We generated transgenic mice expressing a single-chain β2-microglobulin (β2m)-H-2Dd. The cell-surface β2m-H-2Dd molecule was expressed on a β2m-deficient background and reacted with appropriate mAbs. It was of the expected m.w. and directed the normal development of CD8+ T cells in the thymus of a broad TCR repertoire. It also presented both exogenously provided and endogenous peptide Ags to effector CD8+ T cells. In tests of NK cell education and function, it failed to reveal any interaction with NK cells, suggesting that the site of the interaction of NK receptors with H-2Dd was disrupted. Thus, the sites of TCR and NK receptor interaction with H-2Dd are distinct, an observation consistent with independent modes of TCR and NK receptor evolution and function.

The proper expression of MHC class I molecules (MHC-I)4 is critical for the development and function of NK cells (1, 2) and CD8+ T lymphocytes (3, 4). Although both CD8+ T cells and NK cells can recognize the same MHC-I on target cells, the precise requirements are quite different (5). CD8+ T cells use their clonally expressed and somatically rearranged TCR to recognize peptide fragments presented on APC in complex with MHC-I (6, 7, 8). In contrast, the specificity of some NK receptors (NKR) is predominantly influenced by MHC-I itself and is unaffected by the particular peptide bound (9, 10, 11), while some other NKR exhibit a peptide preference (12, 13). NK cells employ a number of different receptors, both activating and inhibitory. Among the best understood NKR are the inhibitory receptors of the Ly49 family in the mouse (14). These NKR transmit inhibitory signals delivered upon engagement of MHC-I on target cells (15, 16, 17). The activation of the mature NK cell is modulated by the interaction of its inhibitory receptor(s) with appropriately conformed MHC-I on target cells. In addition, MHC-I expression of the host controls the proportion of NK cells expressing particular Ly49 receptors, and MHC-I expression tunes the functional potential of the NK cells in a developmental manner (18, 19, 20, 21). Thus, the NK cell exploits the engagement of MHC-I in both education and effector phases.

Crystallographic structures of several TCR/MHC-I complexes indicate a common orientation of the αβ TCR in interacting with amino acid side chains of both the MHC-I molecule and the bound peptide (22, 23, 24, 25, 26). In contrast, relatively little is known about the structure of Ly49 receptors. Ly49A is a member of a family of related proteins encoded by closely linked genes and by amino acid sequence is distantly related to members of the C-type lectin family (14, 27). Ab blocking studies and transfection of target cells with MHC-I-encoding genes indicate that Ly49A is an inhibitory receptor for the H-2Dd and H-2Dk molecules (15, 28) and that this inhibitory effect for H-2Dd results from interactions with the α1α2 domains (11, 15). Other studies indicate that Ly49A binds H-2Dd directly, although the precise location of the Ly49A interaction with the MHC-I molecule is unclear (29, 30, 31).

We previously described the expression and function in transfected cells of single-chain (Sc) forms of the MHC-I molecules, H-2Dd and HLA-A2, which consist of the MHC-I L chain, β2-microglobulin (β2m), covalently linked to the MHC-I H chain through a peptide spacer (32, 33, 34). Others have reported similar Sc constructs of H-2Kd (35) and HLA-A2 (36). Sc H-2Dd molecules were expressed on the surface of transfected cells as detected by flow cytometry and could stimulate T hybridoma cells (B4.2.3) specific for H-2Dd complexed with P18-I10, a decapeptide from the HIV-1 envelope glycoprotein V3 loop (37, 38) in vitro when loaded with peptide. Analysis of the fine specificity of Ag presentation by the Sc molecules revealed only subtle quantitative differences when a panel of synthetic substituted peptide Ags was employed, suggesting that there are no major differences in the way the Sc molecules bind peptide or in the conformation of the resulting MHC/peptide complex (39). These data indicated that the overall conformation of the Sc H-2Dd molecule was adequate for binding peptides and for stimulating H-2Dd-restricted T cells. Because these earlier experiments addressed issues of mature T cell recognition of MHC/peptide complexes, we expected that transgenic expression of the Sc MHC-I molecules would allow questions of T cell and NK cell development and education to be explored. Here, we examine the expression, both in B6 and in B6 β2m−/− mice, of a Sc H-2Dd (ScDd) molecule and examine its recognition in both education and effector phases by T and NK cells. These results lead to conclusions about the nature of the site of TCR and NKR interaction with H-2Dd.

To obtain a chimeric gene encoding a Sc (β2m-spacer-MHC-I H chain) H-2Dd, in a form suitable for expression as a transgene, three different fragments from two plasmids were used: 1) a 5′ fragment (XbaI-BamHI) from the pDd-1 plasmid (40, 41) containing promoter sequences from ∼400 bp upstream of the initiation ATG codon, modified in its BamHI site with Klenow polymerase and dGTP and dATP; 2) a SalI-FspI fragment derived from a cDNA for a Sc H-2Dd construct that codes for β2m, a peptide linker, and the mature H-2Dd H chain protein through to the end of the α2 domain (encoded by exon 3), which had been modified at the SalI site with Klenow polymerase and dTTP and dCTP (32); and 3) an FspI-EcoRI fragment (which extends from the end of exon 3 to the 3′ untranslated region) from pDd-1. These were ligated in two sequential steps into the pBluescript-IIKS −(+) vector.

The DNA construct encoding the Sc H-2Dd molecule was injected into fertilized eggs of homozygous C57BL/6NCr mice to generate transgenic mice. Two transgenic founder mice were backcrossed to B6 mice that contained an induced defect in the expression of β2m (C57BL/6GphTac-{Ko}B2m N5, β2m knockout (β2m−/−) mice (3), maintained and bred at Taconic, Germantown, NY) to generate transgenic mice expressing cell-surface Sc H-2Dd in the absence of other β2m-dependent molecules. (The formal names for the transgenic strains are C57BL/6NCr tScβ2mDd-1 and -2, and we will refer to them here as ScDd-1 and ScDd-2, respectively. When bred onto the β2m−/− strain, we refer to these as ScDd-1 β2m−/− and ScDd-2 β2m−/− as well.) Offspring were screened for H-2 expression with Abs specific for H-2Dd (34-2-12), H-2Db (28-14-8), and H-2Kb (AF6-88.5) by flow cytometry. Most of the experiments reported here have been performed with ScDd-1, and some have been performed with ScDd-2. Surface expression of the transgene-encoded molecule and functional behavior were indistinguishable for the two strains.

The following Abs, purchased from PharMingen, San Diego, CA, were used: anti-CD16/CD32 (2.4G2); FITC-conjugated anti-H-2Dd (34-5-8, α1α2 domain specific), (34-2-12, which binds the α3 domain), (34-4-20), (3-25.4); anti-H-2Kb (AF6-88.5); anti-H-2Db (KH95); anti-CD3ε (145-2C11); anti-CD8 (53-6.7); anti-Ly49A (A1); PE-conjugated CD4 (H129.19); anti-TCR (TCR) Vα2 (B20.1); anti-Vα8 (B21.14); anti-Vα11 (RR8-1); anti-Vβ2 (B20.6); anti-Vβ3 (KJ25); anti-Vβ4 (KT4); anti-Vβ5 (MR9-4); anti-Vβ6 (RR4-7); anti-Vβ7 (TR310); anti-Vβ8 (MR5-2); anti-Vβ9 (MR10-2); anti-Vβ10 (B21.5); anti-Vβ11 (RR3-15); anti-Vβ13 (MR12-3); biotinylated anti-NK1.1 (PK136); anti-TCR Vα3.2 (RR3-16); anti-Vβ12 (MR11-1); anti-Vβ14 (14-2); anti-Vβ17 (KJ23); cychrome-conjugated anti-TCR Cβ (H57-597); and streptavidin-cychrome. FITC-conjugated F(ab′)2 goat anti-mouse IgG mAb was purchased from Jackson ImmunoResearch (West Grove, PA). A1 (anti-Ly-49A), SW5E6 (anti-Ly49C/I), and 4D11 (anti-Ly49G2) were also used. 34-5-8S (anti-α1α2 of H-2Dd) and 34-4-20S (anti-H-2Dd) were obtained from American Type Culture Collection (Manassas, VA). Ab was purified from cell-culture supernatant by protein A or protein G affinity chromatography and conjugated to FITC or biotin by standard protocols.

C57BL/6 (B6), BALB/c, β2m−/−, and D8 (H-2Dd transgenic B6 (42)) mice were obtained from the National Institute of Allergy and Infectious Diseases production facility at Taconic (Germantown, NY). B6.C-H2bm1/By and C57BL6/J-H2bm3/Eg, which carry the H-2Kbm1 and H-2Kbm3 mutations of H-2Kb mice, were purchased from The Jackson Laboratory (Bar Harbor, ME) and are referred to here as B6.bm1 and B6.bm3, respectively.

