The NK cell inhibitory receptor Ly49A recognizes the mouse MHC class I molecule H-2Dd and participates in the recognition of missing self. Previous studies indicated that the determinant recognized by Ly49A exists in α1/α2 domain of H-2Dd. Here we have substituted polymorphic as well as conserved residues of H-2Dd α1/α2 domain (when compared with H-2Kd, which does not interact with Ly49A). We then tested the ability of the H-2Dd mutants to interact with Ly49A by soluble Ly49A tetramer binding and NK cell cytotoxicity inhibition assays. Individual introduction of mutations converting the H-2Dd residue into the corresponding H-2Kd residue (N30D, D77S, or A99F) in H-2Dd partially abrogated the interaction between Ly49A and H-2Dd. Introduction of the three mutations into H-2Dd completely abolished Ly49A recognition. Individual introduction of D29N or R35A mutation into the residues of H-2Dd that are conserved among murine MHC class I severely impaired the interaction. The crystal structure of H-2Dd reveals that D77 and A99 are located in the peptide binding groove and that N30, D29, and R35 are in the interface of the three structural domains of MHC class I: α1/α2, α3, and β2-microglobulin. These data suggest that Ly49A can monitor mutations in MHC class I inside and outside of the peptide binding groove and imply that inhibitory MHC class I-specific receptors are sensitive to mutations in MHC class I as well as global loss of MHC class I. Our results also provide insight into the molecular basis of Ly49A to distinguish MHC class I polymorphism.

Natural killer cells are a group of lymphocytes that spontaneously lyse certain tumor targets (1). The finding that cell lines defective for MHC class I expression were efficiently lysed by NK cells led to the missing self hypothesis: NK cells kill cells that lack normal expression of MHC class I molecules (2). Recent studies have established that NK cells express inhibitory receptors for MHC class I molecules (3, 4). The receptors can be classified into two groups by their structural characteristics. One group that includes human killer cell Ig-like receptors is characterized by two or three Ig-like domains in the extracellular region and configuration of type I transmembrane proteins (5, 6). NK cell MHC class I receptors of Ig-type have thus far been found only in primates, but not in rodents. gp49B1, which belongs to the Ig superfamily, is expressed on mouse NK cells (7, 8); however, the ligand for gp49B1 may not be MHC class I. Another group of NK cell MHC class I receptors is characterized by disulfide-linked dimer of type II transmembrane protein with an extracellular region homologous to carbohydrate recognition domain of C-type lectin. This group includes the Ly49 family in mice (9, 10, 11, 12) and rats (13) and CD94/NKG2 heterodimer found in humans (14) and rodents (15, 16).

The inhibitory receptors of both groups have one or two immune receptor tyrosine-based inhibitory motifs (ITIM)3 characterized by an IXYXXL sequence in the cytoplasmic region (17). Upon ligand recognition, the Tyr residue in the ITIM is phosphorylated and recruits the Src homology domain-containing protein tyrosine phosphatase 1 with Src homology 2 domains. The recruited Src homology domain-containing protein tyrosine phosphatase 1 is believed to dephosphorylate unknown critical substrates to shut down positive signaling. There are activating type of MHC class I receptors of C-type lectin and Ig-type with extracellular domains similar to those of the inhibitory receptors. Instead of cytoplasmic ITIM, the activating receptors have characteristic charged amino acid residues in the transmembrane region, and by that associate with DAP12, an adapter component with cytoplasmic immune receptor tyrosine-based activating motif (18). Engagement of such receptors initiates an activation signaling cascade analogous to T or B cell receptor signaling.

Mouse NK cells express the Ly49 family of receptors, which includes >10 members (9, 10, 11, 12, 19). Several lines of evidence demonstrate that Ly49A, a primary member of Ly49 family, is an inhibitory receptor that specifically recognizes a conformational epitope on H-2Dd. Introduction of H-2Dd, but not H-2Kd or Ld, functionally protects C1498 lymphoma cells from lysis by Ly49A+ NK cells (20). Also, Ly49A expression on NK cells is modulated in mice expressing H-2Dd (21). Physical interaction is supported by studies of Chinese hamster ovary cells transfected with Ly49A that bind H-2Dd-transfected C1498 cells (22). The recognition of α1/α2 domain of H-2Dd is supported by two lines of indirect evidence. First, recognition of H-2Dd by Ly49A is inhibited by the α1/α2 domain-specific Ab 34-5-8S, but not by 34-2-12S Ab, which recognizes α3 domain of H-2Dd (20). Second, Ly49A recognizes the natural mutant MHC class I molecule dm1, which has α1 and the N-terminal half of α2 domain of H-2Dd and the rest of the regions from H-2Ld, which is not a ligand for Ly49A (23). Ly49A recognizes only the peptide-bound form of H-2Dd molecules, but there is no apparent specificity for peptides as long as they have the anchoring residues required to bind H-2Dd (24, 25), in contrast with peptide-specific recognition of MHC class I molecules by TCRs. Ly49A has a functional C-type lectin domain that can bind the polyanionic carbohydrate, fucoidan (26). However, Ly49A can recognize the H-2Dd that lacks carbohydrate moiety (23). Recently, Tormo et al. reported the crystal structure of the Ly49A/H-2Dd complex (27), providing firm evidence for specific interaction of Ly49A with H-2Dd. The structure reveals the presence of two possible binding sites on H-2Dd for Ly49A. The authors suggested that one of the binding sites (site 1) that is formed by the N-terminal end of α1 α helix and the C-terminal end of α2 α helix of H-2Dd might be the functional binding site involved in the inhibitory recognition of H-2Dd by Ly49A expressed on NK cells.

