Immunophenotypic and proteomic characterization of cord blood (CB) CD56^bright and CD56^dim NK cells: potential role of CB CD56^Dim mediating GVL effect

4316

NK cells are characterized by absence of CD3 but expression of CD56^dim (90%, cytotoxic) and CD56^{brigh t} (10%, mediator) (Shankaran et al Nature 2001). NK subsets carry out their respective functions based on their repertoire of NK receptors (NKRs) (Moretta et al Annu Rev Immunol 2001). Despite an immaturity in (spell out initials when first used) CB T-cell immunity, there is a similar leukemic relapse rate following UCBT vs. UBMT (Cairo et al Blood 1997). There is, however, a paucity of data comparing PB vs. CB NK subsets and NKR expression. In this study, we positively selected PB vs. CB CD56⁺ cells using magnetic beads (Miltenyi) and sorted into CD3^-/CD56^bright and CD3^-/CD56^dim and NKR expression measured (CD16, CD158a {KIR2DL1} CD158a,h {KIR2DL1 and KIR2DS1}, CD158b {KIR2DL2}, CD161, NKG2A, NKG2C, NKG2D, Nkp44, NKp46).Additionally,250 μg protein from cell extracts of CB and PB CD56^Dim were used to perform quantitative differential protein profiling using tandem mass spectrometry (MS/MS)(Lim et al Lab Invest 2004) CD56⁺ selection yielded >89% purity (PB-96%, CB-89%). There was no statistical difference (mean + SEM) in NKR expression between the CB CD56^dim and PB CD56^dim. The PB vs. CB CD56^bright had significant increased expression of KIR2DL2 (20.31 + 2.27 vs. 9.43 + 0.82, p< 0.012) only. The CB CD56^{dim vs. bright} had increased CD16 expression (85.06 + 6.75 vs. 40.91 + 5.74, p< 0.004) only. Proteomic data showed that 33 and 37 proteins were over and under expressed by > 2 fold between CB and PB CD56^dim NK cells. The functional categories of differentially expressed proteins were binding (46%), catalytic (17%), signaling (15%), transcription (15%), enzyme (3%), structural (2%), and transport (2%). CB compared to PB underexpressed Inositol 1,4,5-triphosphate receptor type 3, mitogen-activated protein kinase 5, and natural cytotoxicity triggering receptor 3 precursor by 8.33, 6.67, and 2.5 fold decrease, respectively. The CB compared to PB overexpressed neurogenic locus notch homolog protein 2 precursor, pleckstrin homology domain-containing family A member 1, and transcriptional repressor NF-X1 by 16.67, 7.14, and 5.26 fold increase, respectively. While PB vs. CB CD56^dim and PB vs. CB CD56^bright had similar NKR expression, CB vs. PB CD56^dim NK cells had protein expression differences in 70 proteins, mostly in catalytic and binding proteins, which may influence their cytotoxic capacity. We hypothesize that CB CD56^dim, in part contribute to the GVL effect post HLA disparate UCBT, and that differentially expressed proteins in CB vs. PB CD56^dim may contribute to disparate cytotoxic activity.