocompatibility class I interactions were blocked by pan-HLA antibodies, demonstrating that these cells are functionally competent. Blockade of a single inhibitory receptor resulted in slightly increased killing, whereas combined LIR-1 and NKG2A blockade consistently resulted in increased NK cell cytotoxicity. Dual blockade of NKG2A and LIR-1 led to significant killing of targets by resting KIR(-) NK cells, demonstrating that this population is not hyporesponsive. Together these results suggest that alloreactivity of a significant fraction of KIR(-) NK cells is mediated by NKG2A and LIR-1. Thus strategies to interrupt NKG2A and LIR-1 in combination with anti-KIR blockade hold promise for exploiting NK cell therapy in acute leukemias.

Figure 1. NK cells express KIR, NKG2A, and LIR-1 and leukemia targets minimally express HLA-G transcripts

A) NK cells were evaluated by flow cytometry for the expression of class I recognizing inhibitory receptors (n=42). Each CD56⁺/CD3⁻ NK cell population was gated to evaluate the expression of NKG2A and LIR-1 on KIR (using a cocktail recognizing KIR2DL1/S1, KIR2DL2/L3/S2 and KIR3DL1) expressing cells. Shown is a representative example from a donor with a i) KIR and ii) NKG2A dominant phenotype. B) NK cells were then enriched into a KIR⁺ and KIR⁻ population using anti-PE conjugated immunomagnetic beads recognizing a cocktail of anti-KIR mab. Representative examples of a KIR⁺ and KIR⁻ population are shown. KIR⁻ NK cells co-express a high proportion of NKG2A and less LIR-1. C) RT-PCR and Southern blot analysis was used to test for alternatively spliced HLA-G on nine leukemia samples, the erythroleukemia cell line K562, two normal peripheral blood mononuclear cell populations (PBMC) and a positive control choriocarcinoma cells line (JEG-3). β-actin was used as an internal control.

Figure 2. KIR, NKG2A and LIR-1 blockade all contribute to cytotoxicity against primary AML

A) Primary AML blasts from 2 patients, AML4 (i, ii) (tested with 5 allogeneic NK cell donors) and AML3 (iii, iv) (tested with 6 allogeneic donors), were investigated for their susceptibility to cytolysis mediated by resting (i, iii) NK cells at effector to target (E:T) ratio of 5:1. B) Degranulation of resting (ii, iv) NK cells was determined by CD107a expression following co-incubation with AML4 (ii) (n=3 NK cell donors) and AML3 blasts (iv) (n=6 NK cell donors). Mean ± SEM from different NK cell donors are shown. Pan-HLA blockade was compared to blockade of KIR, NKG2A and LIR-1 individually or in combinations. Statistically significant differences between groups are marked (* p< 0.05; ** p<. 0.01; *** p<0.005; **** p<0.0005). HLA-B and -C ligand status of leukemia blasts are shown. There was KIR ligand match in three donors and mismatch in 2 KIR ligands in three donors used for AML3.

Figure 3. KIR⁻ NK cells exhibit significant alloreactivity against primary AML blasts

Primary AML blasts (AML3) were investigated for their susceptibility to KIR enriched (KIR⁺) and KIR⁻ resting (n=4 NK cell donors) NK cells in A) cytotoxicity and B) degranulation assays. There was KIR ligand match in one KIR⁺ donor and mismatch in 1 KIR-ligand in one and in 2 KIR-ligands in two KIR⁺ donors.

Figure 4. KIR⁻ NK cells exhibit significant alloreactivity against primary ALL blasts

Primary ALL blasts (ALL9) were investigated for their susceptibility to KIR enriched (KIR⁺) and KIR⁻ resting (n=4 NK cell donors) NK cells in A) cytotoxicity and B) degranulation assays.

Figure 5. Enhanced killing by NK cells 100 days after allogeneic transplantation by NKG2A and LIR-1 blockade

PBMC from patients 100 days after transplantation were tested against NK cells susceptible AML (n=3) and ALL (n=2) primary blasts alone or with anti-KIR, LIR-1, NKG2A blockade or the combination (Combo). A) Representative and B) aggregate data are shown. P-values less than 0.05 are shown (*).