Our recent studies have indicated that another site on H-2Dd, termed site 2, forms the functional interaction site with Ly49A (28). However, other studies, including investigations by our laboratory, also indicate that there may be functional consequences of H-2Dd residues that are not directly contacted by Ly49A. Specifically, we previously localized regions of H-2Dd that are important for recognition by Ly49A by using a series of recombinant MHC class I molecules between H-2Dd and Kd (23). The important areas include β-sheet regions that form the bottom of α1/α2 domain rather than α helices of α1/α2 domains, which form the functional binding site for Ly49A by the crystal structure data and our more recent mutagenesis data (28). Furthermore, other investigators have reported that mutations in residues that affect the peptide binding cleft may also affect Ly49A interaction (29, 30). Therefore, the contribution of H-2Dd residues that do not directly contact Ly49A requires further investigation.

To address this issue in more detail, we produced a panel of H-2Dd mutants with a single or multiple mutations to examine the role of each residue in the interaction with Ly49A. The H-2Dd mutants expressed on C1498 lymphoma cells were tested for their ability to interact with Ly49A by binding of soluble Ly49A (sLy49A) complex and by functional inhibition of killing. We found that simultaneous introduction of three specific mutations inside and outside of the peptide binding groove totally disrupted the epitope on H-2Dd that is recognized by Ly49A. We also demonstrate that mutations in conserved residues also impair the epitope of H-2Dd required for recognition by Ly49A. Taken together, our results suggest that Ly49A recognizes a conformational epitope that is sensitive to substitution in residues in α1/α2 domains, including both polymorphic and conserved residues. We discuss the molecular mechanism by which Ly49A detects mutations and polymorphism in MHC class I in light of the recently identified functional binding site of Ly49A on H-2Dd (28).

C57BL/6 mice were obtained from CLEA Japan (Tokyo, Japan).

C1498 cells were maintained in RPMI 1640 supplemented with 10% heat-inactivated FCS (Sigma, St. Louis, MO), 25 mM HEPES (Dojindo, Kumamoto, Japan), 100 μg/ml penicillin, 100 U/ml streptomycin, 2 mM glutamine, and 5 μM 2-ME. Dd5 was established by transfecting C1498 cells with wild-type Dd expression vector as described previously (23). Dd5 and C1498 cells transfected with Dd mutant cDNA were maintained in the same medium supplemented with 0.5 mg/ml G418 sulfate (Life Technologies, Rockville, MD). L cells, generous gifts from Dr. Ted H. Hansen (Washington University, St. Louis, MO), were maintained in DMEM supplemented with 10% heat-inactivated FCS, 25 mM HEPES, 100 μg/ml penicillin, 100 U/ml streptomycin, 2 mM glutamine, nonessential amino acids (Life Technologies), and 1 mM sodium pyruvate (Life Technologies).

Ly49A+ or Ly49A lymphokine-activated killer (LAK) cells were prepared from splenocytes from 8- to 16-wk-old C57BL/6J mice as described previously (20). Recombinant human IL-2, a generous gift form Ajinomoto Corporation (Tokyo, Japan), was used for culture. Both populations were >98% positive for NK1.1 by FACS analysis. Ly49A+ cell preparations were >97% positive for anti-Ly49A Ab (A1) and Ly49A preparations contained <2% Ly49A+ cells.

sLy49A was prepared as described elsewhere (48). Briefly, the extracellular domain of Ly49A with N-terminal biotinylation sequence tag (31) was expressed in Eschericia coli using an efficient T7 RNA polymerase-based system (32). The recombinant protein was refolded in vitro by dilution (33) and purified by cation exchange and gel filtration column chromatography. The sLy49A was biotinylated by biotin ligase BirA (Avidity, Denver, CO). sLy49A tetramer was formed by incubating the biotinylated sLy49A with R-PE-conjugated streptavidin (BD PharMingen, San Diego, CA) at the molar ratio of 4:1.

34-5-8S (anti-H-2Dd α1/α2), 34-2-12S (anti-H-2Dd α3) (34), A1 (anti-Ly49A) (35), PK136 (anti-NK1.1) (36), 53-6.7 (anti-mouse CD8α) (37), and H57-597 (anti-mouse TCR β) (38) were purified from culture supernatants by protein A or protein G affinity column chromatography.

Mutations were introduced using the Altered Sites II site-directed mutagenesis kit (Promega, Madison, WI) according to the manufacturer’s protocol. Multiple mutations were introduced by several rounds of mutations using two selectable markers, ampicillin and tetracycline resistance. Sequences of the primers used for site-directed mutagenesis will be provided on request. Sequences of all mutants were verified by sequencing both strands using ABI373A, ABI377 sequencers, or LS2000 DNA sequencer (Shimadzu, Kyoto, Japan). H-2Dd cDNAs with mutations were directionally subcloned into an expression vector pHβAPr-neo driven by human β-actin promoter as described previously (23).

C1498 cells were transfected with electroporation using ECM600 (BTX, San Diego, CA) under the following conditions: 0.4 cm gap, 300 V, 1,000 μF, and 720 ohm resistance. After electroporation, the cells were cultured for 2 days and then selected in the presence of 1 mg/ml G418 sulfate and cloned by limiting dilution as described previously (23). The clones were assayed by flow cytometry, and clones expressing an equivalent level of H-2Dd to wild-type Dd transfectant were selected for further analysis. L cells were transiently transfected with 1 μg of plasmid and 8 μl of Transfectamine (Life Technologies) in OPTI-MEM (Life Technologies) and were analyzed by flow cytometry after 48 h of culture.