The expression of the ScDd transgene and lack of normal level of H-2Db and H-2Kb were documented by direct immunofluorescence analyses using FITC-conjugated 34-2-12, 3-25.4 (anti-H-2Dd), KH95 (anti-H-2Db), and AF6-88.5 (anti-H-2Kb) mAbs and indirect staining using biotinylated 28-14-8 (anti-H-2Db) followed by PE-conjugated streptavidin. To block Ab binding to Fc receptors, all samples were pretreated with anti-CD16/CD32 mAb. Percentages of CD4+ and CD8+ T cells from lymph nodes and spleen were determined among the TCR-positive cells. To study expression of NKR on NK cells, spleen cells were treated with ACK lysis buffer (Biofluids, Rockville, MD), passed over nylon wool columns to remove T cells, and stained with PE-conjugated anti-NK1.1 and FITC-conjugated anti-Ly49A, C/I, or G2 mAbs.

Cells were surface-biotinylated using NHS-LC-biotin (Pierce, Rockfiord, IL) as previously described (43). The cells were then solubilized with 1% Nonidet P-40 in 10 mM Tris-HCl, pH 7.2, 140 mM NaCl, 1 mM PMSF, 5 mM iodoacetamide, 1 mM sodium orthovanadate, and Complete protease inhibitor mixture (Boehringer Mannheim, Indianapolis, IN). Nuclei were removed by centrifugation, and lysates from 5 × 107 cells were precleared with protein G-Sepharose and then subjected to immunoprecipitation with 10 μg of 34-2-12 and 150 μl of a 10% slurry of protein G-Sepharose beads (Pharmacia Biotech, Uppsala, Sweeden). Beads were washed and boiled in 1% SDS and 10 mM iodoacetamide for 5 min, and eluted proteins were then separated on a 14% SDS-polyacrylamide gel and transferred to an Immobilon P membrane (Millipore, Bedford, MA). Biotinylated proteins were visualized with streptavidin-HRP (Zymed, South San Francisco, CA) and enhanced chemiluminescence (Amersham, Chicago, IL).

ScDd β2m−/− and D8 mice were immunized by i.p. injection of 2 × 107 pfu of vaccinia virus expressing HIV-1 envelope glycoprotein gp160IIIB (vPE16, the gift of P. Earl and B. Moss) (44). Three weeks later, splenocytes were cultured at 5 × 106/ml in 24-well culture plates in complete T cell medium (RPMI 1640 containing 10% FBS, 2 mM l-glutamine, penicillin (100 U/ml), streptomycin (100 mg/ml), and 50 μM 2-ME). Three days later, the cultures were supplemented with one-tenth volume of T-Stim (Collaborative Biomedical Products, Bedford, MA) as a source of IL-2. Spleen cells were stimulated in vitro for 7 days with 1 μM P18IIIB-I10 (RGPGRAFVTI) peptide together with 4 × 106 irradiated (3300 rad) syngeneic spleen cells as APC. Cytolytic activity of CTL lines was measured by a standard 4-h 51Cr-release assay (45). SEs of the mean of triplicate cultures were all <5% of the mean.

Primary mixed lymphocyte cultures were established essentially as described previously (46). Briefly, 2.5 × 107 responder splenocytes and 2.5 × 107 irradiated stimulator cells (3000 rad from a 137Cs source) were cultured together in 20 ml of complete T media with 5% FCS in upright T-25 flasks (Corning Glass Works, Corning, NY) for 5 days. Fresh NK effector cells were depleted of RBC, and nylon wool nonadherent splenocytes were taken from groups of mice that had been treated 20–24 h previously with 150 μg poly I:C (Sigma, St. Louis, MO), an NK stimulator. Four-day lymphokine-activated killer (LAK) NK effector cells were prepared by a procedure based upon that of Chadwick and Miller (47). Briefly, splenocytes were depleted of RBC by hypotonic lysis and passed over nylon wool. Nylon wool nonadherent cells were cultured for 4 days in RPMI 1640 plus 10% FCS, supplements (including 50 μM 2-ME), and 400 ng/ml recombinant human IL-2 (Chiron, Emeryville, CA). Adherent LAK (A-LAK) cells were prepared as described (15).

For target cell preparation, splenocytes were cultured in complete T medium containing 5% FCS for 24–30 h in 24-well plates (Falcon Plastics, Lincoln Park, NJ) at 2 × 106 cells/ml with 2 μg/ml Con A (Sigma). On the day of assay, one-tenth volume of 1 M methyl-α-d-mannopyranoside (Sigma) in RPMI 1640 or PBS was added to target cell cultures to block Con A sites before labeling for 1–2 h in 100 μl of 10 mCi/ml [51Cr]Na2CrO4 (Amersham, Arlington Heights, IL) in PBS. All points were determined in triplicate using 5 × 103 to 1 × 104 target cells per well.

Bone marrow cells were obtained by flushing the lumina of the tibiae and femora of donor mice with HBSS under aseptic conditions. Suspensions of 3 × 106 cells were injected i.v. (subocular under anesthesia or by tail vein) into groups of six (or occasionally five) irradiated (900 rad from a 137Cs source) hosts. Five days after inoculation with bone marrow cells, mice were injected i.p. with 3 mCi of [125I] iododeoxyuridine (125IUdR; Amersham). Then, 20–24 h after injection of radiolabel, the mice were sacrificed and incorporation of radioactivity into the spleens was determined by gamma spectroscopy.

The extracellular portion of Ly49A (amino acids 67 to 262) was expressed in bacteria as inclusion bodies, solubilized, refolded, and purified as described in detail elsewhere.5 The purified protein was chemically biotinylated with NHS-LC-biotin (Pierce) and further purified. Lymph node cells were stained for flow cytometry analysis using the biotinylated Ly49A and PE-streptavidin.

To explore the expression of the ScDd and other MHC-I molecules in β2m−/− transgenic mice, lymph node cells from ScDd β2m−/−, β2m−/−, BALB/c, and B6 mice were stained with mAb against H-2Dd, H-2Db, and H-2Kb. As expected, lymph node cells of the β2m−/− transgenic mice expressed no dectectable H-2Kb or H-2Db above the level observed in β2m−/− nontransgenic cells (Fig. 1,A). In contrast, mice homozygous for the ScDd transgene, expressed in the β2m−/− background, revealed apparently normal levels of H-2Dd epitopes as indicated by reactivity with Abs 34-2-12 (α3 domain), and 3-25.4 (α1α2 domain) (Fig. 1 A) as well as with 34-5-8 (α1α2 domain) and 34-4-20 (data not shown).

FIGURE 1.

Expression of ScDd in the absence of normal β2m. A, Lymph node cells from ScDd β2m−/−, BALB/c, β2m−/−, and B6 mice were stained with FITC-conjugated-mAbs against H-2Dd, Db, and Kb molecules and PE-conjugated-CD4 mAb. Analysis for the expression of MHC-I molecules was performed on the gated CD4-positive cells. B, Lymph node cells from C3H/HeJ, ScDd β2m−/−, β2m−/−, and B6 mice were incubated with biotinylated 28-14-8 (anti-α3 of H-2Db) and stained with PE-streptavidin. C, Immunoprecipitation of transgenic Sc H-2Dd. Spleen cells from ScDd β2m−/−, B6, and BALB/c mice were surface-biotinylated and lysed. The lysates were precipitated with 34-2-12 (anti-α3 of H-2Dd) and protein-G Sepharose beads, washed, separated on 14% SDS polyacrylamide gel, and transferred to Immobilon P membrane. The biotinylated proteins were visualized with strepavidin-HRP and enhanced chemiluminescent reaction; lane 1, BALB/c; lane 2, ScDd β2m−/−; lane 3, B6.

FIGURE 1.

Expression of ScDd in the absence of normal β2m. A, Lymph node cells from ScDd β2m−/−, BALB/c, β2m−/−, and B6 mice were stained with FITC-conjugated-mAbs against H-2Dd, Db, and Kb molecules and PE-conjugated-CD4 mAb. Analysis for the expression of MHC-I molecules was performed on the gated CD4-positive cells. B, Lymph node cells from C3H/HeJ, ScDd β2m−/−, β2m−/−, and B6 mice were incubated with biotinylated 28-14-8 (anti-α3 of H-2Db) and stained with PE-streptavidin. C, Immunoprecipitation of transgenic Sc H-2Dd. Spleen cells from ScDd β2m−/−, B6, and BALB/c mice were surface-biotinylated and lysed. The lysates were precipitated with 34-2-12 (anti-α3 of H-2Dd) and protein-G Sepharose beads, washed, separated on 14% SDS polyacrylamide gel, and transferred to Immobilon P membrane. The biotinylated proteins were visualized with strepavidin-HRP and enhanced chemiluminescent reaction; lane 1, BALB/c; lane 2, ScDd β2m−/−; lane 3, B6.

