A potent mechanism to secure NK cell tolerance is their expression of MHC class I-specific inhibitory receptors. On ligation of these receptors, NK functions are downregulated. Consequently, when MHC class I molecules are lost, target cells are killed. This recognition strategy has been termed "missing self" because the lack of the body's self markers leads to NK cell susceptibility (Bix et al. 1991; Hoglund et al. 1988; Karre et al. 1986; Moretta et al. 2002). In humans, inhibitory signals in response to MHC class I encounter are delivered primarily by killer immunoglobulin like (KIR) receptors on NK cells. Given that NK cells can lose tolerance against self cells in case of low MHC class I expression, it has been postulated that genetic polymorphisms in the KIR haplotypes could contribute to risk for autoimmunity.
Humans contain several genes encoding activating and inhibitory forms of KIR. Activating KIRs that contain two extracellular immunoglobulin domains and a short cytoplasmatic tail are designated KIR2DS. The corresponding inhibitory KIRs are called KIR2DL. Ligands for inhibitory KIR2DL receptors are HLA-C molecules that fall into two functional groups: KIR2DL1 binds to HLA-C with a Lys80 residue (HLA-Cw4 group), whereas KIR2DL2 and KIR2DL3 recognize HLA-C with an Asn80 residue (HLA-Cw3 and related alleles). Inhibitory KIRs can also contain three extracellular domains and are then called KIR3DL. The relative activity of activating and inhibitory KIRs on each NK cell is one parameter, albeit not the only one, that controls the activation of human NK cells (Moretta et al. 2002).
KIR genes are very polymorphic and are also clonally expressed in a stochastic manner on human NK cells. Furthermore, although there are at least 14 described KIR genes, not all genes are present in all individuals. It is thus likely that different NK cells, in the same individual or between individuals, are differentially capable of forming inhibitory KIR-HLA interactions, which may lead to differences in the amount of activating signal needed to trigger effector functions, for example, in autoimmunity. Consistent with this view, Martin et al. showed that the risk for developing psoriatic arthritis was highest among subjects that carried activating KIRs and at the same time lacked HLA-Cw ligands for some of their inhibitory KIRs (Martin et al. 2002; Nelson et al. 2004). Similarly, KIR2DS2 in absence of KIR2DL2 was more common in patients with scleroderma compared to healthy controls (Momot et al. 2004). KIR2DS2 has also been shown to associate with vasculitis in patients with RA (Yen et al. 2001) and with susceptibility to psoriasis vulgaris (Suzuki et al. 2004). In addition, psoriasis vulgaris was found to be associated with KIR2DS1 (Luszczek et al. 2004). These studies point toward activating KIRs conferring a more activated phenotype of immune responses, presumably by decreasing the activation threshold for KIR-expressing lymphocytes. These activating haplotypes thus confer an increased risk of developing autoimmunity. Also in IDDM, increased numbers of activating KIR genes compared to healthy controls have been demonstrated and the association with disease was clearest when combined with presence of HLA-C ligand and in DQ2 and/or DQ8 individuals (van der Slik et al. 2003).
Genetic associations between alleles of NK receptors and risk of developing autoimmune diseases bear a scientific weight, but the functional consequence of these associations is not easily studied. One drawback is that genetic studies do not take into account which cells potentially express the receptors under study. In the case of KIR receptors, they are expressed both by NK and T cells. It is also unknown at which level the KIR/MHC matching may operate. One possibility is in the target organ. Target cells could potentially downmodulate MHC class I molecules because of viral targeting, which could break NK cell tolerance in some NK cells and trigger NK cell-mediated target cell destruction. Alternatively, MHC class I downregulation on other cells in the body could trigger a helper function of NK cells, as has been described in cancer models (Kelly et al. 2002). Future studies will have to be set up to study this question specifically.
Was this article helpful?