Cells were stained with 10 μg/ml purified Abs for 30 min. Then cells were washed three times with HBSS containing 0.1% BSA and 0.1% sodium azide and were stained with 10 μg/ml FITC-goat anti-mouse IgG F(ab′)2 (ICN Pharmaceuticals, Costa Mesa, CA) for 15 min. The cells were washed two times with the same buffer and analyzed on FACSCalibur with CellQuest software (BD Biosciences, San Jose, CA). For staining with sLy49A tetramer, cells were incubated on ice for 30 min with the 20 μg/ml sLy49A tetramer, then washed three times with the buffer. The cells were fixed with 1% paraformaldehyde in PBS and then analyzed as described above. Ten thousand events of cells gated by forward and side scattering were acquired for analysis. Binding of sLy49A tetramer to each mutant H-2Dd was calibrated with the expression of H-2Dd detected by 34-2-12S Ab in the following formula: sLy49A tetramer-binding index = [(mean fluorescence intensity (MFI) of sLy49A tetramer-stained cells) − (MFI of streptavidin-PE-stained cells)]/[(MFI of 34-2-12S-stained cells) − (MFI of control Ab-stained cells)]. sLy49A tetramer binding of each H-2Dd mutant is expressed as the relative value of the binding index as compared with that of wild-type H-2Dd, which was adjusted to 100.

Cytotoxicity of Ly49A+ or Ly49A LAK cells against C1498 transfectants was determined with a standard 51Cr release assay at the E:T ratios of 4 and 20 as described (23). Radioactivity released into supernatant was measured by liquid scintillation counting with Microbeta (Wallac, Turku, Finland). The percentage of specific cytotoxicity was calculated as described (23). Where Ab was included in the assay, LAK cells were incubated with Abs for 15 min at room temperature before addition of target cells.

The previous study examining a series of recombinant MHC class I molecules between H-2Dd, which is a ligand for Ly49A, and H-2Kd, which is not a ligand for Ly49A, demonstrated the importance of the two regions, 1–52 and 90–107 of H-2Dd, for the recognition of H-2Dd by Ly49A (23). These regions contained multiple residues that are different between Dd and Kd (Fig. 1). To evaluate these findings further, we compared the sequences of α1/α2 domains of H-2Dd and Dk (which bind Ly49A) and those of H-2Kd and Kb (which do not bind Ly49A) for residues that might be informative in mutagenesis experiments (Fig. 1). In the region of H-2Dd from 1 to 52, the N30D mutation was generated because Asp30 is uniquely found in Kd and Kb, whereas Dd and Dk have Asn30. The region from 90 to 107 involves the antigenic epitope of the anti-H-2Dd Ab 34-5-8S (39), which can block the interaction between Ly49A and H-2Dd (20). To determine the epitope of 34-5-8S Ab as well as the critical residues for Ly49A binding, we chose L95F, A99F, E104G, and G107W in this region, even though these residues are not conserved in another Ly49A ligand, H-2Dk. We also included D77S, which potentially affects peptide binding.

FIGURE 1.

Amino acid sequence alignment of the α1/α2 domains of H-2Dd, Dk, Kd, and Kb and location of mutated residues in H-2Dd. A, Amino acid sequences of α1/α2 domain of H-2Dd, Dk, Kd, and Kb were aligned using GENETYX-MAC (Software Development, Tokyo, Japan). The sequence of each MHC class I was obtained from GenBank with the following accession numbers: H-2Dd (P01900), H-2Dk (AAA53201), H-2Kd (P01902), and H-2Kb (P01901). Conserved residues are boxed. Residues that were mutated are indicated by ▵ (polymorphic residues) or ▴ (conserved residues) on top of the sequence. B, Structure of H-2Dd (PDB accession: 1bii) α1/α2 domain is shown in ribbon model. Residues that were mutated are shown in stick with numbers. The graphic was prepared with Swiss-PdbViewer (47 ).

FIGURE 1.

Amino acid sequence alignment of the α1/α2 domains of H-2Dd, Dk, Kd, and Kb and location of mutated residues in H-2Dd. A, Amino acid sequences of α1/α2 domain of H-2Dd, Dk, Kd, and Kb were aligned using GENETYX-MAC (Software Development, Tokyo, Japan). The sequence of each MHC class I was obtained from GenBank with the following accession numbers: H-2Dd (P01900), H-2Dk (AAA53201), H-2Kd (P01902), and H-2Kb (P01901). Conserved residues are boxed. Residues that were mutated are indicated by ▵ (polymorphic residues) or ▴ (conserved residues) on top of the sequence. B, Structure of H-2Dd (PDB accession: 1bii) α1/α2 domain is shown in ribbon model. Residues that were mutated are shown in stick with numbers. The graphic was prepared with Swiss-PdbViewer (47 ).

Close modal

We transfected C1498 lymphoma cells with the H-2Dd mutant constructs. The transfected clones were individually assayed for expression of H-2Dd by flow cytometry with 34-2-12S Ab, which recognizes the α3 domain (40) where we did not introduce any mutation. Clones with levels of expression similar to the wild-type H-2Dd transfectant were chosen for further analysis (Fig. 2). Because we only obtained W97R mutant transfectants with very low expression, W97R was omitted from further analysis. The panel of the transfectants was then examined for reactivity with 34-5-8S Ab, which recognizes a conformational epitope of H-2Dd in α1/α2 domain that is sensitive to peptide binding (41). Most of the H-2Dd mutant transfectants were stained equally with the two mAbs, 34-2-12S and 34-5-8S, whereas the transfectants with E104G or G107W mutant showed diminished staining with 34-5-8S Ab (Fig. 2). Transient transfection of L cells with these mutant constructs showed similar results (data not shown). These results suggest that the epitope of the 34-5-8S Ab contains the β hairpin loop that includes Glu104 and Gly107 and are consistent with the previous finding that the epitope recognized by 34-5-8S Ab is sensitive to the replacement of the region from 90 to 107 of H-2Dd with the similar region of H-2Kd (39). Because Ly49A recognition of H-2Dd is affected by 34-5-8S Ab, these data suggest that Glu104 and Gly107 are part or are in the proximity of the Ly49A epitope.

