HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Michaëlsson, J. Articles by Kärre, K. Search for Related Content PubMed PubMed Citation Articles by Michaëlsson, J. Articles by Kärre, K.

MHC Class I Recognition by NK Receptors in the Ly49 Family Is Strongly Influenced by the ₂-Microglobulin Subunit¹

Microbiology and Tumor Biology Center, Karolinska Institutet, Stockholm, Sweden

	Abstract

Top Abstract Introduction Materials and Methods Results Discussion References

 
NK cell recognition of targets is strongly affected by MHC class I specific receptors. The recently published structure of the inhibitory receptor Ly49A in complex with H-2D^d revealed two distinct sites of interaction in the crystal. One of these involves the ₁, ₂, ₃, and ₂-microglobulin (₂m) domains of the MHC class I complex. The data from the structure, together with discrepancies in earlier studies using MHC class I tetramers, prompted us to study the role of the ₂m subunit in MHC class I-Ly49 interactions. Here we provide, to our knowledge, the first direct evidence that residues in the ₂m subunit affect binding of MHC class I molecules to Ly49 receptors. A change from murine ₂m to human ₂m in three different MHC class I molecules, H-2D^b, H-2K^b, and H-2D^d, resulted in a loss of binding to the receptors Ly49A and Ly49C. Analysis of the amino acids involved in the binding of Ly49A to H-2D^d in the published crystal structure, and differing between the mouse and the human ₂m, suggests the cluster formed by residues Lys³, Thr⁴, Thr²⁸, and Gln²⁹, as a potentially important domain for the Ly49A-H-2D^d interaction. Another possibility is that the change of ₂m indirectly affects the conformation of distal parts of the MHC class I molecule, including the ₁ and ₂ domains of the heavy chain.

	Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

 
Natural killer cells are lymphocytes with the capacity to attack a variety of targets, such as certain tumors, virus-infected cells, and bone marrow-derived cells (1). They are regulated by activating and inhibitory receptors; some of the former and most of the latter receptors recognize MHC class I molecules on surrounding cells. The receptors can be divided into two major groups, comprising molecules with C-type lectin-like folds and a type II membrane topology (2, 3), and receptors of the Ig superfamily with a type I membrane topology (4). Both groups contain inhibitory as well as activating receptors. A delicate balance between triggering and inhibitory signals determines the action of the NK cell. Murine NK cells predominantly express receptors of the C-type lectin-like fold family, namely, Ly49 and CD94/NKG2. The Ly49 receptors consists of at least 13 members that are expressed at the mRNA level (Ly49A-P) (5) with more or less defined MHC class I specificity (6). Most of the Ly49 receptors have been characterized as inhibitory; they contain an immunoreceptor tyrosine-based inhibitory motif in their cytoplasmic tail and, upon ligation with the correct MHC class I molecule, they deliver inhibitory signals to the cell (3). There are also activating Ly49 receptors, which lack immunoreceptor tyrosine-based inhibitory motifs, and instead interact with adaptor molecules with immunoreceptor tyrosine-based activation motifs, e.g., DAP-12 (3). The MHC class I specificity of the Ly49 receptors is overlapping, and one Ly49 receptor can bind to several different MHC class I molecules. Conversely, one MHC class I molecule can bind to several different Ly49 receptors.

Class I MHC molecules are expressed on virtually all nucleated cells and present peptides derived from intracellular proteins to T cells (7). As noted above, MHC class I molecules also have a strong influence on the regulation of NK cell activity, mainly by providing inhibitory signals after initial triggering by activating receptors (8, 9, 10, 11). The MHC class I molecule is a trimolecular complex consisting of heavy chain, light chain (₂-microglobulin, ₂m),³ and peptide. ₂m is a soluble 12 kDa globular protein noncovalently associated with the MHC class I heavy chain. It is a relatively conserved protein, and displays 69% homology at the amino acid level between mouse and human (see Fig. 5) (12). This allows for cross-species association of ₂m with the heavy chain. This was first demonstrated by culturing human and murine cells in FBS, which led to an exchange of their endogenous ₂m for exogenous bovine ₂m (b₂m) (13, 14). The degree of ₂m exchange in MHC class I in mouse cell lines incubated with exogenous human ₂m (h₂m) can be up to 90% (15). Furthermore, changes in the ₂m subunit, e.g., by an exchange from murine ₂m (m₂m) to h₂m, have been demonstrated to affect the conformation of the ₁ and ₂ domains in the heavy chain in several earlier studies (16, 17, 18, 19, 20, 21, 22, 23).

View larger version (12K): [in this window] [in a new window]   FIGURE 5. Sequence comparison between m2m and h2m. Residues in m2m participating in hydrogen bonds with Ly49A are shaded in gray. Sequence alignment was performed by using "BLAST 2 sequences"(www.ncbi.nlm.nih.gov/blast/bl2seq/bl2.html).

Several studies have probed the impact of changes in ₂m on the stability, as well as on the structure and function, of MHC class I molecules in the context of recognition by Abs and T cells (16, 17, 18, 19, 20, 21, 22, 23, 27, 28, 29). To date no one has addressed the role of ₂m in MHC class I recognition by Ly49 receptors. Our interest in this problem was in part attracted by a discrepancy between two earlier studies using MHC class I tetramers to study Ly49 receptor specificity. Both studies demonstrated that Ly49A binds to H-2D^d, whereas Ly49C had a broader specificity and bound to, e.g., H-2K^b and H-2D^b (30, 31). However, H-2D^b tetramers were reported to bind Ly49A in one study (31) but not in the other (30). One difference, of several, between the studies was that the former used m₂m and the latter h₂m in the H-2D^b tetramers. This, together with the available structural data, prompted us to test the role of ₂m in the interaction with Ly49 receptors. In addition, a mouse expressing a transgene single-chain H-2D^d, with m₂m and peptide covalently linked to the heavy chain, failed to educate NK cells in vivo, and soluble Ly49A did not bind to the single chain H-2D^d expressed on cells derived from the lymph node. The covalent modification close to ₂m was proposed as one potential reason for the lack of Ly49A reactivity, either by steric hindrance or by a conformational change (32).