Close modal

Although β2m−/− cell lines and mice are profoundly deficient in expression of MHC-I molecules, it is well known that low levels of H-2Db are detectable by flow cytometry or by sensitivity to lysis by H-2Db-specific alloreactive CTL (48, 49, 50). As shown in Fig. 1,B, we detected low levels of H-2Db in both β2m−/− and in ScDd β2m−/− mice expressing the ScDd transgene. Presence of the ScDd transgene did not perturb the expression of H-2Db in β2m−/− mice (Fig. 1 B).

With biochemical techniques we examined the Sc molecules made in these transgenics. Although we expected that the Sc H-2Dd molecules expressed at the cell surface were the basis of the serological reactivity, it was possible that the Sc molecules were proteolytically cleaved in the spacer region, leading to cell surface H-2Dd2m heterodimers. H-2Dd molecules expressed on the surface of cells of the transgenic mice were exclusively detected as molecules with a molecular mass of 60–65 kDa as determined by immunoprecipitation from splenocytes using 34-2-12 mAb (anti-α3 domain of H-2Dd) (Fig. 1 C). In contrast, the normal MHC-I H-2Dd H chain on the spleen cells from BALB/c was detected as a 50-kDa protein. Thus, the H chain of transgenic ScDd is expressed as a molecular species that is covalently linked with β2m on the cell surface, and there is no evidence for proteolytic cleavage of the Sc molecule. (Because the 34-2-12 Ab used for the immunoprecipitation is α3 domain specific and reacts with the isolated H-2Dd α3 domain (51), this Ab would have detected membrane molecules truncated in the β2m, spacer, α1, or α2 regions.)

It is well known that normal expression of MHC-I molecules in the thymus is necessary for the normal maturation of CD8+ T cells, that is, the progression of immature CD4+CD8+ cells to mature single positive cells. Animals defective in β2m expression, and as a result lacking normal cell-surface MHC-I expression, show a profound decrease in the number of CD8+ (single positive) cells in the thymus as well as in lymph node and spleen (3, 4). In addition, animals defective in TAP expression, and thus deficient in the delivery of self-peptides to MHC-I, show similarly abnormal CD8+ T cell development (52). To assess the selection of CD8+ T cells by ScDd in the absence of normal expression of other MHC-I molecules, we analyzed thymocytes, lymph node cells, and splenocytes from B6, B6 β2m−/−, and ScDd β2m−/− mice for the presence of mature CD8+ T cells by flow cytometry (Fig. 2). Unlike nontransgenic β2m−/− thymocytes, those from ScDd β2m−/− mice contained mature CD8+ cells in numbers similar to those seen in B6 thymuses (Fig. 2). As reported by others, normal B6 animals showed a significant proportion of CD4 and CD8 single positive cells in the thymus (10.65 and 1.70%, respectively), lymph node (62.39 and 36.32% of TCR Cβ+ cells), and spleen (65.39 and 32.29% of TCR Cβ+). In contrast, thymocytes from β2m−/− animals showed a profound decrease in the proportion of CD8 single positive cells in each tissue (0.01, 0.21, and 0.14% in thymus, lymph node, and spleen, respectively) and a compensatory increase in the proportion of CD4 single positive cells in the peripheral lymph nodes and spleen. There was no apparent change in total number of CD4+ cells. Despite the lack of proper expression of MHC-I molecules other than the ScDd in the transgenic β2m−/− animals, substantial numbers of CD8+ T cells were detected in the thymus, lymph nodes, and spleen of ScDd β2m−/− mice (1.70, 33.69, and 31.63% of total, respectively), indicating their normal migration to peripheral lymphoid organs (Fig. 2).

FIGURE 2.

Phenotypic analysis of differentiation of CD8+ T cells in transgenic mice. Thymocytes, lymph node cells, and spleen cells were prepared from β2m−/− and ScDd β2m−/− mice at the age of 6 wk. The expression of CD4 and CD8 was analyzed by two-color fluorescence as described in Materials and Methods. The CD4/CD8 expression in lymph node and spleen cells was analyzed among cells expressing the H57-597 (Cβ) marker. Values for each of the quadrants represent the percentage of the gated cells expressing the indicated markers.

FIGURE 2.

Phenotypic analysis of differentiation of CD8+ T cells in transgenic mice. Thymocytes, lymph node cells, and spleen cells were prepared from β2m−/− and ScDd β2m−/− mice at the age of 6 wk. The expression of CD4 and CD8 was analyzed by two-color fluorescence as described in Materials and Methods. The CD4/CD8 expression in lymph node and spleen cells was analyzed among cells expressing the H57-597 (Cβ) marker. Values for each of the quadrants represent the percentage of the gated cells expressing the indicated markers.

Close modal

We considered the possibility that relatively few CD8+ clones might mature in ScDd β2m−/− mice and expand to fill the available space. If such were the case, the diversity of the expressed TCR would be expected to be low. We examined the Vα and Vβ TCR repertoires of CD8+ T cells using available mAbs. CD8+ T cells in lymph nodes from ScDd β2m−/− exhibited a diverse repertoire of TCR, indicating that these CD8+ T cells do not represent expansion of a small, oligoclonal population (Fig. 3 A). Quantitative differences in the level of expression of different TCR V region genes might reveal subtle differences in the TCR repertoire selected by the ScDd on the β2m−/− background as compared with those repertoires selected by either B6 or B10.D2.

FIGURE 3.

Sc H-2Dd functions normally in thymic selection and in allo- and Ag-specific recognition. A, CD8+ T cells were gated and analyzed for expression of various Vα and Vβ segments of TCRs by flow cytometry. The percentage of CD8+ T cells expressing TCRs were determined in three mice analyzed independently and the mean and SE are plotted. B and C, The indicated cells from primary mixed lymphocyte cultures were used for a cytolysis assay using Con A-stimulated spleen cells from the indicated strains as targets. In B, triangles represent effector cells of ScDdB6 raised against bm3, and circles are for D8 effector cells raised against bm3. Filled circles and triangles are for bm3 targets, and open circles and triangles are for self-targets. D, ScDd β2m−/− (circles) and D8 (squares) mice were immunized by i.p. injection with vaccinia virus expressing gp160IIIB (vPE16) as described in Materials and Methods. Three weeks later, spleen cells from immunized mice were stimulated in vitro with P18-I10 peptide-pulsed irradiated syngeneic spleen cells for 7 days as described. Cytolytic activity of CTL was measured in a 4-h assay using 51Cr-labeled P815 target cells pulsed with P18-I10 (solid markers) for 2 h or without peptide (open markers).

FIGURE 3.

Sc H-2Dd functions normally in thymic selection and in allo- and Ag-specific recognition. A, CD8+ T cells were gated and analyzed for expression of various Vα and Vβ segments of TCRs by flow cytometry. The percentage of CD8+ T cells expressing TCRs were determined in three mice analyzed independently and the mean and SE are plotted. B and C, The indicated cells from primary mixed lymphocyte cultures were used for a cytolysis assay using Con A-stimulated spleen cells from the indicated strains as targets. In B, triangles represent effector cells of ScDdB6 raised against bm3, and circles are for D8 effector cells raised against bm3. Filled circles and triangles are for bm3 targets, and open circles and triangles are for self-targets. D, ScDd β2m−/− (circles) and D8 (squares) mice were immunized by i.p. injection with vaccinia virus expressing gp160IIIB (vPE16) as described in Materials and Methods. Three weeks later, spleen cells from immunized mice were stimulated in vitro with P18-I10 peptide-pulsed irradiated syngeneic spleen cells for 7 days as described. Cytolytic activity of CTL was measured in a 4-h assay using 51Cr-labeled P815 target cells pulsed with P18-I10 (solid markers) for 2 h or without peptide (open markers).

Close modal

We also wished to explore the extent of overlap in the TCR repertoires selected by the ScDd molecules and normal H-2Dd. To do this, we stimulated D8 (B6, native H-2Dd transgenic) splenocytes with ScDd B6 (B6, ScDd transgenic) splenocytes and vice versa in MLR (Fig. 3,B). T cells that develop in an environment in which their TCR are selected by the ScDd molecule but retain discriminating activity against the native two-chain H-2Dd molecule would be expected to be present and be stimulated in the ScDd anti-D8 MLR and should be capable of killing D8 targets. Similarly, T cells expressing TCR that are selected on H-2Dd yet remain reactive against unique epitopes of ScDd would be expected to be stimulated in the D8 anti-ScDd B6 MLR and show cytolysis against ScDd B6 targets. A low level of lysis of target cells of the stimulator mouse strain indicates that few such cells are present and/or they are not particularly reactive. CTL stimulated in both the D8 anti-ScDd B6 MLR and the ScDd B6 anti-D8 MLR exhibited very low cytolytic activity against target cells of the stimulator type (Fig. 3,B). This was not due to a general inability of these mice to respond to any MHC-I difference, as B6.bm3 stimulators elicited substantial activity from the same responders against H-2bm3 target cells (Fig. 3,B). H-2bm1 stimulators also elicited substantial activity from the same responders (data not shown). This also was not due to the inability of H-2Dd or ScDd proteins to stimulate responses generally, as B6 cells transgenic for either the native two chain or the ScDd elicited substantial CTL activity from nontransgenic B6 splenocytes (Fig. 3 C). Furthermore, the anti-H-2Dd alloreactive CTL raised against either the Sc or the native molecule do not effectively distinguish native H-2Dd from the Sc. Taken together, the serological and functional data indicate that ScDd participates in positive selection of CD8+ T cells with a broad repertoire and that this repertoire is functionally similar to that induced by the native H-2Dd.