FIGURE 2.

Expression of H-2Dd mutants on transfected C1498 cells. Untransfected C1498 cells or transfected with each indicated H-2Dd mutant cDNAs were stained with 34-2-12S Ab (anti-H-2Dd α3, thin line), 34-5-8S Ab (anti-H-2Dd α1/α2, bold line) or isotype-matched control Ab (dotted line) then with FITC-goat anti-mouse IgG F(ab′)2.

FIGURE 2.

Expression of H-2Dd mutants on transfected C1498 cells. Untransfected C1498 cells or transfected with each indicated H-2Dd mutant cDNAs were stained with 34-2-12S Ab (anti-H-2Dd α3, thin line), 34-5-8S Ab (anti-H-2Dd α1/α2, bold line) or isotype-matched control Ab (dotted line) then with FITC-goat anti-mouse IgG F(ab′)2.

Close modal

The panel of H-2Dd mutant transfectants was assayed for binding of sLy49A tetramer and for killing by Ly49A+ LAK cells. sLy49A tetramer binds H-2Dd with similar specificity as Ly49A expressed on NK cells (48). Surprisingly, neither E104G nor G107W mutation, which affected 34-5-8S binding (Fig. 2), showed any significant effect on the binding of sLy49A tetramer to H-2Dd (Fig. 3). The L95F mutation also showed no significant effect on sLy49A tetramer binding. In contrast, introduction of N30D, D77S, or A99F mutation into H-2Dd significantly reduced the binding of sLy49A tetramer to H-2Dd (Fig. 3). In killing assays, the protective activity of the N30D mutant against killing by Ly49A+ LAK cells was partially abrogated, whereas the other single mutants in the polymorphic residues showed similar protective activity to wild-type H-2Dd (Fig. 4).

FIGURE 3.

sLy49A tetramer staining of H-2Dd mutant transfectants. C1498 cells transfected with the H-2Dd mutants were stained with 20 μg/ml sLy49A tetramer. Binding of sLy49A tetramer to each H-2Dd mutant transfectant is expressed as binding relative to wild-type H-2Dd transfectants (normalized to 100), as described under Materials and Methods.

FIGURE 3.

sLy49A tetramer staining of H-2Dd mutant transfectants. C1498 cells transfected with the H-2Dd mutants were stained with 20 μg/ml sLy49A tetramer. Binding of sLy49A tetramer to each H-2Dd mutant transfectant is expressed as binding relative to wild-type H-2Dd transfectants (normalized to 100), as described under Materials and Methods.

Close modal
FIGURE 4.

Protecitive activity of H-2Dd mutants from killing by Ly49A+ LAK cells. C1498 cells transfected with the H-2Dd mutants were labeled with 51Cr and were subjected to a 4-h killing assay by Ly49A+ LAK cells (left panel) or Ly49A LAK cells (right panel) in the absence (○) or presence of anti-Ly49A (•) or control Ab (▵). Means ± SD of triplicates are shown. Representative results from three independent experiments with similar results are shown.

FIGURE 4.

Protecitive activity of H-2Dd mutants from killing by Ly49A+ LAK cells. C1498 cells transfected with the H-2Dd mutants were labeled with 51Cr and were subjected to a 4-h killing assay by Ly49A+ LAK cells (left panel) or Ly49A LAK cells (right panel) in the absence (○) or presence of anti-Ly49A (•) or control Ab (▵). Means ± SD of triplicates are shown. Representative results from three independent experiments with similar results are shown.

Close modal

To examine the combined effect of these mutations that affected sLy49A tetramer binding, we introduced either two or all of the N30D, D77S, and A99F mutations into the H-2Dd heavy chain and established stable transfectants (Fig. 2). Introduction of either one of D77S and A99F mutation together with N30D into H-2Dd decreased the sLy49A binding to H-2Dd compared with N30D single mutant H-2Dd. Introduction of all three mutations into H-2Dd at the same time completely abrogated the sLy49A binding (Fig. 3). Simultaneous introduction of both D77S and A99F into H-2Dd did not have significant effects that differed from either D77S or A99F single mutation. In the killing assay, a similar cooperative effect among N30D, D77S, and A99F mutations in H-2Dd on the abrogation of Ly49A-mediated protection of the tumor cell killing was observed (Fig. 4).

These results well explain the previous observation that exchange of the region 1–52 of H-2Dd with that of H-2Kd partially inhibits H-2Dd recognition by Ly49A and that exchange of the region 1–107 of H-2Dd with that of H-2Kd completely abrogated the H-2Dd recognition by Ly49A (23). The results also indicate that Ly49A binding to H-2Dd does not depend on the 34-5-8S epitope, even though 34-5-8S Ab blocks the Ly49A binding to H-2Dd.