Here we describe how the change from m₂m to h₂m in three different MHC class I tetramers, H-2D^b, H-2D^d, and H-2K^b, reduces the binding to the receptors Ly49A and Ly49C. This has implications for understanding Ly49-MHC class I interactions, as well as for technical aspects of in vitro NK cell assays.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

 
Mice and cell lines

Ly49A-transgenic (tg) B6 mice (B6VA49A; founder 26) and BALB/c mice were bred and maintained at the Microbiology and Tumor Biology Center, Karolinska Institutet (Stockholm, Sweden). Animal care was in accordance with institutional guidelines. The Ly49A-tg B6 mice have been described elsewhere (33). RNK16 is a spontaneous leukemic rat NK cell line, and C1R is a human B cell lymphoblastoid cell line. Ly49A-RNK16 transfectants have been described elsewhere (34), as have Ly49C-C1R transfectants (35). The hybridomas HB-76, HB-87, HB-102, (anti-H-2D^d), HB-158, HB-41 (anti-H-2K^b), B22.249 (anti-H-2D^b), and HB-120 (anti-HLA A, B, and C) were purchased from American Type Culture Collection (Manassas, VA). Cell lines were cultured in RPMI 1640 supplemented with 10% FCS, 50 µg/ml streptomycin, 100 µg/ml penicillin, and 2 mM L-glutamine at 37°C with 7.5% CO₂. Hybridomas were cultured in DMEM supplemented with 15% FCS, hypoxanthine/aminopterin/thymidine, sodium pyruvate, 50 µg/ml streptomycin, 100 µg/ml penicillin, and 2 mM L-glutamine. The preparation of nylon wool-nonadherent (NWNA) spleen cells were prepared as described earlier (36).

Generation of MHC class I tetramers

Production of soluble MHC class I heavy chain fused with a BirA substrate peptide (H-2K^b_Bsp, H-2D^b_Bsp, H-2D^d_Bsp), m₂m, and h₂m, and the in vitro refolding of the MHC_Bsp/₂m/peptide complexes have been described previously (37, 38). The generation of the H-2D^d_Bsp cDNA construct has also been described (31). The H-2D^d_Bsp construct encodes aa 1–280 of H-2D^d, ending with a proline, followed by a glycine, serine, and Bsp 41/50 (described by P. Schatz et al.) (39). The H-2K^b_Bsp (40) and the H-2D^b_Bsp (41) constructs were gifts from J. D. Altman (Emery University, Atlanta, GA) and T. N. Schumacher (The Netherlands Cancer Institute, Amsterdam, The Netherlands), respectively. The H-2K^b_Bsp construct encodes aa 1–279, ending with a serine, followed by a glycine, serine, and Bsp 41/50 (39). The H-2D^b_Bsp construct encodes aa 1–280, ending with a threonine, followed by a glycine-serine linker (12 aa) and Bsp 85 (39). The HLA-E_Bsp cDNA construct was a gift from V. Braud (Institute of Molecular Medicine, John Radcliffe Hospital, Oxford, U.K.). The m₂m cDNA was a gift from P. J. Travers (Birkbeck College, London, U.K.), and the h₂m cDNA was provided by E. Y. Jones (Oxford Center for Molecular Science, Oxford, U.K.). Both m₂m and h₂m cDNA encode aa 1–99, plus an additional methionine at position one in h₂m, and a methionine and a glycine at position 1 and 2 in m₂m. Synthetic peptides (HIV-1-derived P18-I10 (RGPGRAFVTI), OVA_257–264 (SIINFEKL), lymphocytic choriomeningitis virus (LCMV) gp33 (KAVYNFATM), and B7L (VMAPRTVLL) were purchased from Research Genetics (Huntsville, AL). Briefly, the MHC_Bsp, ₂m, and peptide were refolded by dilution in the presence of protease inhibitors (2 µg/ml leupeptin, 2 µg/ml pepstatin A, and 0.2 mM PMSF) for 48 h. The same batch of heavy chain and peptide was used for the refoldings of each MHC. The refolded complexes were purified by size exclusion chromatography on a Superdex 200 10/30 column (Pharmacia Biotech, Uppsala, Sweden) and biotinylated by incubation with BirA enzyme and biotin (Avidity, Denver, CO) according to the instructions of the manufacturer. Free biotin was removed by gel filtration using NAP-5 desalting columns (Pharmacia Biotech). The MHC complexes were then quickly frozen in liquid nitrogen and stored at -70°C. Tetramers were produced in parallel by stepwise mixing biotinylated MHC/₂m/peptide complexes with neutravidin-PE (Molecular Probes, Eugene, OR) at a 4:1 molar ratio. The MHC class I tetramers generated were H-2D^b_Bsp refolded with m₂m and LCMV gp33 (H-2D^b/m₂m), H-2D^b_Bsp refolded with h₂m and LCMV gp33 (H-2D^b/h₂m), H-2D^d_Bsp refolded with m₂m and P18-I10 (H-2D^d/m₂m), H-2D^d_Bsp refolded with h₂m and P18-I10 (H-2D^d/h₂m), H-2K^b_Bsp refolded with m₂m and OVA_257–264 (H-2K^b/m₂m), H-2K^b_Bsp refolded with h₂m and OVA_257–264 (H-2K^b/h₂m), and HLA-E_Bsp refolded with h₂m and B7L (HLA-E). Equal quality of the paired MHC class I tetramers (with either m₂m or h₂m) was controlled by a gel-shift assay (data not shown), as well as by staining hybridomas expressing an Ab specific for the MHC class I molecule used (see Results). Furthermore, the quality of the H-2D^b tetramers was controlled by additional staining of T cell clones specific for H-2D^b/LCMV gp33.

Abs and flow cytometry

FITC-labeled DX5 mAb was purchased from PharMingen (Stockholm, Sweden). A1 (anti-Ly49A) and SW5E6 (anti-Ly49C/I) were purified from hybridoma supernatants over a protein G column (Pharmacia Biotech). All stainings were made in 50 µl on ice for 60 min, followed by three washes in PBS supplemented with 2% FCS. Stainings performed in PBS alone were similar to stainings in PBS supplemented with 2% FCS. NWNA spleen cells were incubated with an anti-FcRIII mAb (2G4) before staining with mAbs and MHC tetramers. MHC tetramers were used at concentrations ranging from 2 to 40 µg/ml. All stainings with control tetramers were performed at 40 µg/ml. Dead cells were excluded on the basis of forward and side scatter. Cells were analyzed on a FACScan or a FACSort cytometer (BD Biosciences, Mountain View, CA).

