CD8 T cells express a single dominant KIR (1 KIR⁺ cells) often distinct from that of NK cells in the same individual. (A) Distribution of KIR2DL1-, KIR2DL3-, and KIR3DL1-expressing NK cells and CD45RA⁺CD57⁺ CD8 T cells within cells expressing 1-3 KIRs (n = 43; ***P < .001, 1-way ANOVA with Bonferroni multiple comparison test; mean ± SEM). The gating strategy used to identify NK cells included gating on CD56^dim+CD3⁻CD14⁻ lymphocytes after having excluded dead cells and double events. (B) The observed frequencies of CD45RA⁺CD57⁺ CD8 T cells coexpressing 2 KIRs (KIR2DL1 + KIR2DL3, KIR2DL1 + KIR3DL1, KIR2DL3 + KIR3DL1) is plotted against those given by the product of the expression frequencies for the respective KIRs. Results are shown as frequency out of total KIR-expressing cells. A perfect fit to the product rule appears as a gray line, illustrating a 1:1 relation between observed and expected frequencies. K values were derived from the slope of the linear regression of observed data relative to a perfect fit with the product rule. P values were derived from a linear regression analysis. (C) Frequency of the most expressed KIR, out of KIR2DL1, KIR2DL3, and KIR3DL1, within 1-KIR⁺ cells as the frequency out of total 1 KIR⁺ cell for NK cells and CD45RA⁺CD57⁺ CD8 T cells (n = 43; ***P < .001, paired Student t test; median). (D) Correlation between single KIR dominance within 1 KIR⁺ CD45RA⁺CD57⁺ CD8 T cell and pan-KIR expression of CD45RA⁺CD57⁺ CD8 T cells. (E) Expression of KIR2DL1, KIR2DL3, and KIR3DL1 on NK cells and CD45RA⁺CD57⁺ CD8 T cells (n = 43; n.s. indicates not significant; 1-way ANOVA with Bonferroni multiple comparison test; median). (F) Correlation of pan-KIR expression on CD45RA⁺CD57⁺ CD8 T cells and CD56^dim NK cells. (G) Schematic pie charts of KIR2DL1, KIR2DL3, and KIR3DL1 expression within 1 KIR⁺ NK cell and CD45RA⁺CD57⁺ CD8 T cells for 3 representative individuals.