In addition to these polymorphic residues, we introduced mutations into conserved charged residues in α1 domain in the neighborhood of Asn30 to evaluate their contribution to Ly49A binding. These mutants include D29N, R35A, E41A, E46A, and R48A (Fig. 1). The H-2Dd mutant molecules were expressed on C1498 cells (Fig. 2) and were tested for the binding of sLy49A tetramer and the ability to inhibit killing by Ly49A+ LAK cells. Introduction of D29N mutation completely abolished the ability of H-2Dd to interact with Ly49A in the sLy49A binding and the functional protection assay (Figs. 3 and 4). Introduction of R35A mutation severely impaired the ability of H-2Dd to interact with Ly49A in both of the assays. Mutations in Glu41, Glu46, or Arg48 did not show any significant effect on the Ly49A-H-2Dd interaction. Thus, our results show the ability of Ly49A to detect mutations in various regions of H-2Dd to monitor expression of normal MHC class I.

We demonstrated that the introduction of point mutations inside and outside of peptide binding groove impaired the ability of H-2Dd to protect tumor target cells from lysis by Ly49A+ LAK cells. In missing self hypothesis, NK cells monitor somatic cells for expression of MHC class I molecules (2). When NK cells find cells with loss of MHC class I expression or expression of aberrant MHC class I, NK cells are activated to kill those cells. Our results suggest that Ly49A can monitor a wide range of mutations inside and outside of the peptide binding groove of MHC class I molecules. Importantly, these data imply that inhibitory MHC class I specific receptors are sensitive to mutations in various regions of MHC class I, such as might occur in tumorigenesis, as well as global loss of MHC class I. These data also provide insight into structural basis of Ly49A specificity for polymorphic MHC class I.

N30D, D29N, and R35A mutations located outside of the peptide binding groove affected the recognition of H-2Dd by Ly49A (Figs. 3 and 4). These mutations can be interpreted in light of the crystal structure of the Ly49A/H-2Dd complex reported by Tormo et al. (27) and recent identification of the functional Ly49A-binding site on H-2Dd (28). The crystal structure provided the two possible binding sites on H-2Dd for Ly49A dimer. One site that consists of the NH2-teminal end of α1 α helix and the COOH-terminal of α2 α helix has been suggested to be a functional binding site for Ly49A because of the polymorphism and geometry (27). In contrast with this possibility, we recently established that Ly49A functionally binds the other site on H-2Dd to inhibit NK cell cytotoxicity (28). The functional binding surface is contributed by the three structural domains that constitute H-2Dd, α1/α2 and α3 domains and β2-microglobulin (β2m), and was referred to as site 2 by Tormo et al. (27). Importantly, Asn30, Asp29, and Arg25 are at the junction of these three domains and are also in the neighborhood of the functional binding site for Ly49A (Fig. 5). Especially, the side chain of Arg35 (H-2Dd heavy chain) is involved in hydrogen bonding with β2m in the crystal structure of H-2Dd (42, 43). Disruption of this hydrogen bond by R35A mutation might change the orientation of β2m against α1/α2 domain and thereby severely impair the Ly49A binding to H-2Dd. In the crystal structure, Asp29 of H-2Dd heavy chain makes a salt bridge with Arg228 of Ly49A in the functional binding site. Therefore, complete loss of the binding by introduction of D29N mutation into H-2Dd heavy chain is also consistent with this structure model.

FIGURE 5.

H-2Dd mutations in the Ly49A/H-2Dd complex. The structure of the Ly49A/H-2Dd complex (PDB ID: 1QO3) resolved by Tormo et al. (27 ) is shown in ribbon model. The two possible binding sites are shown. We have recently demonstrated that site 2 is the functional binding site related to inhibition of NK cell cytotoxicity (28 ). H-2Dd heavy chain, β2m, and peptide are colored in green, yellow, and orange, respectively. Two chains of a Ly49A dimer are colored in dark blue or sky blue. H-2Dd residues shown to affect Ly49A interaction in this study are colored red and are shown in space filling model with label. The H-2Dd residues involved in the 34-5-8S epitope are colored blue and are shown in space filling model. The graphics image was prepared with Swiss-PdbViewer (47 ).

FIGURE 5.

H-2Dd mutations in the Ly49A/H-2Dd complex. The structure of the Ly49A/H-2Dd complex (PDB ID: 1QO3) resolved by Tormo et al. (27 ) is shown in ribbon model. The two possible binding sites are shown. We have recently demonstrated that site 2 is the functional binding site related to inhibition of NK cell cytotoxicity (28 ). H-2Dd heavy chain, β2m, and peptide are colored in green, yellow, and orange, respectively. Two chains of a Ly49A dimer are colored in dark blue or sky blue. H-2Dd residues shown to affect Ly49A interaction in this study are colored red and are shown in space filling model with label. The H-2Dd residues involved in the 34-5-8S epitope are colored blue and are shown in space filling model. The graphics image was prepared with Swiss-PdbViewer (47 ).

Close modal

D77S and A99F mutations affected H-2Dd interaction with Ly49A even though the residues are located inside the peptide binding groove (Fig. 3). The effect of D77S and A99F mutations on Ly49A interaction with H-2Dd was not evident in the killing assay when introduced alone (Fig. 4). However, cointroduction of these mutations together with N30D mutation significantly impaired the inhibitory recognition by Ly49A as well as the binding of Ly49A tetramer. This observation might reflect a threshold of ligand receptor interaction required for inhibition of killing: the effect of D77S and/or A99F single or double mutations was small so that the effect was insufficient to be detected by functional inhibition assay unless the N30D mutation was also present.