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

 
To test whether ₂m was involved in the recognition of MHC class I by Ly49 receptors, we produced MHC class I tetramers refolded with either h₂m or m₂m and measured their binding to Ly49-expressing cells by FACS. Murine MHC class I is stabilized by h₂m at least as well as by m₂m, and h₂m has been used in murine MHC class I tetramers earlier (27, 30, 40, 42, 43, 44, 45).

Binding of H-2D^b to both Ly49A and Ly49C is markedly reduced when m₂m is replaced with h₂m

A discrepancy in studies of Ly49 receptor binding specificity of H-2D^b suggested to us that the binding of H-2D^b to Ly49A, but not to Ly49C, was influenced by the ₂m subunit (30, 31). To test this hypothesis it was necessary to investigate whether a change from m₂m to h₂m in H-2D^b would affect binding to both receptors or only to Ly49A. Consequently, H-2D^b/m₂m and H-2D^b/h₂m tetramers were produced and tested for binding to Ly49 receptor-expressing cells in parallel experiments. The same batch of MHC class I heavy chain and the same peptide were used in each refolding, leaving ₂m as the variable parameter.

As hypothesized, H-2D^b/h₂m tetramers failed to bind the Ly49A-RNK16 transfectant (Fig. 1A), whereas the binding of H-2D^b/m₂m tetramers, although weak, was reproducible in a number of experiments (Fig. 1A). No staining of nontransfected RNK16 cells was observed (data not shown). Ly49A-RNK16 transfectants incubated with a control tetramer (H-2K^b/m₂m) were also negative (shown as 0 µg/ml tetramer in Fig. 1A). Because MHC class I tetramer binding is sensitive to the level of receptor expression, we also stained cells with a higher expression level of Ly49A to see whether the H-2D^b/h₂m tetramer would then bind Ly49A. Ly49A-tg B6 mice express Ly49A at a high level on CD2⁺ cells, including T and NK cells (33). DX5⁺ NWNA spleen cells from Ly49A-tg B6 mice were not stained by H-2D^b/h₂m tetramers (Fig. 1B). However, the H-2D^b/m₂m tetramers did clearly bind to Ly49A on the DX5⁺ NWNA Ly49A-tg B6 spleen cells (Fig. 1B). This binding was not observed with a control tetramer (HLA-E; shown as 0 µg/ml tetramer in Fig. 1B), and it could be completely blocked by an anti-Ly49A mAb (A1) (data not shown).

View larger version (27K): [in this window] [in a new window]   FIGURE 1. Binding of H-2Db tetramers to both Ly49A and Ly49C is markedly reduced after replacement of m2m with h2m. A, Binding of H-2Db/m2m () and H-2Db/h2m tetramers () to Ly49A-RNK16 transfectants. Staining with H-2Kb/m2m tetramers was used as control (shown as 0 µg/ml tetramer in the graph), and was similar to staining without tetramer (data not shown). No binding was seen by H-2Db tetramer to nontransfected RNK16 cells (data not shown) B, Binding of H-2Db/m2m () and H-2Db/h2m tetramers () to DX5+ NWNA spleen cells from Ly49A-tg B6 mice. Staining with HLA-E tetramers was used as control (shown as 0 µg/ml tetramer in the graph), and was similar to staining without tetramers (data not shown). The binding of H-2Db/m2m tetramers could be completely blocked by an anti-Ly49A mAb (A1) (data not shown). C, Binding of H-2Db/m2m () and H-2Db/h2m tetramers () to Ly49C-C1R transfectants. Staining with H-2Dd/m2m tetramers was used as control (shown as 0 µg/ml tetramer in the graph), and was similar to staining without tetramers (data not shown). No binding was seen by H-2Db tetramers to nontransfected C1R cells (data not shown). D, Binding of H-2Db/m2m () and H-2Db/h2m tetramers () to DX5+ NWNA spleen cells from BALB/c mice. Staining with HLA-E tetramers was used as control (shown as 0 µg/ml tetramer in the graph), and was similar to staining without tetramers (data not shown). The binding of H-2Db/m2m tetramers could be completely blocked by an anti-Ly49C/I mAb (5E6) (data not shown). Note that this graph is plotted as percent positive cells of the DX5+ population, and not mean fluorescence intensity against concentration of tetramers. Positive cells were defined as cells staining brighter than the control tetramer, as indicated by the vertical line in H. E, Binding of H-2Db/m2m () and H-2Db/h2m tetramers () to the hybridoma B22.249. Staining with HLA-E tetramers was used as control (shown as 0 µg/ml tetramer in the graph), and was similar to staining without tetramers (data not shown). No staining of a control hybridoma (HB-76) by H-2Db tetramers was observed (data not shown). F, Binding of H-2Db/m2m () and H-2Db/h2m tetramers () to a CD8+ T cell clone specific for H-2Db in complex with the peptide LCMV gp33. Staining with HLA-E tetramers was used as control (shown as 0 µg/ml tetramer in the graph), and was similar to staining without tetramers (data not shown). G, FACS histogram of DX5+ spleen cells from a Ly49A-tg B6 mouse after staining with H-2Db/m2m tetramers (solid bold line), H-2Db/h2m tetramers (solid thin line), and a control HLA-E tetramer (dashed line). The bimodal staining pattern is in part due to endogenous Ly49A expression by a fraction of the NK cells, and in part to a higher Ly49A transgene expression on DX5+ T cells. All tetramers were used at 40 µg/ml. H, FACS histogram of DX5+ spleen cells from BALB/c mice after staining with H-2Db/m2m tetramers (solid bold line), H-2Db/h2m (solid thin line), and a control HLA-E tetramer (dashed line). All tetramers were used at 40 µg/ml. Staining with the control tetramer was similar to staining with PBS alone (data not shown). A-E, Means ± SD of at least three experiments. F, Mean ± SD of two experiments. The stainings were made in parallel experiments. Note that different scales have been used for clarity.

