SIGNAL TRANSDUCTION BY INHIBITORY RECEPTORS
ON HUMAN NATURAL KILLER CELLS



Dr. 
Kerry S. Campbell KERRY S. CAMPBELL, Ph.D.,
Associate Member (from June 1998)

GEORGE HII, B.S., Scientific Technician (from September 1998)


Natural Killer (NK) cells constitute about 10 to15% of peripheral blood leukocytes in humans. Research over the past decade has heightened our awareness of the important roles that NK cells can play in innate immune responses. As cellular effectors of innate immunity, they do not require prior antigen exposure to generate rapid cytotoxicity and cytokine responses directed toward certain endogenous cells that have been altered by virus infection or transformation to cancer cells. The key target cell surface markers that NK cells utilize to distinguish altered cells from normal cells are the major histocompatibility (MHC) class I (MHC-I) molecules normally expressed on the surface of virtually all cells of the body. The lack of MHC-I surface expression on a specific target cell is interpreted as "missing self" by NK cells, which triggers focused lysis of that target cell and initiates the secretion of cytokines, including interferon-g, tumor necrosis factor, and granulocyte/macrophage colony stimulating factor. The cytokine release by NK cells can greatly influence immune responsiveness and differentiation of other lymphocytes.



Extracted pic [2]

FIGURE 1. Structures of the two classes of MHC Class I-binding inhibitory receptors found on human NK cells. Basic structures are shown including orientation of the amino (NH3) and carboxyl (COOH) termini, extracellular C-type lectin or immunoglobulin (Ig)-like structural domains, disulfide linkages (SS), intracellular ITIM sequences (YxxL), and MHC-I ligands (HLA-A, -B, -C, and -E).

MHC-I recognition by NK cells is due to the surface expression of inhibitory receptors that bind MHC-I. Human NK cells express two families of MHC-I-binding inhibitory receptors, the killer cell inhibitory receptors (KIR) and the CD94/NKG2A heterodimers (Figure 1). KIR are type I transmembrane proteins that bind to classical MHC-I (human leukocyte antigen [HLA]-A, -B and -C) molecules, while the CD94/NKG2A receptors consist of type II transmembrane proteins that bind to the non-classical HLA-E (Figure 1). KIR are products of a multigene family consisting of slightly polymorphic receptors with specific MHC-I binding profiles. Engagement of KIR and CD94/NKG2A inhibitory receptors with MHC-I dominantly arrests activation signals derived from numerous receptors interacting with cell surfaces, such as CD2, CD16, NKR-P1, integrins, and several recently identified receptors. When an NK cell conjugates with a rare cell lacking MHC-I, however, the inhibitory receptors are not engaged with ligand and the unsuppressed positive signals trigger targeted lysis of that conjugated cell (Figure 2).

A subgroup of receptors in the KIR family, sometimes referred to as KAR for killer cell activating receptors, is comprised of truncated proteins that ironically transduce stimulatory signals. The truncated receptor chains lack functional elements in their cytoplasmic domains and uniquely exhibit lysine residues in their transmembrane domains. These lysines enable interactions with aspartic acid residues within the transmembrane domains of a signalling chain, designated DAP12. Thus, association with DAP12 provides activation signalling capacity to the KAR. The role of these activating forms of this receptor family is currently unclear, but they may be involved in triggering the phosphorylation of the immunoreceptor tyrosine-based inhibitory motif (ITIM) domains of the inhibitory versions, which is required for recruitment of the src homology (SH) domain-containing protein tyrosine phosphatase (PTP), SHP-1. Improved understanding of the specific mechanisms by which inhibitory and activating forms of KIR tune the NK cell activation threshold is critical for understanding the biological responses of NK cells during viral infection and cancer.


Extracted pic [1]

FIGURE 2. Schematic of KIR-mediated inhibition of target cell lysis. The two NK cell conjugation conditions depicted reflect encounter of an MHC-I expressing target cell, in which ligated KIR recruits intracellular SHP-1 phosphatase to inhibit activation receptor signals, and encounter of a MHC-I-deficient cell, in which KIR are not engaged and inhibitory signals are absent.