The crystal structure of H-2Dd revealed that Ala99 and Asp77 are located in a critical position to bind two anchoring residues of the peptide, Pro at position 3 and Arg at position 5, respectively (42, 43). Consequently, H-2Dd with these mutations might not bind peptide or acquire new peptide specificities. Two lines of evidence, albeit indirect, suggest that the mutants H-2Dd with D77S and/or A99F mutation on the cell surface are expressed with bound peptides. First, H-2Dd molecules with one or both of D77S and A99F mutations were expressed on the cell surface at levels comparable to wild-type H-2Dd (Fig. 2). The notion that expression of classical MHC class I molecules requires complex formation with peptides (44) suggests that the H-2Dd mutants are peptide-bound. Secondly, the H-2Dd mutants were stained by 34-5-8S Ab, which recognizes an epitope sensitive to peptide binding (41), to the same extent as staining by mAb 34-2-12S (Fig. 2), which recognizes α3 domain of H-2Dd (40). In contrast with D77S and/or A99F mutants, the inefficient expression of the W97R mutant (Fig. 2) likely results from the impaired ability of the mutant to bind peptide. This is supported by our observation that expression of the mutant was rescued by incubation of the cells at 28°C (N. Matsumoto, unpublished observation) that induce peptide-independent expression of MHC class I (24, 25). Moreover, the W97R mutant was least reactive with 34-5-8S when expressed on L cells (data not shown).

Recently, Waldenström et al. reported that simultaneous introduction of S73W and D156Y mutation into H-2Dd partially impairs H-2Dd-mediated protection against killing by Ly49A+ NK cells (29). These substitutions putatively introduce a ridge inside the peptide binding groove that is found in H-2Db. Their interpretation is consistent with our current finding that substitution in peptide binding groove influenced the recognition of H-2Dd by Ly49A. However, we could not detect any decrease in protection by the S73W, D156Y double mutant in our system using C1498 cells (N. Matsumoto, unpublished data). The discrepancy might be due to the difference in level of expression of mutant H-2Dd or the difference in balance of activating and inhibitory ligands on the target cells used in these experimental systems. Correa et al. have shown that a stronger inhibitory signal is required to overcome a stronger activating signal by comparing natural killing and Ab-dependent cellular cytotoxicity (45). During the preparation of this manuscript, Nakamura et al. reported that introduction of triple mutations, W97Q, A99S, and W114L, into the peptide binding groove of H-2Dd completely abrogates recognition by Ly49A (30). This also underscores the importance of the peptide binding groove for Ly49A recognition of H-2Dd. In the crystal structure, the COOH-terminal end of α1 α helix and the NH2-terminal end of α2 α helix are engaged by Ly49A with hydrogen bonds at the functional binding site. This structure explains the peptide dependency of Ly49A binding to H-2Dd and the sensitivity of the Ly49A binding to the mutations in the peptide binding groove, observed in this and previous studies (29, 30).

Our finding that E104G or G107W mutations impaired staining with 34-5-8S Ab suggests that 34-5-8S recognizes the β hairpin loop, including Glu104 and Gly107, and indicates the essential role of these residues for efficient binding of the Ab. This is consistent with the previous finding that the 34-5-8S epitope is located in the region between 90 and 107 (39). The current finding that H-2Dd with E104G or G107W mutation was fully recognized by Ly49A (Figs. 3 and 4) clearly segregates the Ly49A epitope on H-2Dd from the 34-5-8S epitope, despite the previous result that 34-5-8S Ab inhibits H-2Dd recognition by Ly49A (20). However, the β hairpin loop, which contains the 34-5-8S epitope, is located at the edge of the functional binding site of Ly49A on H-2Dd, and the Ab is therefore in good proximity to sterically hinder Ly49A binding (Fig. 5). Therefore, the identification of the 34-5-8S epitope and the inhibition of the Ly49A-H-2Dd interaction by 34-5-8S strongly support the recent identification of the functional Ly49A binding site as site 2 (28).

Ly49A is able to distinguish polymorphic MHC class I molecules. Ly49A can recognize H-2Dd, Dk, and Dp, but not H-2Db, Kb, Kd, and Ld (20, 46). Introduction of N30D, D77S, and A99F, substitution of three Dd residues to Kd-type completely abolished Dd interaction with Ly49A (Figs. 3 and 4). This observation explains the previous finding that the R6 recombinant MHC class I molecule, which has residues 1–107 from Kd, is not able to protect target cell killing by Ly49A+ NK cells and that the R1-R5 recombinant MHC class I molecules, all of which have residues 1–52 from Kd and the residues 90–107 from H-2Dd, partially protect target cells from killing by Ly49A+ NK cells (23).

Our results well explain the inability of Ly49A to interact with H-2Kd and provide insight into how Ly49A distinguishes polymorphic MHC class I molecules. However, the residues found in H-2Kd are not always associated with other H-2 molecules that are not ligands for Ly49A. Additional residues that determine the reactivity of each polymorphic MHC class I molecule with Ly49A remain to be explored by further site-directed mutagenesis on H-2Dd to fully understand the basis of MHC class I specificity of Ly49A.

We thank D. H. Margulies for helpful discussion, T. Hansen for L cells, and S. Hanum and S. Pan for technical assistance on site-directed mutagenesis.

1

This work was supported by grants-in-aid for scientific research from the Ministry of Education, Science, and Culture of Japan (12672107), a grant for research on health sciences focusing on drug innovation from the Japan Health Science Foundation (12259), and grants from the National Institutes of Health. W.M.Y. is an investigator of the Howard Hughes Medical Institute.<./>

3

Abbreviations used in this paper: ITIM, immuno-receptor tyrosine-based inhibitory motifs; LAK, lymphokine-activated killer; sLy49A, soluble Ly49A; β2m, β2-microglobulin; MFI, mean fluorescence intensity.