Binding of H-2D^d refolded with h₂m to Ly49A is impaired

We further investigated whether the observed influence of the ₂m subunit on H-2D^b binding to Ly49A and Ly49C applied also to other Ly49-MHC class I interactions. The involvement of the ₂m subunit in site 2 in the structure of Ly49A in complex with H-2D^d (24) suggested to us that a change from m₂m to h₂m could also affect the interaction between H-2D^d and Ly49A. Ly49A-RNK16 transfectants were stained by H-2D^d/m₂m tetramers (Fig. 2A), but not by a control tetramer (H-2K^b/m₂m) (shown as 0 µg/ml tetramer in Fig. 2A). Moreover, no staining by H-2D^d/m₂m of nontransfected RNK16 cells was observed (data not shown). This was expected, as similar observations were made earlier (30, 31). In contrast, H-2D^d/h₂m tetramers failed to stain the same Ly49A-transfected cell line (Fig. 2A). H-2D^d/h₂m tetramers stained HB-102, hybridoma cells expressing Ig molecules specific for H-2D^d, to a similar extent as the H-2D^d/m₂m tetramers (Fig. 2C). No binding was observed when incubating HB-102 with a control tetramer (HLA-E; shown as 0 µg/ml tetramer in Fig. 2C) or when we incubated an H-2D^b-specific hybridoma (B22.249) with the H-2D^d tetramers (data not shown). Incubation of NWNA spleen cells from Ly49A-tg B6 mice with H-2D^d/h₂m tetramer failed to stain the DX5⁺ cells, whereas there was still a strong binding of H-2D^d/m₂m tetramer (Fig. 2B). A control tetramer (HLA-E) did not stain the Ly49A-positive cells (shown as 0 µg/ml tetramer in Fig. 2B), and the binding of H-2D^d/m₂m tetramers could be completely blocked by an anti-Ly49A mAb (A1) (data not shown). Importantly, we observed a weak, but reproducible, staining of Ly49A-positive T cells by the H-2D^d/h₂m tetramer (data not shown). The expression level of Ly49A on these cells is even higher than on the DX5⁺ cells. This binding was not observed when staining with a control tetramer (HLA-E), and could be blocked by the addition of A1 mAb (data not shown). In conclusion, ₂m also influences the binding of H-2D^d to Ly49A.

View larger version (31K): [in this window] [in a new window]   FIGURE 2. Binding of H-2Dd tetramers to Ly49A is markedly reduced after replacement of m2m with h2m. A, Binding of H-2Dd/m2m () and H-2Dd/h2m tetramers () to Ly49A-RNK16 transfectants. Staining with H-2Kb/m2m tetramers was used as control (shown as 0 µg/ml tetramer in the graph), and was similar to staining without tetramers (data not shown). No binding was seen by H-2Dd tetramers to nontransfected RNK16 cells (data not shown). B, Binding of H-2Dd/m2m () and H-2Dd/h2m tetramers () to DX5+ NWNA spleen cells from Ly49A-tg B6 mice. Staining with HLA-E tetramers was used as control (shown as 0 µg/ml tetramer in the graph), and was similar to staining without tetramers (data not shown). The binding of H-2Dd/m2m tetramers could be completely blocked by an anti-Ly49A mAb (A1) (data not shown). C, Binding of H-2Dd/m2m () and H-2Dd/h2m tetramers () to the hybridoma HB-102. Staining with HLA-E tetramers was used as control (shown as 0 µg/ml tetramer in the graph), and was similar to staining without tetramers (data not shown). No staining of a control hybridoma (B22.249) by H-2Dd tetramers was observed (data not shown). D, FACS histogram of DX5+ spleen cells from a Ly49A-tg B6 mouse after staining with H-2Dd/m2m tetramers (solid bold line), H-2Dd/h2m tetramers (solid thin line), and a control HLA-E tetramer (dashed line). The bimodal staining pattern is in part due to endogenous Ly49A expression by a fraction of the NK cells, and in part to a higher Ly49A transgene expression on DX5+ T cells. All tetramers were used at 40 µg/ml. A-C, Means ± SD of at least three experiments. The stainings were made in parallel experiments. Note that different scales have been used for clarity.

Binding to Ly49C is markedly reduced when H-2K^b is refolded with h₂m

Franksson et al. (35) reported that H-2K^b/OVA (SIINFEKL) conferred better protection against Ly49C⁺ NK cells than H-2K^b/SEV (FAPGNYPAL). In contrast, Su et al. (46) reported that "empty" (peptide-receptive) H-2K^b molecules conferred better protection than H-2K^b presenting the OVA peptide. Although the picture is not complete, these studies clearly show that the interaction between Ly49C and H-2K^b is influenced by the content of the MHC class I peptide binding groove. There is no evidence for such a specific peptide influence on the interaction between Ly49A and H-2D^d. This implies that the H-2K^b-Ly49C interaction differs in some aspects from that of Ly49A and H-2D^d. To test the role of the ₂m subunit in this interaction, we refolded H-2K^b tetramers with either h₂m or m₂m and compared their binding to Ly49C-expressing cells. We chose to refold the H-2K^b complexes with an OVA-derived peptide, known to permit strong binding of H-2K^b to Ly49C (31).

H-2K^b/m₂m tetramers clearly stained Ly49C-C1R transfectants (Fig. 3A), as published earlier (31). Interestingly, H-2K^b/h₂m tetramers displayed a severely reduced binding to the Ly49C-C1R transfectants (Fig. 3A). The binding of H-2K^b/h₂m was close to the negative control, although we observed, in one of four experiments, a slight binding at higher H-2K^b/h₂m tetramer concentrations (data not shown). Similar results were obtained when staining NWNA spleen cells from BALB/c mice. H-2K^b/m₂m tetramers clearly stained DX5⁺ spleen cells, whereas H-2K^b/h₂m tetramers did not stain the DX5⁺ population above background levels (Fig. 3B). To test the integrity and quality of both tetramers we stained HB-41 hybridoma cells expressing an anti-H-2K^b mAb. Both H-2K^b tetramers stained HB-41 cells to a similar level, but no staining was observed with a control HLA-E tetramer (shown as 0 µg/ml tetramer in Fig. 3C) or when staining a control hybridoma (B22.249) (data not shown). The hybridoma HB-158, producing the mAb AF6-88.5, has been demonstrated by Kuhns and Pease to be specific for an epitope dependent partially on the ₂m subunit (23). The staining of HB-158 by H-2K^b/h₂m tetramers was severely reduced, but not completely abolished, in comparison to staining by H-2K^b/m₂m tetramers (Fig. 3D), confirming the results of Kuhns and Pease (23) using a direct binding assay.

