A KIR-TLR connection

Killer immunoglobulin-like receptors (KIRs) are well-known for binding human leukocyte antigen (HLA) and allowing natural killer (NK) cells to monitor HLA levels on host cells. Unexpectedly, Sivori and colleagues have found that certain KIRs can also bind microbial nucleic acid and deliver it to endosome resident Toll-like receptors (TLRs), identifying a completely novel role for KIR family receptors.¹ 

KIRs were originally discovered based on their ability to inhibit the effector functions of NK cells upon engaging specific HLA allotypes expressed on the cell surface of potential target cells. Consequently, inhibitory KIRs prevent NK cells from killing host cells that express HLA molecules at normal levels, allowing the specific detection of cells in which HLA expression is low. This so-called “missing-self” recognition represents an indirect recognition strategy used by NK cells to detect infected or transformed cells, which escape the direct antigen/HLA recognition by T cells. Besides the T-cell receptor, KIRs thus represent the prototype cell-surface receptor specific for HLA molecules. In a surprising twist, reported in this issue of Blood, Sivori et al find that certain KIRs can also directly bind unmethylated microbial cysteine guanine dinucleotide (CpG) DNA.¹ 

KIRs are composed of 2 or 3 extracellular immunoglobulin (Ig) domains whereby HLA binding is mediated by the junction between the 2 membrane proximal Ig domains (see figure). The precise role of the third, membrane distal Ig-domain, the so-called D0 domain has remained enigmatic, even though it does play a role in HLA binding. This D0 domain in 3 Ig-domain KIRs, such as KIR3DL2, is now shown to directly bind CpG DNA (see figure). In contrast to HLA binding, KIR-mediated DNA recognition does not inhibit but rather stimulates NK cells to efficiently produce cytokines, such as Interferon-γ. Exactly how DNA binding induces KIR internalization rather than KIR clustering and inhibitory signaling is currently unknown. Irrespective of the mechanism, KIRs are shown to deliver DNA to endosomes where the acidic pH induces detachment and allows transfer to endosome resident TLR9. Even though other means exist to translocate DNA to endosomes, these new data suggest that certain KIRs permit human NK cells to directly recognize and efficiently respond to infection.

Mouse NK cells use a set of structurally distinct, but functionally equivalent Ly49 receptors for the recognition of major histocompatibility complex (MHC) class I molecules. A recurrent issue in the NK-cell field is whether human and mouse receptors perform equivalent functions. Here, despite the structural differences, one Ly49 family member, Ly49Q, may indeed perform a function similar, if not identical, to the one reported for KIR3DL2. Ly49Q binds certain MHC class I allotypes²  and was recently reported to play an important role in TLR9 signaling. Rather than playing a role in NK cells, Ly49Q is important for efficient TLR9-mediated production of type I Interferon by plasmacytoid dendritic cells.³  Even though it is not known whether Ly49Q binds DNA, the data provide intriguing evidence that structurally distinct human and murine MHC class I receptors are involved in TLR signaling.

What is the significance of the KIR-TLR connection in health and/or disease? The new data raise the possibility that immune cells co-expressing specific KIR and TLR9, such as NK cells and a subset of T cells, play a prominent role in directly recognizing and responding to pathogens or to host cell death. As one possible pathologic role, KIR3DL2 is selectively expressed in Sézary syndrome,⁴  a cutaneous T-cell lymphoma, which is associated with skin-related infections. Sivori et al speculate that the tumorigenic process may be initiated by infection, which chronically triggers KIR3DL2 mediated T-cell proliferation.

Finally, multiple genetic studies have shown an association of inhibitory KIR with diseases as diverse as infection, cancer, and autoimmunity. In most cases, the strongest influence on disease outcome is due to KIR-HLA gene interactions.⁵  One study showed a significant correlation between carriers of a particular KIR3DL2 allele and the induction of an efficient NK-cell response to Plasmodium falciparum–infected erythroctyes.⁶  Because erythrocytes lack HLA molecules and based on the new findings reported by Sivori et al, it is tempting to speculate that the recognition of DNA causally linked the expression of KIR3DL2 to NK-cell activation, perhaps providing the first evidence for the benefits of KIR-TLR connection.