The generation of a diverse repertoire of CD8+ T cells indicated that the ScDd molecule was effective in the presentation of self-Ag (presumably as peptides) to developing thymic cells. In addition, the ability of cells expressing ScDd to elicit allospecific CTL from B6 T cells indicated that the molecule is effective in Ag presentation to mature T cells. We also wished to evaluate the ability of the ScDd molecules to effectively participate in an immune response by presenting foreign Ags, such as those generated by a viral infection, to specific CD8+ T cells. ScDd β2m−/− and D8 mice were infected with recombinant vaccinia virus directing the expression of the HIV-1 gp160 envelope glycoprotein (vPE16). Three weeks following infection, spleen cells were restimulated in vitro by syngeneic spleen cells loaded with gp160-derived H-2Dd-restricted peptide, P18-I10 (37), and then were tested for effector function in a 51Cr-release assay. CTL induced in ScDd β2m−/− mice could kill P18-I10-loaded P815 cells expressing wild-type H-2Dd on the cell surface, whereas these cells could not lyse P815 target cells that had not received the sensitizing peptide (Fig. 3 D). ScDd β2m−/− mice developed a P18-I10 specific, H-2Dd-restricted CD8+ CTL response that was comparable to that of D8 mice in extent of cytolysis. Furthermore, these cells were able to recognize peptide-loaded native H-2Dd efficiently. Such CD8+ CTL had been generated in vivo with the peptide produced endogenously from the full-length gp160, and these T cells had not been exposed to native H-2Dd-expressing APC before the cytolysis assay. These findings clearly demonstrate that CD8+ T cells selected by peptide in ScDd β2m−/− mice are capable of mounting an effective peptide-specific anti-viral cytotoxic response. Most importantly, CTL induced by ScDd expressed on APC of the transgenic mice interact effectively with native H-2Dd/P18-I10 complexes, that is, they are unable to distinguish native H-2Dd from ScDd.

Native H-2Dd interacts with some of the Ly49 receptors on NK cells, in particular Ly49A (15) and Ly49G2 (16), and plays a critical role in the function and development of NK cells as well as CD8+ T cells (14). To explore the effects of the ScDd transgene in the development of NK cells, we tested whether the presence of ScDd could alter NK cell development in transgenic mice. Expression of native H-2Dd is sufficient to alter the NK cell specificity: H-2Dd, when expressed as a transgene in B6 mice, confers upon the NK cells in these mice the ability to reject B6 bone marrow grafts in vivo and the ability to lyse B6 target cells in vitro (53, 54). Additionally, expression of MHC-I molecules in normal mice is sufficient to permit development of NK cells capable of rejecting β2m−/− bone marrow in vivo and killing β2m−/− target cells in vitro (2, 55).

Surprisingly, we found that expression of the ScDd transgene in β2m−/− mice did not confer any similar function upon the NK cells in these mice. Poly I:C-stimulated NK cells from ScDd β2m−/− mice could not kill β2m−/− Con A blasts as measured in cytotoxicity assays (Fig. 4,A), and ScDd β2m−/− mice accepted β2m−/−-derived bone marrow donor cells (Fig. 4,B). NK cells raised in short-term in vitro cultures supplemented with IL-2 show similar cytotoxicity patterns (data not shown). Thus, NK cells of ScDd β2m−/− mice are defective in either their development and/or their effector function. To determine whether NK cells of ScDd β2m−/− mice were properly educated by ScDd in vivo, we studied expression both of the level and the percentage of Ly49 receptors on NK cells in ScDd β2m−/− mice. The proportion of NK cells expressing receptors of the Ly49 family would be expected to be reduced in the presence of an MHC-I ligand, and each NK cell would also be expected to express a lower level of the given receptor (18, 19, 21, 56). The level of expression and the percentage of NK cells expressing Ly49A, Ly49C/I, and Ly49G2 on NK cells from ScDd β2m−/− mice were indistinguishable from those of NK cells from β2m−/− mice (Table I). This contrasts with the surface expression of NKR of D8 mice that have significantly fewer Ly49A- and Ly49G2-expressing NK cells. These cells also exhibit a lower cell surface density of Ly49A when compared with B6 animals. Similarly, ScDd B6 mice are not significantly different from the B6 parental line. These results further support the conclusion that NK cells of ScDd β2m−/− and of ScDd B6 mice could not be properly educated by ScDd in vivo during their development.

FIGURE 4.

ScDd is defective in promoting the education of NK cells. Fresh poly I:C-stimulated NK cells from B6, ScDd β2m−/−, and β2m−/− mice were generated as described in Materials and Methods and tested on target Con A-activated lymphoblasts from the indicated strains A, Bone marrow grafts were performed in the indicated donor → host combinations as described in Materials and Methods (B and C). ScDd β2m−/− mice do not reject bone marrow cells from β2m−/− (B), nor do ScDd B6 animals reject grafts from B6 (C). Each group of recipients contains five or six mice in the bone marrow transplantation. The results in B and C each show two independent bone marrow transplantation experiments.

FIGURE 4.

ScDd is defective in promoting the education of NK cells. Fresh poly I:C-stimulated NK cells from B6, ScDd β2m−/−, and β2m−/− mice were generated as described in Materials and Methods and tested on target Con A-activated lymphoblasts from the indicated strains A, Bone marrow grafts were performed in the indicated donor → host combinations as described in Materials and Methods (B and C). ScDd β2m−/− mice do not reject bone marrow cells from β2m−/− (B), nor do ScDd B6 animals reject grafts from B6 (C). Each group of recipients contains five or six mice in the bone marrow transplantation. The results in B and C each show two independent bone marrow transplantation experiments.

Close modal
Table I.

Cell-surface expression of NKR in transgenic and knockout mice

NKR
MouseLy49A (A1)Ly49C/I (5E6)Ly49G2 (4D11)
%aMFIb%MFI%MFI
β2m−/− 24.6 ± 2.5 117.3 ± 25.4 65.7 ± 2.8 118.8 ± 42.5 61.6 ± 1.7 131.4 ± 75.9 
ScDd β2m−/− 23.6 ± 0.9 107.0 ± 16.6 64.2 ± 3.9 138.0 ± 29.2 64.5 ± 1.0 126.6 ± 64.9 
C57BL/6 19.5 ± 1.0 100 ± 41.5 43.4 ± 2.2 100 ± 31.1 47.1 ± 2.7 100 ± 46.7 
B6 tgDd (D8) 16.2 ± 1.4 26.5 ± 7.2 39.9 ± 2.9 97.3 ± 34.9 40.5 ± 3.5 75.4 ± 35.8 
ScDd B6 19.9 ± 2.3 65.0 ± 11.7 40.0 ± 3.6 103.4 ± 45.9 56.1 ± 6.6 103.8 ± 34.2 
NKR
MouseLy49A (A1)Ly49C/I (5E6)Ly49G2 (4D11)
%aMFIb%MFI%MFI
β2m−/− 24.6 ± 2.5 117.3 ± 25.4 65.7 ± 2.8 118.8 ± 42.5 61.6 ± 1.7 131.4 ± 75.9 
ScDd β2m−/− 23.6 ± 0.9 107.0 ± 16.6 64.2 ± 3.9 138.0 ± 29.2 64.5 ± 1.0 126.6 ± 64.9 
C57BL/6 19.5 ± 1.0 100 ± 41.5 43.4 ± 2.2 100 ± 31.1 47.1 ± 2.7 100 ± 46.7 
B6 tgDd (D8) 16.2 ± 1.4 26.5 ± 7.2 39.9 ± 2.9 97.3 ± 34.9 40.5 ± 3.5 75.4 ± 35.8 
ScDd B6 19.9 ± 2.3 65.0 ± 11.7 40.0 ± 3.6 103.4 ± 45.9 56.1 ± 6.6 103.8 ± 34.2 
a

The percentage of NK cells expressing the indicated NKR among the NK1.1+ splenocytes is indicated.

b

MFI, Mean fluorescence intensity. The mean ± SD of three different mice of each strain is indicated. MFI values are normalized to C57BL/6 = 100%.