View larger version (26K): [in this window] [in a new window]   FIGURE 3. Binding of H-2Kb tetramers to Ly49C is markedly reduced after replacement of m2m with h2m. A, Binding of H-2Kb/m2m () and H-2Kb/h2m tetramers () to Ly49C-C1R transfectants. Staining with H-2Dd/m2m tetramers was used as control (shown as 0 µg/ml tetramer in the graph), and was similar to staining without tetramers (data not shown). No binding was seen by H-2Kb tetramers to nontransfected C1R cells (data not shown). B, Binding of H-2Kb/m2m () and H-2Kb/h2m tetramers () to DX5+ NWNA spleen cells from BALB/c mice. Staining with HLA-E tetramers was used as control (shown as 0 µg/ml tetramer in the graph), and was similar to staining without tetramers (data not shown). Note that this graph is plotted as percent positive cells of the DX5+ population and not mean fluorescence intensity against concentration of tetramers used. Positive cells were defined as cells staining brighter than the control tetramer, as indicated by the vertical line in E. C, Binding of H-2Kb/m2m () and H-2Kb/h2m tetramers () to the hybridoma HB-41. Staining with HLA-E tetramers was used as control (shown as 0 µg/ml tetramer in the graph), and was similar to staining without tetramers (data not shown). No staining of a control hybridoma (B22.249) by H-2Kb tetramers was observed (data not shown). D, Binding of H-2Kb/m2m () and H-2Kb/h2m tetramers () to the hybridoma HB-158. Staining with HLA-E tetramers was used as control (shown as 0 µg/ml tetramer in the graph), and was similar to staining without tetramers (data not shown). E, FACS histogram of DX5+ spleen cells from a BALB/c mouse after staining with H-2Kb/m2m tetramers (solid, bold line), H-2Kb/h2m tetramers (solid, thin line), and a control HLA-E tetramer (dashed line). All tetramers were used at 40 µg/ml. A-E, Means ± SD of at least three experiments. The stainings were made in parallel experiments. Note that different scales have been used for clarity.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Tormo et al. (24) reported two sites of interaction between receptor and ligand in the cocrystal of Ly49A and H-2D^d. ₂m contributes to 25% of the second site of interaction, site 2 (Fig. 4A). Here, one of the subunits of the Ly49A dimer lies against the surfaces of the -sheets (65% of the interface) that form the bottom of the peptide binding platform, ₃ (15% of the interface) and ₂m. Interactions at this site have been described as primarily polar, with 26 direct hydrogen bonds (24). Many other contacts between ₂m and Ly49A are bridged by numerous water molecules trapped at the interface. There is 30% difference in the amino acid sequence between m₂m and h₂m (Fig. 5). Four residues of the m₂m subunit (Lys³, Thr⁴, Gln²⁹, and Lys⁵⁸) are involved in seven direct hydrogen bond interactions with five residues (Asp¹⁹³, Arg¹⁵⁷, Val²⁴⁸, and Asp²²⁹, and Asn²⁴²) in Ly49A at site 2 (Fig. 4B) (24). Three of these residues (Lys³, Thr⁴, and Gln²⁹) form a cluster with other amino acids (Gln⁶, Thr²⁸, and residue 85; Ala in BALB/c, and Asp in C57BL/6). All of the amino acids in this cluster, except residue 4, differ between m₂m and h₂m. Thus there is an important difference between the mouse and the human, in this region of ₂m (Figs. 4 and 5). The most striking difference is residue 29, where glutamine in the mouse, involved in three hydrogen bonds with Ly49A, is substituted by a glycine in the human. It is tempting to speculate that these residues in ₂m influence the binding of Ly49 receptors to MHC class I molecules. Note that residues Arg¹⁵⁷, Asp¹⁹³, Asp²²⁹, and Val²⁴⁸ in Ly49A, which interact with this cluster, also are involved in the contact at site 1. There are additional residues contributing to the molecular contacts between Ly49A and H-2D^d at both sites, e.g., Asp²⁴¹, Asn²²⁶, Arg²²⁸, Asn²⁴², Asn²⁴⁴, Asp²⁴⁶, Gln²⁴⁷, and Phe²⁴⁹ (24).

View larger version (46K): [in this window] [in a new window]   FIGURE 4. Illustration of how 2m is involved in the binding of the Ly49A NK receptor to H-2Dd at site 2. A, In the published cocrystal m2m is responsible for 25% of the surface contact, and there are seven hydrogen bonds between the 2m subunit of H-2Dd and the Ly49A-2 subunit of the receptor dimer. Only the subunit of the Ly49A dimer that is involved in the binding with 2m at site 2 is depicted. The MHC heavy chain, the 2m subunit, the peptide binding in the 12 cleft, and the Ly49A subunit are colored in cyan, black, red, and lime green, respectively. The interface of interest is highlighted by a red square, and is described more in detail in B. B, Four residues of the m2m subunit (Lys3, Thr4, Gln29, and Lys58) are involved in seven direct hydrogen bond interactions with five residues (Asp193, Arg157, Val248, Asp229, and Asn242) in Ly49A at site 2. Three of these residues (Lys3, Thr4, and Gln29) form a cluster with other amino acids (Gln6, Thr28, and residue 85), all of which, with the exception of position 4, differ between m2m and h2m. The most striking difference is residue 29, where glutamine in the mouse, involved in three hydrogen bonds with Ly49A, is substituted by a glycine in the human. The MHC heavy chain, the 2m subunit, and the Ly49A subunit are colored in cyan, black, and green, respectively. The seven hydrogen bonds are depicted as dashed lines.

In summary, the replacement of m₂m with h₂m could have a direct effect on the Ly49A interaction at site 2, or an indirect effect at site 1. Therefore, our result can not be interpreted in favor of one of the two contact sites as the physiologically important one. It is not excluded that both are important, i.e., that the same H-2D^d molecule can be recognized, at the same time, by two different Ly49A receptors, either in trans (between NK and target cell), in cis (between molecules on the NK cell surface itself), or in both cis and trans.