DEVELOPMENT OF A MODEL SYSTEM TO EXAMINE SIGNALLING CROSSTALK BETWEEN KIRS AND ACTIVATION RECEPTORS DURING TARGET CELL CONJUGATION. HII, CAMPBELL, in collaboration with LUTZa

A key element of signal transduction studies is the use of a model system that mimics physiological signalling conditions. One of the most relevant biological functions of NK cells is the process of natural cytotoxicity, whereby MHC-I-negative target cells are recognized and killed upon conjugation. We have obtained subclones of the NK-92 cell line that will be conjugated with MHC-I-deficient and MHC-I-bearing target cells to study KIR signal transduction. The NK-92 subclones are ideal for use as model cells for the following reasons. 1) They exhibit potent cytotoxicity toward target cell lines, which can be efficiently inhibited by expressing specific MHC-I molecules on the same target cells via transfection. 2) They express both an inhibitory and an activating KIR reactive with distinct available monoclonal antibodies. 3) They express an inhibitory KIR (KIR3DL1) that efficiently blocks cytotoxicity when engaged with HLA-B MHC-I molecules expressed on transfected target cells. Recognition of HLA-B is a distinct advantage, since expression of HLA-B (as opposed to HLA-A or -C) by transfection in the target cells does not induce the coexpression of HLA-E. This eliminates the influence of parallel negative signalling through the HLA-E-binding CD94/NKG2A inhibitory receptors on NK cells in target cell conjugation experiments. 4) They are much easier to grow than primary NK cell clones. 5) They lack expression of the FcgRIII (CD16) activation receptor; antibody treatment of CD16+ NK cells normally requires the complicated use of F(ab')2 antibody fragments to avoid activation by co-ligation of CD16. 6) They provide the opportunity to generate stable transfectants with retroviral vectors.

The NK-92 model system is being optimized to establish the most efficient KIR-mediated inhibitory signal transduction. Once optimized, this system will provide a versatile tool that should prove useful for many years of productive research applications. In addition to developing a cellular model system to study, we are developing methods to manipulate the system by introducing cDNA constructs for protein expression. These expression methods will be used to express mutant signalling effector proteins as tools to disrupt normal signalling pathways. We are exploring the use of retroviral and vaccinia viral vectors to express proteins in the NK-92 subclones.

DEFINING THE FUNCTIONAL ROLES FOR SHP-1 RECRUITMENT DURING KIR-MEDIATED INHIBITION OF NATURAL CYTOTOXICITY. HII, CAMPBELL, in collaboration with COLONNAb

Evidence from us and others indicates that the predominant molecular effector mediating "negative signalling" by inhibitory KIR is the recruitment of the cytosolic protein tyrosine phosphatase, SHP-1 (Figure 2). SHP1 contains a catalytic domain and two SH2 domains, which are protein interaction modules that specifically bind to two phosphorylated ITIMs [(I/V)xYxx(L/V)] within the cytoplasmic domains of KIRs. The recruitment of SHP-1 by tyrosine phosphorylated ITIM is believed to inhibit signal transduction events downstream from the activation of protein tyrosine kinases (PTK) through dephosphorylating PTK and their substrates.

Only a handful of observations have been made regarding apparent SHP-1 substrates that are dephosphorylated when KIR are ligated, and some of the reported observations are contradictory. In addition, few laboratories have analyzed these events by physiologically engaging KIR in NK cells during actual target cell interactions. We are using the model system described above to study the functional impact of SHP-1 recruitment by inhibitory KIR during target cell conjugation. To determine the impact of SHP-1 recruitment to KIR, we are seeking to identify proteins that are phosphorylated in NK cells conjugated with MHCI-deficient target cells, but not phosphorylated in NK cells conjugated with HLA-B-expressing target cells. The identification of specific proteins that are not phosphorylated when KIR are engaged will help determine the exact stage(s) of NK cell activation signalling at which SHP-1 intervenes to abrogate the response.

PUBLICATIONS

The following publications represent work carried out prior to K.S.C. joining Fox Chase:

BAUCH, A., CAMPBELL, K. S., RETH, M. Interaction of the CD5 cytoplasmic domain with the Ca2+/calmodulin-dependent kinase II d. Eur. J. Immunol. 28:2167-2177, 1998.

CALVO, J., VILDÀ, J. M., PLACES, L., SIMARRO, M., PADILLA, O., ANDREU, D., CAMPBELL, K. S., AUSSEL, C., LOZANO, F. Human CD5 signalling and constitutive phosphorylation of C-terminal serine residues by casein kinase II. J. Immunol. 161:6022-6029, 1998.

CAMPBELL, K.S., CELLA, M., CARRETERO, M., LÓPEZ-BOTET, M., COLONNA, M. Signalling through human killer cell activating receptors triggers tyrosine phosphorylation of an associated protein complex. Eur. J. Immunol. 28:599-609, 1998.

CAMPBELL, K. S., COOPER, S., DESSING, M., YATES, S., BUDER, A. Interaction of p59fyn kinase with the dynein light chain, Tctex-1, and co-localization during cytokinesis. J. Immunol. 161:1728-1737, 1998.

a   M. Colonna: Basel Institute for Immunology, Basel, Switzerland CH-4005

b   C. Lutz: University of Iowa, Iowa City, IA 52242

Illustrations or unpublished data in these reports should not be used without permission of the author.


Fox Chase Cancer Center Scientific Report 1998