1
Trinchieri, G..
1989
. Biology of natural killer cells.
Adv. Immunol.
47
:
187
2
Karre, K., H. G. Ljunggren, G. Piontek, R. Kiessling.
1986
. Selective rejection of H-2-deficient lymphoma variants suggests alternative immune defence strategy.
Nature
319
:
675
3
Lanier, L. L..
1998
. NK cell receptors.
Annu. Rev. Immunol.
16
:
359
4
Yokoyama, W. M..
1999
. Natural Killer Cells. W. E. Paul, ed.
Fundamental Immunology
4th Ed.
575
Lippincott-Raven, Philadelphia.
5
Long, E. O., D. N. Burshtyn, W. P. Clark, M. Peruzzi, S. Rajagopalan, S. Rojo, N. Wagtmann, C. C. Winter.
1997
. Killer cell inhibitory receptors: diversity, specificity, and function.
Immunol. Rev.
155
:
135
6
Colonna, M., J. Samaridis.
1995
. Cloning of immunoglobulin-superfamily members associated with HLA-C and HLA-B recognition by human natural killer cells.
Science
268
:
405
7
Wang, L. L., I. K. Mehta, P. A. LeBlanc, W. M. Yokoyama.
1997
. Mouse NK cells express gp49B1, a structural homologue of human killer inhibitory receptors.
J. Immunol.
158
:
13
8
Rojo, S., D. N. Burshtyn, E. O. Long, N. Wagtmann.
1997
. Type I transmembrane receptor with inhibitory function in mouse mast cells and NK cells.
J. Immunol.
158
:
9
9
Yokoyama, W. M., L. B. Jacobs, O. Kanagawa, E. M. Shevach, D. I. Cohen.
1989
. A murine T lymphocyte antigen belongs to a supergene family of type II integral membrane proteins.
J. Immunol.
143
:
1379
10
Chan, P. Y., F. Takei.
1989
. Molecular cloning and characterization of a novel murine T cell surface antigen, YE1/48.
J. Immunol.
142
:
1727
11
Wong, S., J. D. Freeman, C. Kelleher, D. Mager, F. Takei.
1991
. Ly-49 multigene family: new members of a superfamily of type II membrane proteins with lectin-like domains.
J. Immunol.
147
:
1417
12
Smith, H. R., F. M. Karlhofer, W. M. Yokoyama.
1994
. Ly-49 multigene family expressed by IL-2-activated NK cells.
J. Immunol.
153
:
1068
13
Rolstad, B., J. T. Vaage, C. Naper, D. Lambracht, K. Wonigeit, E. Joly, G. W. Butcher.
1997
. Positive and negative MHC class I recognition by rat NK cells.
Immunol. Rev.
155
:
91
14
Lazetic, S., C. Chang, J. P. Houchins, L. L. Lanier, J. H. Phillips.
1996
. Human natural killer cell receptors involved in MHC class I recognition are disulfide-linked heterodimers of CD94 and NKG2 subunits.
J. Immunol.
157
:
4741
15
Vance, R. E., D. M. Tanamachi, T. Hanke, D. H. Raulet.
1997
. Cloning of a mouse homolog of CD94 extends the family of C-type lectins on murine natural killer cells.
Eur. J. Immunol.
27
:
3236
16
Ho, E. L., J. W. Heusel, M. G. Brown, K. Matsumoto, A. A. Scalzo, W. M. Yokoyama.
1998
. Murine Nkg2d and Cd94 are clustered within the natural killer complex and are expressed independently in natural killer cells.
Proc. Natl. Acad. Sci. USA
95
:
6320
17
Burshtyn, D. N., A. M. Scharenberg, N. Wagtmann, S. Rajagopalan, K. Berrada, T. Yi, J. P. Kinet, E. O. Long.
1996
. Recruitment of tyrosine phosphatase HCP by the killer cell inhibitor receptor.
Immunity
4
:
77
18
Lanier, L. L., B. C. Corliss, J. Wu, C. Leong, J. H. Phillips.
1998
. Immunoreceptor DAP12 bearing a tyrosine-based activation motif is involved in activating NK cells.
Nature
391
:
703
19
McQueen, K. L., J. D. Freeman, F. Takei, D. L. Mager.
1998
. Localization of five new Ly49 genes, including three closely related to Ly49c.
Immunogenetics
48
:
174
20
Karlhofer, F. M., R. K. Ribaudo, W. M. Yokoyama.
1992
. MHC class I alloantigen specificity of Ly-49+ IL-2-activated natural killer cells.
Nature
358
:
66
21
Karlhofer, F. M., R. Hunziker, A. Reichlin, D. H. Margulies, W. M. Yokoyama.
1994
. Host MHC class I molecules modulate in vivo expression of a NK cell receptor.
J. Immunol.
153
:
2407
22
Daniels, B. F., F. M. Karlhofer, W. E. Seaman, W. M. Yokoyama.
1994
. A natural killer cell receptor specific for a major histocompatibility complex class I molecule.
J. Exp. Med.
180
:
687
23
Matsumoto, N., R. K. Ribaudo, J. P. Abastado, D. H. Margulies, W. M. Yokoyama.
1998
. The lectin-like NK cell receptor Ly-49A recognizes a carbohydrate-independent epitope on its MHC class I ligand.
Immunity
8
:
245
24
Correa, I., D. H. Raulet.
1995
. Binding of diverse peptides to MHC class I molecules inhibits target cell lysis by activated natural killer cells.
Immunity
2
:
61
25
Orihuela, M., D. H. Margulies, W. M. Yokoyama.
1996
. The natural killer cell receptor Ly-49A recognizes a peptide-induced conformational determinant on its major histocompatibility complex class I ligand.
Proc. Natl. Acad. Sci. USA
93
:
11792
26