MHC class I tetramers are excellent tools to investigate the specificity of various MHC class I binding receptors, but the analysis is not without pitfalls; tetramer binding is highly dependent on the level of receptor expression, with sharp threshold effects. There is always a risk of false negative results due to too low expression of the receptor of interest. An earlier study of MHC class I-Ly49 receptor interaction based on MHC class I tetramer binding reported seemingly controversial results to ours (30). In the study by Hanke et al. (30), both H-2D^b/h₂m and H-2K^b/h₂m tetramers were found to bind to Ly49C. In our study we could observe virtually no binding of MHC class I refolded with h₂m to the above receptors. However, we could observe some binding of H-2D^d/h₂m to T cells in a Ly49A-tg mouse with high expression of Ly49A. Differences in the level of receptor expression in the two systems is one possible explanation for the differing results, demonstrating that the effect of the change of the ₂m subunit is not absolute when it comes to Ly49-MHC class I interactions. However, it is clear from our study that changing the ₂m subunit clearly can decrease the binding between MHC class I tetramers and Ly49 receptors.

The valency of the MHC tetramer can vary between batches, affecting the binding to the receptors expressed on the cells. Therefore, it is important to produce the compared MHC tetramers under the same conditions, including usage of the same batch of heavy chain and peptide, measuring the concentrations and doing the tetramerizations in parallel. Furthermore, it is important to verify that production under these stringent conditions results in MHC tetramers of comparable quality. In this study, we have primarily used staining of specific B cell hybridomas as a control. The affinity between the MHC and the Ig molecules produced by the B cell hybridoma is higher than the one between, e.g., Ly49 receptors and MHC class I; thus, the former interaction may be less sensitive to possible differences in MHC class I valency between tetramer batches. Therefore, we have also stained T cell clones specific for H-2D^b complexed with a peptide from LCMV gp33. Considering the stringency of the production, the staining of the B cell hybridomas, and the control staining of the specific T cell clones, it is unlikely that our results depend on a systematic difference in quality between tetramer batches based on h₂m vs m₂m.

Considering the high homology (76%) between h₂m and b₂m, and the similar affinity for murine MHC class I molecules, our results implicate that precaution should be taken when interpreting results from assays based on cells grown in bovine serum. Up to 90% of the MHC class I molecules have been demonstrated to exchange m₂m to h₂m when incubating murine tumor cells with fluorescent h₂m (15). This is likely to be true also for b₂m, which can be present in large amounts in FCS. Given the effect on MHC class I tetramer binding to Ly49 receptors, the exchange from m₂m to b₂m or h₂m could reduce the protective capacity by MHC class I molecules and, thereby, lead to false results also in functional assays. For example, this effect could potentially explain why MHC class I specificity is more distinct in in vivo rejection assays than in in vitro assays of NK cell cytotoxicity when tumor targets are used (49).

In summary, the influence of the ₂m subunit on Ly49-MHC class I interactions can be mediated by direct interference within a contact site that involves ₂m, such as site 2 in the Ly49A-H-2D^d interaction, by structural alterations distal to ₂m, or both. Additional structural investigations will be needed to elucidate the effect of h₂m on murine MHC class I structures, and the effect on Ly49 receptor binding. Today there are no available structures for comparison of the same MHC class I/peptide combination refolded with m₂m and h₂m, respectively. Such studies are required to pinpoint the motifs responsible for the difference in binding of MHC class I tetramers to Ly49 receptors, imposed by the ₂m subunit. They would also enlighten other conformational aspects of the MHC class I structure, important for recognition by NK cells, Abs, and T cells.

Note added in proof.

During the revision of this manuscript, Matsumoto et al. (50) published data demonstrating a role of ₂m in recognition of H-2D^d by Ly49A. They show that Ly49A tetramers do not bind significantly to cells expressing H-2D^d complexed with h₂m. Furthermore, they demonstrate that the interaction between H-2D^d and Ly49A can be blocked by an anti-m₂m mAb, both in binding assays and functional assays. The data by Matsumoto et al. support the notion that ₂m is important for Ly49-MHC interactions, and we can in this study confirm and extend the importance of ₂m in recognition by Ly49 receptors.

	Acknowledgments

 
We thank Drs. J. D. Altman, T. N. Schumacher, and V. Braud for providing the H-2K^b_Bsp, H-2D^b_Bsp, and HLA-E_bsp cDNA, respectively. We also thank C. L. Sentman, L. Fahlén, and L. Öberg for kind help and for providing the B6VA49A-tg mice.

	Footnotes

 
¹ This project was supported by a grant from the "Network for Inflammation Research" funded by the Swedish Foundation for Strategic Research (to J.M.), as well as grants from the Swedish Cancer Society (to K.K.), the Swedish Medical Research Council (to K.K.), Karolinska Institutet (to A.A.), Robert Lundbergs minnesstiftelse (to A.A.), Svenska Sällskapet för Medicinsk Forskning (to A.A.), and Stiftelsen Lars Hiertas minne (to A.A.).

² Address correspondence and reprint requests to Dr. Jakob Michaëlsson, Microbiology and Tumor Biology Center, Karolinska Institutet, Box 280, SE-171 77, Stockholm, Sweden. E-mail address: jakob.michaelsson{at}mtc.ki.se

³ Abbreviations used in this paper: ₂m, ₂-microglobulin; m₂m, murine ₂m; h₂m, human ₂m; b₂m, bovine ₂m; NWNA, nylon wool-nonadherent; tg, transgenic; LCMV, lymphocytic choriomeningitis virus.

⁴ Waldenström, M., A. Achour, J. Michaëlsson, A. Rölle, and K. Kärre. The role of an exposed loop in the ₂ domain in the mouse MHC class I H-2D^d molecule for recognition by the monoclonal antibody 34-5-8S and the NK cell receptor Ly49A. Submitted for publication.

Received for publication November 20, 2000. Accepted for publication April 3, 2001.

