The functional binding site for the C-type lectin-like natural killer cell receptor Ly49A spans three domains of its major histocompatibility complex class I ligand

Abstract

Natural killer (NK) cells express receptors that recognize major histocompatibility complex (MHC) class I molecules and regulate cytotoxicity of target cells. In this study, we demonstrate that Ly49A, a prototypical C-type lectin-like receptor expressed on mouse NK cells, requires species-specific determinants on beta2-microglobulin (beta2m) to recognize its mouse MHC class I ligand, H-2D(d). The involvement of beta2m in the interaction between Ly49A and H-2D(d) is also demonstrated by the functional effects of a beta2m-specific antibody. We also define three residues in alpha1/alpha2 and alpha3 domains of H-2D(d) that are critical for the recognition of H-2D(d) on target cells by Ly49A. In the crystal structure of the Ly49A/H-2D(d) complex, these residues are involved in hydrogen bonding to Ly49A in one of the two potential Ly49A binding sites on H-2D(d). These data unambiguously indicate that the functional effect of Ly49A as an MHC class I-specific NK cell receptor is mediated by binding to a concave region formed by three structural domains of H-2D(d), which partially overlaps the CD8 binding site.

Figure 1

Anti-β2m antibody inhibits Ly49A interaction with its MHC class I ligand H-2D^d. (A) Reversal of H-2D^d/Ly49A–mediated inhibition of target cell killing by anti-β2m antibody. Killing of H-2D^d–transfected mouse lymphoma C1498 cells by Ly49A⁺ NK cells in the presence of various antibodies was examined by standard ⁵¹Cr-release assay at an E/T ratio of 20:1. Anti–mouse β2m or anti–H-2D^d antibodies were used as Fab or F(ab′)₂ fragments to prevent ADCC. Means ± SD of triplicate studies are shown. (B) Inhibition of H-2D^d–transfected cells binding to Ly49A-transfected cells by anti-β2m antibody. Binding of H-2D^d–transfected C1498 cells to Ly49A-transfected CHO cells was examined in the presence of various antibodies. The antibodies were used at 10 μg/ml. Means ± SD of triplicate studies are shown.

Figure 2

H-2D^d complexed with human β2m cannot protect target cell killing by Ly49A⁺ NK cells. H-2D^d–transfected mouse lymphoma C1498 cells were cultured in the presence (A and C) or absence of human β2m (B and D). Then the cells were tested for binding of anti–mouse β2m antibody S19.8 (shaded area), anti–H-2D^d α3 antibody 34-2-12S (thin line), anti–human β2m antibody BBM1 (bold line), or control antibody (dotted line) by flow cytometry (A and B). The cells were tested for killing by Ly49A⁺ NK cells in the presence of various antibodies (C and D) at an E/T ratio of 20:1. Anti-β2m or anti–H-2D^d antibodies were used as Fab or F(ab′)₂ fragments, respectively.

Figure 3

sLy49A tetramer binds H-2D^d complexed with mouse β2m but not with human β2m. β2m-defective human Daudi cells were stably transfected with mouse β2m (A and C) or human β2m (B and D) together with H-2D^d heavy chain. The cells were assayed for binding of anti–H-2D^d antibody 34-2-12S (A and B, bold line), control antibody (A and B, thin line), PE-labeled sLy49A tetramer (C and D, bold line), or streptavidin-PE (C and D, thin line) by flow cytometry. Staining of these cells with 34-5-8S gave similar results to 34-2-12S (data not shown).

Figure 5

Binding of sLy49A tetramer to H-2D^d mutant transfectants. Each of the H-2D^d mutant transfectants was stained with sLy49A tetramer or streptavidin-PE alone. Binding of sLy49A tetramer to each H-2D^d mutant transfectant is expressed as relative value to wild-type H-2D^d transfectants as described in Materials and Methods. The mutants in site 1 and site 2 are represented by white and black bars, respectively. The other mutants and wild-type are represented by gray bars; R6A, D122A, and K243A mutations, which virtually disrupted sLy49A tetramer binding, are located in site 2. Definition for site 1 and site 2 followed that of Tormo et al. 21 and is shown in Fig. 7.

Figure 4

Expression of H-2D^d mutants on transfected C1498 cells. Individual Ala substitution of each residue was introduced into the indicated residues of H-2D^d. Each mutant is described as original amino acid residue in one letter code followed by residue number and by A, the one letter code for Ala. C1498 cells that were untransfected or transfected with wild-type H-2D^d or individual H-2D^d mutants other than E227A mutant were stained with 34-2-12S antibody (bold lines) or control antibody (broken lines). C1498 cells transfected with H-2D^d E227A mutant were stained with 34-5-8S (bold lines) or control antibody (broken lines), because E227A mutation disrupted the 34-2-12S epitope.

Figure 7

H-2D^d mutations and antibody epitopes in the Ly49A/H-2D^d complex. An MHC class I molecule and Ly49A molecules are depicted as molecular surface and ribbon diagrams, respectively (the coordinates were provided by Dr. D.H. Margulies, National Institutes of Health, Bethesda, MD). The bottom view is orthogonal to the top view. The molecular surface of H-2D^d heavy chain and β2m are yellow and pink, respectively. The residues of H-2D^d that affected Ly49A binding upon Ala substitution are red or orange and are labeled, whereas those that did not affect Ly49A binding upon Ala substitution are dark green (those in site 1) or light green (others). The residues involved in antigenic epitopes recognized by the anti–mouse β2m antibody S19.8 and the anti–H-2D^d α1/α2 antibody 34-5-8S are blue. The glycosylation site Asn86 is magenta. The graphics image was prepared with Swiss-PdbViewer (reference 59).

Figure 6

Inhibition of Ly49A⁺ LAK cell–mediated killing of tumor cells by H-2D^d mutants. Each H-2D^d mutant transfectant was assayed for killing by Ly49A⁺ LAK cells in the absence or presence of anti-Ly49A antibody or control antibody at an E/T ratio of 20:1. Inhibitory activity of each H-2D^d mutant is expressed as percent inhibition of target cell killing mediated by Ly49A in the formula: % inhibition of killing = (% specific cytotoxicity in the presence of the anti-Ly49A antibody A1 − % specific cytotoxicity in the absence of the antibody)/(% specific cytotoxicity in the presence of antibody 100×). Killing assay at an E/T ratio of 4:1 gave similar results (data not shown). Control antibody did not show any significant effect on killing by Ly49A⁺ LAK cells (data not shown). The mutants in site 1 and site 2 are represented by white and black bars, respectively. The other mutants and wild-type are represented by gray bars.