NK cells might require interactions with other molecules whose expression is dependent on β2m−/−, such as MHC-Ib molecules, to be properly educated and active in killing target cells. It is also possible that only a single MHC-I molecule expressed at proper levels is not sufficient to drive NK development. To investigate these possibilities, we employed ScDd B6 mice. These mice are suited to test the above possibilities because they express normal levels of β2m and thus normal levels of all MHC-Ia and MHC-Ib molecules of B6 mice. NK cells from B6 mice transgenic for native H-2Dd (D8) reject B6 bone marrow in vivo and lyse B6 target cells in vitro (53). In addition, cells from D8 mice could kill both B6 and ScDd B6 ConA blasts in the cytotoxicity assay (Fig. 5,A). Furthermore, ScDd B6 hosts could not reject B6 bone marrow grafts (Fig. 4 C). This result is particularly striking because animals expressing a native H-2Dd transgene are capable of rejecting B6 bone marrow grafts. Thus, the ScDd transgene is unable to function like native H-2Dd in the education of NK cells.

FIGURE 5.

Unlike native H-2Dd, ScDd molecules cannot deliver inhibitory or stimulatory signals for NK cell-mediated lysis. IL-2-activated NK cells from D8 mice kill ScDd transgenic target cells and nontransgenic target cells equivalently (A). In bone marrow assays, bone marrow cells from ScDd B6 mice were rejected by B10. D2 mouse recipients (B) and accepted by B6 recipients (C). Each group of recipients contains six mice (or occasionally five) except for one group (n = 3). These results are of two independent bone marrow transplantation experiments. Recipient (B6 + PK136) mice were pretreated with anti-NK1.1 Abs.

FIGURE 5.

Unlike native H-2Dd, ScDd molecules cannot deliver inhibitory or stimulatory signals for NK cell-mediated lysis. IL-2-activated NK cells from D8 mice kill ScDd transgenic target cells and nontransgenic target cells equivalently (A). In bone marrow assays, bone marrow cells from ScDd B6 mice were rejected by B10. D2 mouse recipients (B) and accepted by B6 recipients (C). Each group of recipients contains six mice (or occasionally five) except for one group (n = 3). These results are of two independent bone marrow transplantation experiments. Recipient (B6 + PK136) mice were pretreated with anti-NK1.1 Abs.

Close modal

The MHC-I-dependent resistance to NK lysis is mediated by inhibitory surface NKR that engage target cell MHC-I (57). The native H-2Dd molecule interacts with at least Ly49A and Ly49G2 receptors on NK cells and thus is capable of delivering inhibitory signals to NK cells, preventing or reducing lysis (15, 16, 28, 29). We wished to understand whether the expression of the ScDd transgene-encoded molecules on the target cells resulted in the delivery of inhibitory signals to NK effector cells. In in vitro cytotoxicity assays, short-term cultured NK cells derived from B10.D2 (data not shown) and D8 mice killed ScDd B6 and nontransgenic B6 target cells equivalently (Fig. 5,A) and also lysed ScDd β2m−/− target cells and β2m−/− target cells equivalently. Expression of ScDd on donor bone marrow cells used in grafts was insufficient to prevent rejection of either β2m−/− (data not shown) or B6 bone marrow by B10.D2 hosts (Fig. 5,B). Although it is unclear whether expression of only native H-2Dd on β2m−/− bone marrow grafts would be sufficient to prevent rejection by B10.D2 mice, expression of native H-2Dd as a transgene was sufficient to prevent rejection of B6 marrow by B10.D2 mice (Fig. 5 B), as previously shown by Öhlén and colleagues (53). Taken together, these data indicate that the ScDd transgene, in contrast to the H-2Dd transgene of D8 animals, has little or no function with respect to NK cell inhibition.

The rejection of D8 bone marrow by B6 is dependent on an NK1.1+ cell population, and is believed to be due to engagement of H-2Dd-specific activation receptors on B6 NK cells (53). Although the mechanism of this phenomenon is not completely clear, it may involve recognition by stimulatory receptors that lack cytoplasmic immunoreceptor tyrosine-based inhibitory motifs, such as Ly49D and Ly49H (58, 59). Here we show that B6 mice are unable to reject ScDd B6 bone marrow grafts and confirm the ability of B6 mice to reject D8 grafts (Fig. 5,C). (Because the rejection of bone marrow grafts in this experimental system is due to NK and not CTL activity, allospecific rejection is not observed.) If ScDd B6 bone marrow were capable of eliciting a weak rejection response, one would expect pretreatment of the B6 host with Ab to NK1.1+ cells to reveal a higher level of incorporation of [125I]iododeoxyuridine into the grafted spleens. This was not detected (Fig. 5 C), and the results indicate that the ScDd molecule fails to elicit any stimulatory response from NK cells that may be produced by native H-2Dd.

To explore the mechanism for the functional and developmental deficiency of ScDd-expressing NK cells, we investigated the ability of ScDd to interact directly with Ly49 receptors known to bind native H-2Dd. A soluble biotinylated recombinant form of the Ly49A extracellular domain was employed to probe for binding to lymph node cells expressing various types of MHC-I, and binding was detected by flow cytometry (Fig. 6). Biotinylated Ly49A stained lymph node T cells expressing native H-2Dd (BALB/c, B10.D2, and D8) but did not stain lymph node T cells from B6 or β2m−/− and ScDd β2m−/− mice. This reagent also failed to stain cells from ScDd transgenic mice on either the B6 or β2m−/− background. Thus in a direct assay based on the physical interaction of the recombinant Ly49A with H-2Dd, the ScDd molecule fails to bind Ly49A.

FIGURE 6.

The staining of lymph node cells using biotinylated soluble Ly49A. The lymph node cells from the indicated mouse strains were stained with either biotinylated Ly49A (solid line) or a control biotinylated soluble protein (dashed line), PE-streptavidin, and cychrome-TCR Cβ. These data were analyzed among the TCR Cβ-positive cells of lymph node.

FIGURE 6.

The staining of lymph node cells using biotinylated soluble Ly49A. The lymph node cells from the indicated mouse strains were stained with either biotinylated Ly49A (solid line) or a control biotinylated soluble protein (dashed line), PE-streptavidin, and cychrome-TCR Cβ. These data were analyzed among the TCR Cβ-positive cells of lymph node.

Close modal

Functional assays also suggest that the ScDd protein cannot interact with either Ly49A or Ly49G2. A-LAK cultures of B6 NK cells (15) were preparatively sorted to obtain populations of Ly49A+G2 and Ly49AG2+ NK cells. Neither was able to distinguish B6 target cells from ScDd B6 targets by Ab-dependent cellular cytotoxicity (data not shown). However, both populations were effectively inhibited by D8 target cells due to their known specificity for H-2Dd. Unseparated A-LAK populations were unable to distinguish any of the three target cell populations and lysed them all equivalently (data not shown).

In this paper, we describe transgenic mice expressing a ScDd molecule on somatic cells at densities similar to those of the native H-2Dd molecule in other mouse strains. We have demonstrated that the transgenic ScDd can induce development of a broad spectrum of CD8+ T cells whose TCR repertoire resembles that of the T cells selected by native H-2Dd. In contrast, the expressed product of this transgene is essentially or completely inert to NK cells and their receptors that normally interact with native H-2Dd. The cell surface molecule encoded by the Sc construct did not function like native H-2Dd in any NK assay we employed; it was unable to educate NK cells in the transgenic mice, and was also unable to affect NK cell function in vitro or in vivo or detectably to bind the recombinant Ly49A receptor.

A recent report described the expression of a similar Sc human HLA-A2.1 at normal levels on cells in a transgenic mouse (36). In those experiments, Sc constructs based on either murine or human β2m were expressed in β2m−/− or double β2m−/− H-2Db−/− animals. This transgene also restored a sizable T cell population of functional CD8+ cells when expressed in such MHC-I-deficient mice. As shown in our experiments, transgenic ScDd restored a significant number of CD8+ T cells expressing a broad TCR repertoire in the absence of proper expression of other MHC-Ia and MHC-Ib molecules. Furthermore, these CD8+ T cells could mount a response against specific endogenously processed peptide/MHC complexes in the periphery and efficiently recognized specific peptide presented by native H-2Dd. Although the proportion of CD8+ T cells restored by ScDd and Sc HLA-A2.1 molecules were quite different from each other, the strategy using transgenic Sc molecules was very effective for education of CD8+ T cells in both cases. These results confirm that both the peptide binding site and the region of the MHC-I needed for interaction with TCR are conserved even after the structural modification generating Sc molecules. The similarity of TCR repertoires generated by native and ScDd is demonstrated by the virtual lack of alloreactivity of T cells from D8 for stimulators from ScDd B6 and of T cells stimulated in the reverse MLR (Fig. 3).