	References

Top Abstract Introduction Materials and Methods Results Discussion References

 

1. Yokoyama, W. M.. 1999. Natural killer cells. W. E. Paul, ed. Fundamental Immunology 575. Lippincot-Raven Publishers, Philadelphia.
2. Drickamer, K.. 1999. C-type lectin-like domains. Curr. Opin. Struct. Biol. 9:585.[Medline]
3. Lanier, L. L.. 1998. NK cell receptors. Annu. Rev. Immunol. 16:359.[Medline]
4. Long, E. O.. 1999. Regulation of immune responses through inhibitory receptors. Annu. Rev. Immunol. 17:875.[Medline]
5. Makrigiannis, A. P., S. K. Anderson. 2000. Ly49 gene expression in different inbred mouse strains. Immunol. Res. 21:39.[Medline]
6. Raulet, D. H., W. Held, I. Correa, J. R. Dorfman, M. F. Wu, L. Corral. 1997. Specificity, tolerance and developmental regulation of natural killer cells defined by expression of class I-specific Ly49 receptors. Immunol. Rev. 155:41.[Medline]
7. Townsend, A. R., J. Rothbard, F. M. Gotch, G. Bahadur, D. Wraith, A. J. McMichael. 1986. The epitopes of influenza nucleoprotein recognized by cytotoxic T lymphocytes can be defined with short synthetic peptides. Cell 44:959.[Medline]
8. Ljunggren, H. G., K. Kärre. 1985. Host resistance directed selectively against H-2-deficient lymphoma variants: analysis of the mechanism. J. Exp. Med. 162:1745.[Abstract/Free Full Text]
9. Kärre, 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.[Medline]
10. Ljunggren, H. G., K. Kärre. 1990. In search of the "missing self": MHC molecules and NK cell recognition. Immunol. Today 11:237.[Medline]
11. Gumperz, J. E., P. Parham. 1995. The enigma of the natural killer cell. Nature 378:245.[Medline]
12. III Gates, F. T., J. E. Coligan, T. J. Kindt. 1981. Complete amino acid sequence of murine ₂-microglobulin: structural evidence for strain-related polymorphism. Proc. Natl. Acad. Sci. USA 78:554.[Abstract/Free Full Text]
13. Bernabeu, C., M. van de Rijn, P. G. Lerch, C. P. Terhorst. 1984. ₂-microglobulin from serum associates with MHC class I antigens on the surface of cultured cells. Nature 308:642.[Medline]
14. Kubota, K.. 1984. Association of serum ₂-microglobulin with H-2 class I heavy chains on the surface of mouse cells in culture. J. Immunol. 133:3203.[Abstract]
15. Hochman, J. H., Y. Shimizu, R. DeMars, M. Edidin. 1988. Specific associations of fluorescent ₂-microglobulin with cell surfaces: the affinity of different H-2 and HLA antigens for ₂-microglobulin. J. Immunol. 140:2322.[Abstract]
16. Ferrier, P., C. Layet, D. H. Caillol, B. R. Jordan, F. A. Lemonnier. 1985. The association between murine ₂-microglobulin and HLA class I heavy chains results in serologically detectable conformational changes of both chains. J. Immunol. 135:1281.[Abstract]
17. Mierau, R., P. J. Robinson, A. R. Sanderson, E. Genth, M. Cramer. 1987. Antigenic determinants shared between HLA-A, -B, -C antigens and H-2 class I molecules modified by bovine ₂-microglobulin. [Published erratum appears in 1988 Immunogenetics 27:158.]. Immunogenetics 26:351.[Medline]
18. Nieto, M. C., E. S. Song, D. McKinney, M. McMillan, R. S. Goodenow. 1989. The association of H-2L^d with human ₂-microglobulin induces localized conformational changes in the -1 and -2 superdomain. Immunogenetics 30:361.[Medline]
19. Schmidt, W.. 1988. Monoclonal antibodies recognize interaction determinants of MHC plus foreign ₂-microglobulin. Immunogenetics 28:385.[Medline]
20. Smith, M. H., B. H. Barber. 1990. The conformational flexibility of class I H-2 molecules as revealed by anti-peptide antibodies specific for intracytoplasmic determinants: differential reactivity of ₂-microglobulin "bound" and "free" H- 2K^b heavy chains. Mol. Immunol. 27:169.[Medline]
21. Smith, M. J., T. Basora, J. E. Kieran, M. C. Nieto. 1993. Alteration within a discrete region of the H-2L^d 1 helix upon association with human ₂-microglobulin. Immunol. Cell Biol. 71:145.
22. Schultz, C. S., R. A. Rodriguez, E. A. Chew, C. Dimaano, F. M. Li, M. D. Santos, M. C. Nieto. 1998. Effects of ₂-microglobulin mutations on the ₁ helical region of H-2L^d. Immunogenetics 48:273.[Medline]
23. Kuhns, S. T., L. R. Pease. 1998. A region of conformational variability outside the peptide-binding site of a class I MHC molecule. J. Immunol. 161:6745.[Abstract/Free Full Text]
24. 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.[Medline]
25. Kåse, A., M. H. Johansson, M. Y. Olsson-Alheim, K. Kärre, P. Höglund. 1998. External and internal calibration of the MHC class I-specific receptor Ly49A on murine natural killer cells. J. Immunol. 161:6133.[Abstract/Free Full Text]
26. Andersson, M., S. Freland, M. H. Johansson, R. Wallin, J. K. Sandberg, B. J. Chambers, B. Christensson, U. Lendahl, S. Lemieux, M. Salcedo, H. G. Ljunggren. 1998. MHC class I mosaic mice reveal insights into control of Ly49C inhibitory receptor expression in NK cells. J. Immunol. 161:6475.[Abstract/Free Full Text]
27. Shields, M. J., L. E. Moffat, R. K. Ribaudo. 1998. Functional comparison of bovine, murine, and human ₂-microglobulin: interactions with murine MHC I molecules. Mol. Immunol. 35:919.[Medline]