Daniels, B. F., M. C. Nakamura, S. D. Rosen, W. M. Yokoyama, W. E. Seaman.
1994
. Ly-49A, a receptor for H-2Dd, has a functional carbohydrate recognition domain.
Immunity
1
:
785
27
Tormo, J., K. Natarajan, D. H. Margulies, R. A. Mariuzza.
1999
. Crystal structure of a lectin-like natural killer cell receptor bound to its MHC class I ligand.
Nature
402
:
623
28
Matsumoto, N., M. Mitsuki, K. Tajima, W. M. Yokoyama, K. Yamamoto.
2001
. The functional binding site for the C-type lectin-like NK cell receptor Ly49A spans the three domains of its MHC class I ligand.
J. Exp. Med.
193
:
147
29
Waldenstrom, M., J. Sundback, M. Y. Olsson-Alheim, A. Achour, K. Karre.
1998
. Impaired MHC class I (H-2Dd)-mediated protection against Ly-49A+ NK cells after amino acid substitutions in the antigen binding cleft.
Eur. J. Immunol.
28
:
2872
30
Nakamura, M. C., S. Hayashi, E. C. Niemi, J. C. Ryan, W. E. Seaman.
2000
. Activating Ly-49D and inhibitory Ly-49A natural killer cell receptors demonstrate distinct requirements for interaction with H-2Dd.
J. Exp. Med.
192
:
447
31
Schatz, P. J..
1993
. Use of peptide libraries to map the substrate specificity of a peptide-modifying enzyme: a 13 residue consensus peptide specifies biotinylation in Escherichia coli.
Biotechnology
11
:
1138
32
Studier, F. W., A. H. Rosenberg, J. J. Dunn, J. W. Dubendorff.
1990
. Use of T7 RNA polymerase to direct expression of cloned genes.
Methods Enzymol.
185
:
60
33
Garboczi, D. N., D. T. Hung, D. C. Wiley.
1992
. HLA-A2-peptide complexes: refolding and crystallization of molecules expressed in Escherichia coli and complexed with single antigenic peptides.
Proc. Natl. Acad. Sci. USA
89
:
3429
34
Ozato, K., N. M. Mayer, D. H. Sachs.
1982
. Monoclonal antibodies to mouse major histocompatibility complex antigens.
Transplantation
34
:
113
35
Nagasawa, R., J. Gross, O. Kanagawa, K. Townsend, L. L. Lanier, J. Chiller, J. P. Allison.
1987
. Identification of a novel T cell surface disulfide-bonded dimer distinct from the α/β antigen receptor.
J. Immunol.
138
:
815
36
Koo, G. C., J. R. Peppard.
1984
. Establishment of monoclonal anti-Nk-1.1 antibody.
Hybridoma
3
:
301
37
Ledbetter, J. A., L. A. Herzenberg.
1979
. Xenogeneic monoclonal antibodies to mouse lymphoid differentiation antigens.
Immunol. Rev.
47
:
63
38
Kubo, R. T., W. Born, J. W. Kappler, P. Marrack, M. Pigeon.
1989
. Characterization of a monoclonal antibody which detects all murine α β T cell receptors.
J. Immunol.
142
:
2736
39
Abastado, J. P., C. Jaulin, M. P. Schutze, P. Langlade-Demoyen, F. Plata, K. Ozato, P. Kourilsky.
1987
. Fine mapping of epitopes by intradomain Kd/Dd recombinants.
J. Exp. Med.
166
:
327
40
Evans, G. A., D. H. Margulies, B. Shykind, J. G. Seidman, K. Ozato.
1982
. Exon shuffling: mapping polymorphic determinants on hybrid mouse transplantation antigens.
Nature
300
:
755
41
Otten, G. R., E. Bikoff, R. K. Ribaudo, S. Kozlowski, D. H. Margulies, R. N. Germain.
1992
. Peptide and β2-microglobulin regulation of cell surface MHC class I conformation and expression.
J. Immunol.
148
:
3723
42
Achour, A., K. Persson, R. A. Harris, J. Sundback, C. L. Sentman, Y. Lindqvist, G. Schneider, K. Karre.
1998
. The crystal structure of H-2Dd MHC class I complexed with the HIV-1-derived peptide P18–I10 at 2.4 Å resolution: implications for T cell and NK cell recognition.
Immunity
9
:
199
43
Li, H., K. Natarajan, E. L. Malchiodi, D. H. Margulies, R. A. Mariuzza.
1998
. Three-dimensional structure of H-2Dd complexed with an immunodominant peptide from human immunodeficiency virus envelope glycoprotein 120.
J. Mol. Biol.
283
:
179
44
Townsend, A., C. Ohlen, J. Bastin, H. G. Ljunggren, L. Foster, K. Karre.
1989
. Association of class I major histocompatibility heavy and light chains induced by viral peptides.
Nature
340
:
443
45
Correa, I., L. Corral, D. H. Raulet.
1994
. Multiple natural killer cell-activating signals are inhibited by major histocompatibility complex class I expression in target cells.
Eur. J. Immunol.
24
:
1323
46
Olsson-Alheim, M. Y., J. Sundback, K. Karre, C. L. Sentman.
1999
. The MHC class I molecule H-2Dp inhibits murine NK cells via the inhibitory receptor Ly49A.
J. Immunol.
162
:
7010
47
Guex, N., M. C. Peitsch.
1997
. SWISS-MODEL and the Swiss-PdbViewer: an environment for comparative protein modeling.
Electrophoresis
18
:
2714
48
Matsumoto, N., K. Tajima, M. Mitsuki, and K. Yamamoto. 2001. H-2 allele specificity of the NK cell C-type lectin-like MHC class I receptor Ly49A visualized by soluble Ly49A tetramer. Int. Immunol. In press.