It is valuable to think about our experiments along with those using transgenic mice that express MHC-II molecules covalently linked to a single peptide (60, 61) and those that, as a result of the MHC-II processing defect caused by H-2M deficiency, express MHC-II molecules predominantly in complex with the class II-associated invariant chain peptide (62, 63, 64). These experiments suggest that the normal diversity of self-peptides is not critical for positive selection of a broad TCR repertoire. However, the H-2M−/− animals, which express normal levels of MHC-II, mount a response against cells bearing normal MHC-II, indicative of the differences between their repertoires. The single peptide MHC-II transgenic mice show a similar response to parental MHC-II, but disparities in the level of expression make these experiments more difficult to interpret. Thus, it appears that in the MHC-II-restricted examples discussed above, there are significant differences in the full MHC-II/peptide repertoire of the mutant animals as compared with the parental strains, and the resulting TCR repertoires are quite distinct. In our studies, the ScDd functions well to positively select a broad TCR repertoire, a repertoire that shows little reactivity against native H-2Dd. Thus, the ScDd animals select a repertoire very similar to that selected by the native molecule. We conclude that the ScDd functions properly with respect to presentation of most self-peptides presented by native H-2Dd, presumably because the conformation of the peptide groove of ScDd is well conserved and ScDd present a broad array and an appropriate distribution of MHC/peptide complexes to TCR of developing thymocytes. The ability of ScDd to present peptides derived from the endogeneous processing pathway is consistent with our earlier studies examining the presentation of synthetic peptides by transfected cells expressing ScDd (33). In addition, we showed that the sequence motif of peptides eluted from ScDd molecules was the same as that of peptides derived from native H-2Dd (65).

In contrast to its ability to serve both in the education of and the target cell recognition by CD8+ T cells, the transgenic encoded ScDd did not function with respect to the education of NK cells or inhibition of NK-mediated cytoxicity in a variety of assays. ScDd β2m−/−-derived NK cells did not reject β2m−/− bone marrow grafts in vivo, nor did they kill β2m−/− target cells in vitro (Figs. 4 and 5), indicating that the transgenic ScDd protein was unable to alter the MHC-I reactivity of NK cells (i.e., “educate” the NK cells). Consistent with the failure of ScDd to educate NK cells, the pattern of expression of Ly49 receptors on NK cells in ScDd β2m−/− mice was indistinguishable from that of β2m−/− mice (Table I). Furthermore, expression of the ScDd protein on either B6 or β2m−/− bone marrow grafts was unable to prevent their rejection from B10.D2 (H-2d) mice and unable to reverse their sensitivity to cytolysis in vitro. This contrasts strikingly with the function of the native H-2Dd expressed as a transgene in B6 mice, which rescues the graft from bone marrow rejection.

Recently, a ligand on NK cells for nonclassical MHC-I molecules has been identified. A cell-surface heterodimer consisting of NKG2 and CD94 recognizes HLA-E in humans (66, 67) and Qa-1 in mice (68). Furthermore, a significant proportion of murine NK cells express a receptor that binds to soluble Qa-1 tetramers (69). Because ScDd β2m−/− mice cannot properly express MHC-Ib molecules, we examined whether β2m-dependent MHC-Ib molecules may be critical in education of NK cells in vivo. It was also possible that the expression of only one MHC-Ia molecule might not be sufficient to educate functional NK cells. Thus, additional expression of other classical MHC-I molecules may be required to restore NK cell function from the state found in β2m−/− mice. To explore these possibilities further, we performed bone marrow rejection assays and in vitro cytolysis assays using ScDd B6 mice expressing β2m normally on the cell surface. However, ScDd could not restore the education and function of NK cells even in a β2m+ environment. This conclusion is strengthened by the observation that expression of native H-2Dd as a transgene (in D8) is sufficient to induce all of the NK activities that we assayed and failed to find in ScDd B6 mice or cells. These results are all consistent with the view that ScDd molecules cannot interact properly with any H-2Dd-specific inhibitory NKR and thus that the expression of this molecule is simply not sensed effectively by NK cells.

The failure of biotinylated Ly49A protein to stain cells from the ScDd transgenic animals (Fig. 6) as well as the failure of ScDd on targets to inhibit killing by sorted-Ly49A+G2 NK cells (data not shown), indicates that structural alterations of the ScDd protein prevent it from interacting effectively with NK inhibitory receptors. Using chimeric H-2Kd/H-2Dd molecules, Matsumoto et al. showed that residues 53–65 of the α1 domain and 90–107 in the N-terminal part of the α2 domain of H-2Dd contributed to Ly49A recognition (70). It is possible that these differences indirectly affect NKR binding by influencing the conformation or selection of bound peptides. The specificity of Ly49A for different MHC-I molecules may result from polymorphic residues between reactive and nonreactive MHC-I alleles and/or nonpolymorphic residues having different side chain conformations in different MHC-I molecules (14, 71). Other mutagenesis studies using cultured cell lines transfected with in vitro mutated H-2Dd molecules suggest a role of specific residues in the α1 and α2 domains in Ly49A recognition (72).

Unlike the Ly49A/H-2Dd interaction, very little is known about the interaction of H-2Dd with Ly49G2. The expression level of Ly49G2 on NK cells is not perturbed by expression of the ScDd transgene. This result stands in contrast to the observed changes in the expression of the Ly49G2 receptor among NK cells when expressed in the presence of native H-2Dd (19, 21). Although this is not a direct indication, the fact that we fail to observe a change in Ly49G2 expression in the ScDd transgenic mouse suggests that Ly49G2 also fails to effectively interact with the ScDd molecule. Furthermore, Ly49AG2+ A-LAK cells were not inhibited in their cytolysis of ScDd expressing target cells. Finally, Ly49G2 is expressed on approximately half of NK cells in various mouse strains, and an effective inhibitory interaction of the ScDd protein with Ly49G2 could be expected to alter NK function in bulk NK populations, which was not observed. Thus, we conclude that Ly49G2 also cannot interact effectively with the ScDd protein.

Although the mechanism of the failure in physical interaction between Ly49A or Ly49G2 and ScDd molecules is unclear, this may result from: 1) direct blocking of Ly49A binding by the covalent peptide spacer linking the C terminus of β2m and the N terminus of the H-2Dd H chain in this construct; 2) a conformational change of the Ly49A binding site induced by this covalent link; or 3) the formation of an obligate ScDd/ScDd noncovalent dimer on the cell surface due to “domain swapping” of the tethered β2m whereby β2m covalently linked to one molecule binds to the other, resulting in the sequestration of the Ly49A binding site in the interface between the two heterodimers. (Sc Ab Fv are known to form either dimers (“diabodies”) (73, 74) or trimers depending on the length of the peptide spacer that joins VH and VL (75).) Because our data show no evidence of function with respect to the entire population of NK cells of the mouse, we conclude that this lack of interaction must apply to most or all of the H-2Dd-specific inhibitory NKR, including Ly49A and Ly49G2. In addition, we provide evidence that the lack of interaction extends to H-2Dd-specific stimulatory receptors as well (see Fig. 5 C). We thus conclude that some structure that is necessary for interaction with most or all H-2Dd-specific NK activating and inhibitory receptors is not present in the ScDd protein, while the structures required for effective interaction with a broad range of H-2Dd-specific TCR remain intact. Because the site of interaction of MHC-I/peptide complexes with TCR has been shown to consist of the α1 and α2 helices of the MHC and of exposed side chains of the bound peptide, (22, 23, 24, 25) it seems likely that this structural surface is not significantly distorted in the ScDd molecule. Clearly, the general lack of peptide specificity of Ly49A interaction with H-2Dd as well as the profound difference in the reactivity of ScDd with Ly49A both functionally and in a direct binding assay using the recombinant Ly49A indicate that the TCR and the NKR interact with distinct sites on H-2Dd. In addition, these results support the view that several different activating and inhibitory receptors of the same structural family employ a structurally conserved surface of the molecule to interact with H-2Dd.

In conclusion, transgenic Sc H-2Dd can induce development and function of a large number of mature functional CD8+ T cells expressing a broad TCR repertoire highly similar to the repertoire selected by native H-2Dd but cannot function in the education of NK cells or interact with H-2Dd-specific inhibitory receptors expressed by NK cells. Thus, we demonstrate that TCRs bind a part of the MHC/peptide complex distinct from the site where Ly49A and Ly49G2 bind.

We thank the staff of the National Institute of Allergy and Infectious Diseases Transgenic Mouse Facility; also, Howard Adams, Jose Austin, Kim Beck, and Michelle Klein for technical help, Drs. S. Pack and K. Polakova for assistance, Drs. P. Earl and B. Moss for recombinant vaccinia virus vPE16, Dr. E. Shevach for comments on the manuscript, and Dr. R. Germain for helpful discussions during the course of this work.