28. Shields, M. J., R. Kubota, W. Hodgson, S. Jacobson, W. E. Biddison, R. K. Ribaudo. 1998. The effect of human ₂-microglobulin on major histocompatibility complex I peptide loading and the engineering of a high affinity variant: implications for peptide-based vaccines. J. Biol. Chem. 273:28010.[Abstract/Free Full Text]
29. Shields, M. J., N. Assefi, W. Hodgson, E. J. Kim, R. K. Ribaudo. 1998. Characterization of the interactions between MHC class I subunits: a systematic approach for the engineering of higher affinity variants of ₂-microglobulin. J. Immunol. 160:2297.[Abstract/Free Full Text]
30. Hanke, T., H. Takizawa, C. W. McMahon, D. H. Busch, E. G. Pamer, J. D. Miller, J. D. Altman, Y. Liu, D. Cado, F. A. Lemonnier, et al 1999. Direct assessment of MHC class I binding by seven Ly49 inhibitory NK cell receptors. Immunity 11:67.[Medline]
31. Michaëlsson, J., A. Achour, M. Salcedo, A. Kåse-Sjöström, J. Sundbäck, R. A. Harris, K. Kärre. 2000. Visualization of inhibitory Ly49 receptor specificity with soluble major histocompatibility complex class I tetramers. Eur. J. Immunol. 30:300.[Medline]
32. Chung, D. H., J. Dorfman, D. Plaksin, K. Natarajan, I. M. Belyakov, R. Hunziker, J. A. Berzofsky, W. M. Yokoyama, M. G. Mage, D. H. Margulies. 1999. NK and CTL recognition of a single chain H-2D^d molecule: distinct sites of H-2D^d interact with NK and TCR. J. Immunol. 163:3699.[Abstract/Free Full Text]
33. Fahlen, L., T. L. Öberg, N. K. Brännström, U. Lendahl Khoo, C. L. Sentman. 2000. Ly49A expression on T cells alters T cell selection. Int. Immunol. 12:215.[Abstract/Free Full Text]
34. Nakamura, M. C., E. C. Niemi, M. J. Fisher, L. D. Shultz, W. E. Seaman, J. C. Ryan. 1997. Mouse Ly-49A interrupts early signaling events in natural killer cell cytotoxicity and functionally associates with the SHP-1 tyrosine phosphatase. J. Exp. Med. 185:673.[Abstract/Free Full Text]
35. Franksson, L., J. Sundbäck, A. Achour, J. Bernlind, R. Glas, K. Kärre. 1999. Peptide dependency and selectivity of the NK cell inhibitory receptor Ly-49C. Eur. J. Immunol. 29:2748.[Medline]
36. Coligan, J. E., A. M. Kruisbeek, D. H. Margulies, E. M. Shevach, W. Stober. 1994. In Current Protocols in Immunology Vol. 1: Wiley, New York.
37. Altman, J. D., P. A. H. Moss, P. J. R. Goulder, D. H. Barouch, M. G. McHeyzer-Williams, J. I. Bell, A. J. McMichael, M. M. Davis. 1996. Phenotypic analysis of antigen-specific T lymphocytes. [Published erratum appears in 1998 Science 280:1821.]. Science 274:94.[Abstract/Free Full Text]
38. Achour, A., R. A. Harris, K. Persson, J. Sundbäck, C. L. Sentman, G. Schneider, Y. Lindqvist, K. Kärre. 1999. Murine class I major histocompatibility complex H-2D^d: expression, refolding and crystallization. Acta Crystallogr. D 55:260.[Medline]
39. 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.[Medline]
40. Stevenson, P. G., G. T. Belz, J. D. Altman, P. C. Doherty. 1998. Virus-specific CD8⁺ T cell numbers are maintained during -herpesvirus reactivation in CD4-deficient mice. Proc. Natl. Acad. Sci. USA 95:15565.[Abstract/Free Full Text]
41. Haanen, J. B., M. C. Wolkers, A. M. Kruisbeek, T. N. Schumacher. 1999. Selective expansion of cross-reactive CD8⁺ memory T cells by viral variants. J. Exp. Med. 190:1319.[Abstract/Free Full Text]
42. Pedersen, L. O., A. Stryhn, T. L. Holter, M. Etzerodt, J. Gerwien, M. H. Nissen, H. C. Thogersen, S. Buus. 1995. The interaction of ₂-microglobulin (₂m) with mouse class I major histocompatibility antigens and its ability to support peptide binding: a comparison of human and mouse ₂m. Eur. J. Immunol. 25:1609.[Medline]
43. Murali-Krishna, K., J. D. Altman, M. Suresh, D. J. Sourdive, A. J. Zajac, J. D. Miller, J. Slansky, R. Ahmed. 1998. Counting antigen-specific CD8 T cells: a reevaluation of bystander activation during viral infection. Immunity 8:177.[Medline]
44. Flynn, K. J., G. T. Belz, J. D. Altman, R. Ahmed, D. L. Woodland, P. C. Doherty. 1998. Virus-specific CD8⁺ T cells in primary and secondary influenza pneumonia. Immunity 8:683.[Medline]
45. Vance, R. E., J. R. Kraft, J. D. Altman, P. E. Jensen, D. H. Raulet. 1998. Mouse CD94/NKG2A is a natural killer cell receptor for the nonclassical major histocompatibility complex (MHC) class I molecule Qa-1(b). J. Exp. Med. 188:1841.[Abstract/Free Full Text]
46. Su, R. C., S. K. Kung, E. T. Silver, S. Lemieux, K. P. Kane, R. G. Miller. 1999. Ly-49CB6 NK inhibitory receptor recognizes peptide-receptive H-2K^b. J. Immunol. 163:5319.[Abstract/Free Full Text]
47. Waldenström, M., J. Sundbäck, M. Y. Olsson-Alheim, A. Achour, K. Kärre. 1998. Impaired MHC class I (H-2D^d)-mediated protection against Ly-49A⁺ NK cells after amino acid substitutions in the antigen binding cleft. Eur. J. Immunol. 28:2872.[Medline]
48. 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-2D^d. J. Exp. Med. 192:447.[Abstract/Free Full Text]
49. Höglund, P., J. Sundbäck, M. Y. Olsson-Alheim, M. Johansson, M. Salcedo, C. Ohlen, H. G. Ljunggren, C. L. Sentman, K. Kärre. 1997. Host MHC class I gene control of NK-cell specificity in the mouse. Immunol. Rev. 155:11.[Medline]
50. Matsumoto, N., M. Mitsuki, K. Tajima, W. M. Yokoyama, K. Yamamoto. 2001. 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. J. Exp. Med. 193:147.[Abstract/Free Full Text]

This article has been cited by other articles: