The KIR repertoires of neonatal NK cells are diverse but not biased toward recognition of cognate HLA class I. (A) The frequencies of all 32 KIR and NKG2A receptor combinations (clonotypes) that can be distinguished by the present flow cytometry approach are shown for CB (black bars, n = 90) and PB (white bars, n = 150) as mean with error bars representing SD. Clonotypes are ordered according to the number of expressed receptors starting with no receptor (KIR−NKG2A−), 1 receptor (either a single KIR or NKG2A), 2 receptors (ie, KIR2DL1_3DL1 refers to a clone expressing KIR2DL1 and KIR3DL1 and no other KIR or NKG2A), and so forth. Designation of clonotypes does not include cross-reactivity with stimulatory KIR. (B) Inhibitory receptor repertoires of CB donors with group A/A KIR haplotypes. The order of clonotypes was similar to that in panel A but without consideration of KIR3DL2. Donors were divided into subgroups possessing C1/C1 (white bars, n = 19) and C2/C2 (black bars, n = 11) ligands. (C) Inhibitory receptor repertoires of CB donors with group A/B and B/B KIR haplotypes. Donors were divided into subgroups possessing C1/C1 (white bars, n = 31) and C2/C2 (black bars, n = 22) ligands. (D) Frequency of CB- and PB-derived NK cells expressing KIR2DL1, KIR2DL3, or KIR3DL1 in donors with group A/A KIR haplotypes. For each given KIR, donors were stratified according to the indicated HLA class I ligands. Results are shown as stapled bars with overall frequency of NK cells expressing a given KIR (light gray) and the corresponding fraction of single-KIR+ NK cells (dark gray). Sample distribution was: C1/C1 (n = 19/38 for CB/PB), C1/C2 (n = 20/30), and C2/C2 (n = 11/10) as well as Bw4/Bw4 (n = 5 for CB), Bw4/Bw6 (n = 16), and Bw6/Bw6 (n = 9). Results are shown as mean with error bars representing SD. Data for PB were calculated from the cohort analyzed in Schönberg et al.9 Statistical significance was determined by 2-tailed t test: *P < .05, **P < .01, and ***P < .001.