4

Abbreviations used in this paper: MHC-I, MHC class I; NKR, NK receptor; Sc, single chain; β2m, β2-microglobulin; LAK, lyphokine-activated killer; A-LAK, adherent LAK.

5

Natarajan et al. Submitted for publication.

1
Liao, N. S., M. Bix, M. Zijlstra, R. Jaenisch, D. Raulet.
1991
. MHC class I deficiency: susceptibility to natural killer (NK) cells and impaired NK activity.
Science
253
:
199
2
Höglund, P., C. Ohlen, E. Carbone, L. Franksson, H. G. Ljunggren, A. Latour, B. Koller, K. Kärre.
1991
. Recognition of β2-microglobulin-negative (β2m) T-cell blasts by natural killer cells from normal but not from β2m mice: nonresponsiveness controlled by β2m bone marrow in chimeric mice.
Proc. Natl. Acad. Sci. USA
88
:
10332
3
Zijlstra, M., M. Bix, N. E. Simister, J. M. Loring, D. H. Raulet, R. Jaenisch.
1990
. β2-microglobulin deficient mice lack CD48+ cytolytic T cells.
Nature
344
:
742
4
Koller, B. H., P. Marrack, J. W. Kappler, O. Smithies.
1990
. Normal development of mice deficient in β2m, MHC class I proteins, and CD8+ T cells.
Science
248
:
1227
5
Valiante, N. M., P. Parham.
1996
. NK cells and CTL: opposite sides of the same coin.
Chem. Immunol.
64
:
146
6
Davis, M. M., P. J. Bjorkman.
1988
. T-cell antigen receptor genes and T-cell recognition.
Nature
334
:
395
7
Jorgensen, J. L., P. A. Reay, E. W. Ehrich, M. M. Davis.
1992
. Molecular components of T-cell recognition.
Annu. Rev. Immunol.
10
:
835
8
Ashton-Rickardt, P. G., L. Van Kaer, T. N. Schumacher, H. L. Ploegh, S. Tonegawa.
1993
. Peptide contributes to the specificity of positive selection of CD8+ T cells in the thymus.
Cell
73
:
1041
9
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
10
Salcedo, M., P. Höglund, A. Achour, C. J. Thorpe, H. G. Ljunggren.
1995
. Altered MHC class I presented peptide repertoire is not sufficient to induce NK cell mediated F-1-hybrid resistance.
Mol. Immunol.
32
:
789
11
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
12
Malnati, M. S., M. Peruzzi, K. C. Parker, W. E. Biddison, E. Ciccone, A. Moretta, E. O. Long.
1995
. Peptide specificity in the recognition of MHC class I by natural killer cell clones.
Science
267
:
1016
13
Rajagopalan, S., E. O. Long.
1997
. The direct binding of a p58 killer cell inhibitory receptor to human histocompatibility leukocyte antigen (HLA)-Cw4 exhibits peptide selectivity.
J. Exp. Med.
185
:
1523
14
Lanier, L. L..
1998
. NK cell receptors.
Annu. Rev. Immunol.
16
:
359
15
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
16
Mason, L. H., J. R. Ortaldo, H. A. Young, V. Kumar, M. Bennett, S. K. Anderson.
1995
. Cloning and functional characteristics of murine large granular lymphocyte-1: a member of the Ly-49 gene family (Ly-49G2).
J. Exp. Med.
182
:
293
17
Stoneman, E. R., M. Bennett, J. An, K. A. Chesnut, E. K. Wakeland, J. B. Scheerer, M. J. Siciliano, V. Kumar, P. A. Mathew.
1995
. Cloning and characterization of 5E6(Ly-49C), a receptor molecule expressed on a subset of murine natural killer cells.
J. Exp. Med.
182
:
305
18
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
19
Held, W., J. R. Dorfman, M. F. Wu, D. H. Raulet.
1996
. Major histocompatibility complex class I-dependent skewing of the natural killer cell Ly49 receptor reportoire.
Eur. J. Immunol.
26
:
2286
20
Olsson-Alheim, M. Y., M. Salcedo, H. G. Ljunggren, K. Kärre, C. L. Sentman.
1997
. NK cell receptor calibration: effects of MHC class I induction on killing by Ly49Ahigh and Ly49Alow NK cells.
J. Immunol.
159
:
3189
21
Salcedo, M., A. D. Diehl, M. Y. Olsson-Alheim, J. Sundbäck, L. Van Kaer, K. Kärre, H. G. Ljunggren.
1997
. Altered expression of Ly49 inhibitory receptors on natural killer cells from MHC class I-deficient mice.
J. Immunol.
158
:
3174
22
Garcia, K. C., M. Degano, R. L. Stanfield, A. Brunmark, M. R. Jackson, P. A. Peterson, L. Teyton, I. A. Wilson.
1996
. An αβ T cell receptor structure at 2.5 Å and its orientation in the TCR-MHC complex.
Science
274
:
209
23
Garboczi, D. N., P. Ghosh, U. Utz, Q. R. Fan, W. E. Biddison, D. C. Wiley.
1996
. Structure of the complex between human T-cell receptor, viral peptide and HLA-A2.
Nature
384
:
134
24
Wilson, I. A., K. C. Garcia.
1997
. T-cell receptor structure and TCR complexes.
Curr. Opin. Struct. Biol.
7
:
839
25
Garcia, K. C., M. Degano, L. R. Pease, M. Huang, P. A. Peterson, L. Teyton, I. A. Wilson.
1998
. Structural basis of plasticity in T cell receptor recognition of a self peptide-MHC antigen.
Science
279
:
1166
26
Ding, Y. H., K. J. Smith, D. N. Garboczi, U. Utz, W. E. Biddison, D. C. Wiley.
1998
. Two human T cell receptors bind in a similar diagonal mode to the HLA- A2/Tax peptide complex using different TCR amino acids.
Immunity
8
:
403
27
Yokoyama, W. M., W. E. Seaman.
1993
. The Ly-49 and NKR-P1 gene families encoding lectin-like receptors on natural killer cells: the NK gene complex.
Annu. Rev. Immunol.
11
:
613
28
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
29
Kane, K. P..
1994
. Ly-49 mediates EL4 lymphoma adhesion to isolated class I major histocompatibility complex molecules.
J. Exp. Med.
179
:
1011
30
Brennan, J., D. Mager, W. Jefferies, F. Takei.
1994
. Expression of different members of the Ly-49 gene family defines distinct natural killer cell subsets and cell adhesion properties.
J. Exp. Med.
180
:
2287
31
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
32
Mage, M. G., L. Lee, R. K. Ribaudo, M. Corr, S. Kozlowski, L. McHugh, D. H. Margulies.
1992
. A recombinant, soluble, single-chain class I major histocompatibility complex molecule with biological activity.
Proc. Natl. Acad. Sci. USA
89
:
10658
33
Lee, L., L. McHugh, R. K. Ribaudo, S. Kozlowski, D. H. Margulies, M. G. Mage.
1994
. Functional cell surface expression by a recombinant single-chain class I major histocompatibility complex molecule with a cis-active β2-microglobulin domain.
Eur. J. Immunol.
24
:
2633
34
Lee, L., D. Loftus, E. Appella, D. H. Margulies, M. Mage.
1996
. A recombinant single-chain HLA-A2.1 molecule, with a cis active β2-microglobulin domain, is biologically active in peptide binding and antigen presentation.
Hum. Immunol.
49
:
28
35
Abastado, J. P., D. M. Ojcius, A. Casrouge, P. Yeh, T. N. Schumacher, H. L. Ploegh, P. Kourilsky.
1993
. A soluble, single-chain Kd molecule produced by yeast selects a peptide repertoire indistinguishable from that of cell-surface-associated Kd.
Eur. J. Immunol.
23
:
1776
36
Pascolo, S., N. Bervas, J. M. Ure, A. G. Smith, F. A. Lemonnier, B. Perarnau.
1997
. HLA-A2.1-restricted education and cytolytic activity of CD28+ T lymphocytes from β2 microglobulin (β2m) HLA-A2.1 monochain transgenic H-2Db β2m double knockout mice.
J. Exp. Med.
185
:
2043
37
Takahashi, H., J. Cohen, A. Hosmalin, K. Cease, R. Houghten, J. Cornette, C. DeLisi, B. Mosys, R. Germain, J. Berzofsky.
1988
. An immunodominant epitope of the human immunodeficiency virus envelope glycoprotein gp160 recognized by class I major histocompatibility complex molecule-restricted murine cytotoxic T lymphocytes.
Proc. Natl. Acad. Sci. USA
85
:
3105
38
Shirai, M., C. D. Pendleton, J. A. Berzofsky.
1992
. Broad recognition of cytotoxic T cell epitopes from the HIV-1 envelope protein with multiple class I histocompatibility molecules.
J. Immunol.
148
:
1657
39
Plaksin, D., K. Polakova, M. G. Mage, D. H. Margulies.
1997
. Rigidification of the α2 helix of an MHC class I molecule by a valine to proline mutation in position 165 does not prevent peptide-specific antigen presentation.
J. Immunol.
159
:
4408
40
Evans, G., D. Margulies, B. Shykind, J. Seidman, K. Ozato.
1982
. Exon shuffling: mapping polymorphic determinants on hybrid mouse transplantation antigens.
Nature
300
:
755
41
Margulies, D., G. Evans, K. Ozato, R. Camerini-Otero, K. Tanaka, E. Appella, J. Seidman.
1983
. Expression of H-2Dd and H-2Ld mouse major histocompatibility antigen genes in L cells after DNA-mediated gene transfer.
J. Immunol.
130
:
463
42
Bieberich, C., G. Scangos, K. Tanaka, G. Jay.
1986
. Regulated expression of a murine class I gene in transgenic mice.
Mol. Cell Biol.
6
:
1339
43
Altin, J. G., E. B. Pagler.
1995
. A one-step procedure for biotinylation and chemical cross-linking of lymphocyte surface and intracellular membrane-associated molecules.
Anal. Biochem.
224
:
382
44
Earl, P. L., A. W. Hugin, B. Moss.
1990
. Removal of cryptic poxvirus transcription termination signals from the human immunodeficiency virus type 1 envelope gene enhances expression and immunogenicity of a recombinant vaccinia virus.
J. Virol.
64
:
2448
45
Belyakov, I. M., M. A. Derby, J. D. Ahlers, B. L. Kelsall, P. Earl, B. Moss, W. Strober, J. A. Berzofsky.
1998
. Mucosal immunization with HIV-1 peptide vaccine induces mucosal and systemic cytotoxic T lymphocytes and protective immunity in mice against intrarectal recombinant HIV-vaccinia challenge.
Proc. Natl. Acad. Sci. USA
95
:
1709
46
Raulet, D. H., P. D. Gottlieb, M. J. Bevan.
1980
. Fractionation of lymphocyte populations with monoclonal antibodies specific for LYT-2.2 and LYT-3.1.
J. Immunol.
125
:
1136
47
Chadwick, B. S., S. R. Sambhara, Y. Sasakura, R. G. Miller.
1992
. Effect of class I MHC binding peptides on recognition by natural killer cells.
J. Immunol.
149
:
3150
48
Potter, T. A., C. Boyer, A. M. Verhulst, P. Golstein, T. V. Rajan.
1984
. Expression of H-2Db on the cell surface in the absence of detectable β2 microglobulin.
J. Exp. Med.
160
:
317
49
Potter, T. A., R. A. Zeff, A. M. Schmitt-Verhulst, T. V. Rajan.
1985
. Molecular analysis of an EL4 cell line that expresses H-2Db but not H- 2Kb or β2-microglobulin.
Proc. Natl. Acad. Sci. USA
82
:
2950
50
Bix, M., D. Raulet.
1992
. Functionally conformed free class I heavy chains exist on the surface of β2 microglobulin negative cells.
J. Exp. Med.
176
:
829
51
McCluskey, J., J. Bluestone, J. Coligan, W. Maloy, D. Margulies.
1986
. Serologic and T cell recognition of truncated transplantation antigens encoded by in vitro deleted class I major histocompatibility genes.
J. Immunol.
136
:
1472
52
Van Kaer, L., P. G. Ashton-Rickardt, H. L. Ploegh, S. Tonegawa.
1992
. TAP1 mutant mice are deficient in antigen presentation, surface class I molecules, and CD48+ T cells.
Cell
71
:
1205
53
Öhlén, C., G. Kling, P. Hoglund, M. Hansson, G. Scangos, C. Bieberich, G. Jay, K. Kärre.
1989
. Prevention of allogeneic bone marrow graft rejection by H-2 transgene in donor mice.
Science
246
:
666
54
Hoglund, P., R. Glas, H. G. Ljunggren C. Óhlën, K. Kärre.
1991
. Alteration of the natural killer repertoire in H-2 transgenic mice: specificity of rapid lymphoma cell clearance determined by the H-2 phenotype of the target.
J. Exp. Med.
174
:
327
55
Bix, M., N. S. Liao, M. Zijlstra, J. Loring, R. Jaenisch, D. Raulet.
1991
. Rejection of class I MHC-deficient haemopoietic cells by irradiated MHC-matched mice.
Nature
349
:
329
56
Held, W., D. Cado, D. H. Raulet.
1996
. Transgenic expression of the Ly49A natural killer cell receptor confers class I major histocompatibility complex (MHC)-specific inhibition and prevents bone marrow allograft rejection.
J. Exp. Med.
184
:
2037
57
Ljunggren, H. G., K. Kärre.
1990
. In search of the ‘missing self’: MHC molecules and NK cell recognition.
Immunol. Today
11
:
237
58
George, T. C., L. H. Mason, J. R. Ortaldo, V. Kumar, M. Bennett.
1999
. Positive recognition of MHC class I molecules by the Ly49D receptor of murine NK cells.
J. Immunol.
162
:
2035
59
Nakamura, M. C., P. A. Linnemeyer, E. C. Niemi, L. H. Mason, J. R. Ortaldo, J. C. Ryan, W. E. Seaman.
1999
. Mouse Ly-49D recognizes H-2Dd and activates natural killer cell cytotoxicity.
J. Exp. Med.
189
:
493
60
Ignatowicz, L., J. Kappler, P. Marrack.
1996
. The repertoire of T cells shaped by a single MHC/peptide ligand.
Cell
84
:
521
61
Fukui, Y., T. Ishimoto, M. Utsuyama, T. Gyotoku, T. Koga, K. Nakao, K. Hirokawa, M. Katsuki, T. Sasazuki.
1997
. Positive and negative CD4+ thymocyte selection by a single MHC class II/peptide ligand affected by its expression level in the thymus.
Immunity
6
:
401
62
Fung-Leung, W. P., C. D. Surh, M. Liljedahl, J. Pang, D. Leturcq, P. A. Peterson, S. R. Webb, L. Karlsson.
1996
. Antigen presentation and T cell development in H2-M-deficient mice.
Science
271
:
1278
63
Miyazaki, T., P. Wolf, S. Tourne, C. Waltzinger, A. Dierich, N. Barois, H. Ploegh, C. Benoist, D. Mathis.
1996
. Mice lacking H2-M complexes, enigmatic elements of the MHC class II peptide-loading pathway.
Cell
84
:
531
64
Tourne, S., T. Miyazaki, A. Oxenius, L. Klein, T. Fehr, B. Kyewski, C. Benoist, D. Mathis.
1997
. Selection of a broad repertoire of CD4+ T cells in H-2 Ma0/0 mice.
Immunity
7
:
187
65
Lee, L., L. McHugh, S. Sakuma, D. Loftus, E. Appella, D. H. Margulies, M. Mage.
1995
. Peptide-loaded recombinant single-chain HLA molecules as potential immunogens for T cells. F. Brown, and R. Chanock, and H. S. Ginsburg, and E. Norby, eds.
Vaccines 95
243
Cold Spring Harbor, New York.
66
Braud, V. M., D. S. Allan, C. A. O’Callaghan, K. Soderstrom, A. D’Andrea, G. S. Ogg, S. Lazetic, N. T. Young, J. I. Bell, J. H. Phillips, L. L. Lanier, A. J. McMichael.
1998
. HLA-E binds to natural killer cell receptors CD94/NKG2A, B and C.
Nature
391
:
795
67
Borrego, F., M. Ulbrecht, E. H. Weiss, J. E. Coligan, A. G. Brooks.
1998
. Recognition of human histocompatibility leukocyte antigen (HLA)-E complexed with HLA class I signal sequence-derived peptides by CD94/NKG2 confers protection from natural killer cell-mediated lysis.
J. Exp. Med.
187
:
813
68
Vance, R. E., D. H. Raulet.
1998
. Toward a quantitative analysis of the repertoire of class I MHC- specific inhibitory receptors on natural killer cells.
Curr. Top. Microbiol. Immunol.
230
:
135
69
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
70
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
71
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
72
Waldenström, M., J. Sundbäck, M. Y. Olsson-Alheim, A. Achour, K. Kärre.
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
73
Holliger, P., T. Prospero, G. Winter.
1993
. “Diabodies”: small bivalent and bispecific antibody fragments.
Proc. Natl. Acad. Sci. USA
90
:
6444
74
Poljak, R. J..
1994
. Production and structure of diabodies.
Structure
2
:
1121
75
Kortt, A. A., M. Lah, G. W. Oddie, C. L. Gruen, J. E. Burns, L. A. Pearce, J. L. Atwell, A. J. McCoy, G. J. Howlett, D. W. Metzger, R. G. Webster, P. J. Hudson.
1997
. