Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
, 65 (9), 1047-59

Human NK Cells Maintain Licensing Status and Are Subject to Killer Immunoglobulin-Like Receptor (KIR) and KIR-ligand Inhibition Following Ex Vivo Expansion

Affiliations

Human NK Cells Maintain Licensing Status and Are Subject to Killer Immunoglobulin-Like Receptor (KIR) and KIR-ligand Inhibition Following Ex Vivo Expansion

Wei Wang et al. Cancer Immunol Immunother.

Abstract

Infusion of allogeneic NK cells is a potential immunotherapy for both hematopoietic malignancies and solid tumors. Interactions between killer immunoglobulin-like receptors (KIR) on human NK cells and KIR-ligands on tumor cells influence the magnitude of NK function. To obtain sufficient numbers of activated NK cells for infusion, one potent method uses cells from the K562 human erythroleukemia line that have been transfected to express activating 41BB ligand (41BBL) and membrane-bound interleukin 15 (mbIL15). The functional importance of KIRs on ex vivo expanded NK cells has not been studied in detail. We found that after a 12-day co-culture with K562-mbIL15-41BBL cells, expanded NK cells maintained inhibition specificity and prior in vivo licensing status determined by KIR/KIR-ligand interactions. Addition of an anti-CD20 antibody (rituximab) induced NK-mediated antibody-dependent cellular cytotoxicity and augmented killing of CD20+ target cells. However, partial inhibition induced by KIR/KIR-ligand interactions persisted. Finally, we found that extended co-cultures of NK cells with stimulatory cells transduced to express various KIR-ligands modified both the inhibitory and activating KIR repertoires of the expanded NK cell product. These studies demonstrate that the licensing interactions known to occur during NK ontogeny also influence NK cell function following NK expansion ex vivo with HLA-null stimulatory cells.

Keywords: ADCC; Ex vivo expansion; HLA; Killer immunoglobulin-like receptors; NK cells.

Conflict of interest statement

Disclosures

The authors have no financial conflicts of interest.

Figures

Figure 1
Figure 1. Inhibitory KIR frequencies are affected by donor HLA genotype, but not by ex vivo expansion with K562-mbIL15-41BBL cells
PBMCs from healthy donors were expanded using K562-mbIL15-41BBL cells. A): Cells were pre-gated by lymphocyte, live cells, singlets and NK cells for KIR analyses. B): The percentages of 2DL2/3+ cells in pre- or post-expanded CD3-CD56+ cells from C1/C1 (n=4), C1/C2 (n=3) and C2/C2 (n=5) donors were compared to each other. Each line represents an individual healthy donor. Statistics between C1/C1 and C2/C2 donors are shown. There is no significant difference between C1/C1 and C1/C2 donors, or between C2/C2 and C1/C2 donors. C): The percentages of 3DL1+ cells in pre- or post-expanded CD3-CD56+ cells from Bw4+ donors (n=6) were compared to the percentages of 3DL1+ cells from Bw4- donors (n=4). D): The percentages of 2DL1+ cells in pre- or post-expanded CD3-CD56+ cells from C1/C1 (n=4), C1/C2 (n=3) and C2/C2 (n=5) donors were compared to each other. There is no significant difference between %2DL1+ cells from C1/C1, C1/C2 and C2/C2 donors, in either pre- or post-expanded NK cells. E): The percentages of 2DL1+ cells in pre- or post-expanded CD3-CD56+ cells from 2DL2−2DL3+ (n=6) donors were compared to that for 2DL2+2DL3+ (n=5) donors. There is no significant difference between %2DL1+ cells from 2DL2−2DL3+ and 2DL2+2DL3+ donors, in pre-expanded NK cells, but there is a significant difference in post-expanded NK cells. Two-way ANOVA was used to compare KIR frequencies in NK cells expanded from donors with different genotypes or expanded with various stimulatory cell lines. (*p<0.05, **p<0.01, ***p<0.001, ****p<0.0001)
Figure 2
Figure 2. Inhibitory KIRs on expanded NK cells retain inhibition specificity
A): PBMCs from donor #18 were expanded with K562-mbIL15-41BBL cells. Expanded NK cells were stimulated with either HLA-null 721.221 cells or C1, C2, or Bw4-expressing 721.221 variants. CD107a+ cells were gated for 2DL1single positive (sp), 3DL1sp and 2DL2/3sp NK subsets by flow cytometry, and the percentage of CD107a+ cells was shown within the gate. B): Similar relationships as in Fig. 2A were found for expanded NK cells from 12 healthy donors using K562-mbIL15-41BBL cells, when stimulated with the 721.221-C1, 721.221-C2 and 721.221-Bw4 cell line respectively, in the absence or presence of rituximab (RT: 0.5ug/mL). Single positive (sp), double and triple-positive, as well as iKIR-null NK cell subsets were gated based on KIR2DL1, KIR2DL2/3 and KIR3DL1 expression by flow cytometry analysis. Two-way ANOVA analyses was used to compare normalized CD107a level between NK cells that only express cognate iKIR and NK cells that express different combinations of iKIRs. Statistics are only shown for the RT treatment group (black bars). (ns: not significant, *p<0.05, **p<0.01, ***p<0.001, ****p<0.0001)
Figure 3
Figure 3. Inhibition from iKIR/KIR-ligand interactions can override the licensing effect, and rituximab treatment does not overcome differences due to licensing or iKIR inhibition
A): Normalized CD107a levels of licensed NK cells (2DL2/3sp and 2DL1+2DL2/3+ NK cells, blue bars) expanded from C1/C1 donors (n=4) were compared to normalized CD107a levels of unlicensed NK cells (grey bars) expanded from the same donors, following stimulation with HLA-null, HLA-C1 or C2 expressing 721.221 cells, either with or without rituximab (RT) treatment (0.5ug/mL). B): Normalized CD107a levels of licensed NK cells (blue bars) expanded from C2/C2 donors (n=5) were compared to normalized CD107a levels of unlicensed NK cells (grey bars) expanded from the same donors, following stimulation with HLA-null, HLA-C1 or C2 721.221 cells, either with or without RT treatment (0.5ug/mL). C): Normalized CD107a levels of licensed NK cells (3DL1sp, blue bars) expanded from Bw4+ donors (n=5) were compared to normalized CD107a levels of unlicensed NK cells (iKIR-null, grey bars) expanded from the same donors, following stimulations with HLA-null, HLA-C1, C2 or Bw4 721.221 cells, either with or without RT treatment (0.5ug/mL). In A, B and C, one-way ANOVA analyses (matching by donors) was used to compare CD107a levels between licensed and unlicensed NK cell subsets expanded from the same donor. (ns: not significant, *p<0.05, **p<0.01, ***p<0.001, ****p<0.0001)
Figure 4
Figure 4. Inhibitory and activating KIR frequencies are affected by KIR-ligand expressed on stimulatory cell
PBMCs from healthy donors were expanded using 721.221 HLA-null or 721.221 HLA-C1/C2/Bw4 variants as stimulatory cells, at PBMC: stimulatory ratios of 1:1. A): KIR2DS1 and KIR2DL1 staining were distinguished by a mAb specific for 2DL1 and a mAb recognizing 2DL1 and 2DS1. B): The frequency of 2DL1+2DS1+, 2DL1+2DS1− and 2DL1−2DS1+ cells in NK cells expanded with HLA-null 721.221 cells was compared to the frequency of the same subset in NK cells expanded with HLA-C2 721.221 cells, regardless of expression of other iKIRs on NK cells (n=4). The frequency of 2DL3+ cells and the frequency of 3DL1+ cells in NK cells expanded with HLA-null 721.221 cells were compared to the frequency of the same subset in NK cells expanded with HLA-C1 721.221 or HLA-Bw4 721.221 cells respectively, regardless of expression of other iKIRs on NK cells (n=4). Paired t test or two-way ANOVA (matched by donor) was used to compare KIR frequencies in NK cells expanded with various stimulatory cell lines. (ns: not significant, *p<0.05, **p<0.01, ***p<0.001, ****p<0.0001)

Similar articles

See all similar articles

Cited by 3 PubMed Central articles

Publication types

LinkOut - more resources

Feedback