The effector function of human NK cells is down-regulated via ligation of inhibitory receptors (killer immunoglobulin-like receptors [KIR] and NKG2A) that recognize self MHC. Although most NK cells express “at least one” self MHC receptor, the mechanism producing self-tolerance is not yet understood. In order to precisely evaluate receptor expression to better understand this mechanism we developed and validated quantitative, real time RT-PCR (Q-RT-PCR) assays to measure mRNA levels from individual NKG2 and KIR genes. Comparison of subpopulations of NK cells sorted from blood showed a high expression of NKG2A on CD56+bright cells but virtually no KIR expression except for KIR2DL4. To further define KIR expression patterns on more mature cells we sorted the CD56+dim NK cells into KIR+ and KIR subpopulations using an antibody cocktail that recognizes 6 KIR genes. Comparing KIR gene expression by Q-RT-PCR in these subsets showed much higher transcript levels for the KIR genes included in the sorting cocktail in the KIR+ subpopulation than in the KIR cells (median expression ratio of 221.3; n=43; P=<0.0001). The median expression ratio of genes encoding KIR not recognized by the antibody cocktail was still higher in CD56+dim KIR+ cells than CD56+dim KIR cells (5.05; P=<0.0001). The finding that primary CD56+dim NK cells already expressing one KIR are more likely to express the others is consistent with the variegated KIR expression by NK cell clones and suggests that not all CD56+dim cells express KIR. Further sorting identified a novel subpopulation of CD56+dim NK cells that had neither NKG2A nor KIR at the cell surface. This NKG2AKIR cell subset is committed to the NK cell lineage and comprises 19.4 ± 2.8% of CD56+dim NK cells in healthy donors (n=26). However, the CD56+dim NKG2AKIR NK cells are not intolerant, autoreactive cells. Rather, they are immature, precursor NK cells that neither kill nor produce IFN-γ. When compared with the NKG2A+KIR, NKG2A+KIR+ and NKG2AKIR+ cells the NKG2AKIR subset was the only subpopulation with markedly decreased killing of K562 cells (P<0.05 compared to all others) and which lacked intracellular granzyme measured by flow cytometry. The NKG2AKIR cells also failed to produce IFN-γ even after incubation with IL-12 and IL-18 while the presence of NKG2A correlated with IFN-γ production. The CD56+dim NKG2AKIR NK cell lack of effector function is explained by their developmental immaturity. Importantly, after 14 days of culture in contact with a novel murine embryonic liver cell line (EL08-1D2) and IL-15 the NKG2AKIR cells acquired KIR (29±3.9%) and NKG2A (85±3.8%), and developed cytotoxic and cytokine-producing potential. We demonstrated that although CD56+dim NKG2AKIR NK cells do not have ‘at least one’ inhibitory receptor for engaging autologous MHC class I, they do not yet require a self-tolerance mechanism as they are developmentally immature. KIR expression can be induced by culture of this novel peripheral blood NK cell subset, and correlates with the acquisition of mature function. As CD94/NKG2A is expressed before KIR, the results suggest the following sequence for receptor acquisition in the development of CD56+dim NK cells: NKG2AKIR to NKG2A+KIR to NKG2A+KIR+ to NKG2AKIR+. We conclude that lineage committed NK cells in blood do not require inhibitory self-tolerance mechanisms until they reach a late stage of differentiation.

Disclosure: No relevant conflicts of interest to declare.

Author notes

*

Corresponding author

Sign in via your Institution