The role of Natural Killer (NK) cells in host protection against viral infection and malignant transformation has been well described. NK cells may also lead to a reduction in post-transplant relapse and improved survival in hematopoietic stem cell transplantation (HSCT) for acute myelogenous leukemia (AML). It has been hypothesized that the genotype for the inhibiting killer immunoglobulin-like receptor (KIR) of the hematopoietic stem cell donor in combination with the HLA class I genotype of the recipient could control NK alloreactivity leading to a reduction in post-transplant complications. The KIR gene family encodes however both activating and inhibiting receptors. Here we test the hypothesis that activating KIRs with ligand specificity for HLA class I may contribute to alloreactivity, and potentially could be a genetic factor of significance in allogeneic HSCT. We tested this hypothesis in studies of two pairs of inhibiting and activating KIRs with highly homologous codon sequences in the extracellular domain, namely KIR2DL2/3-KIR2DS2 and KIR2DL1-KIR2DS1. Both the inhibitory 2DL1 and activating 2DS1 have ligand specificity for HLA-Cw group 2, and 2DL2 and 2DL3, have ligand specificity for HLA-Cw group 1, while the activating 2DS2 does not bind in vitro to C1 group. Using an EBV-transformed B-lymphoblastoid cell line (EBV-BLCL) target cell panel homozygous for HLA Class I alleles, we found that NK cells from donors with KIR haplotypes lacking KIR2DS1 or 2DS2 were not cytotoxic to allogeneic EBV-BLCL, independent of the target HLA class I genotype. Polyclonal NK cells obtained from KIR2DS1 positive and C1 group positive donors mediated NK cytotoxicity against C2 positive targets. In contrast, NK cells from KIR2DS1 positive, C2 group homozygous donors displayed minimal cytotoxicity against the C2 group targets (p<0.01). NK clones generated from 2DS1 positive, C2-group negative individuals were cytotoxic to C2-group target cells, while such NK clones could not be obtained from individuals positive for 2DS1 and cognate ligands. Similar findings were made for the relationship between 2DS2, 2DL2/3 and cognate ligand C1 group. Both polyclonal IL-2 propagated NK cells and NK clones from individuals positive for 2DS2 and homozygous for C2 group displayed specific cytotoxicity against C1 positive target cells. The cytotoxicity of 2DS2 positive, C1 group positive NK cells against the C1 positive BLCLs was minimal (p<0.01). These studies demonstrate that 2DS1 and 2DS2 are activating receptors that can induce an alloantigen response. We also present a model for combinations of KIR and HLA genotypes in which the allogeneic function of KIR2DS1 and 2DS2 is consistently seen in donor NK cells. Activating KIR may therefore play a role in allogeneic HSCT, and could contribute to the balance between activating and inhibiting signals for NK cells in HLA-Cw incompatible donor-recipient combinations. Activating KIR interactions with cognate ligand could potentially also play a role in the innate immune response. In the normal host, the increased affinity of the inhibiting KIR isoforms for HLA class I may prevent auto-reactivity, while the activating isoforms may only function in an HLA restricted pattern in context of specific pathogens or transformed cells. It is possible that the low affinity activating KIR may require additional co-stimulating signals that are up-regulated during cellular stress.
Disclosure: No relevant conflicts of interest to declare.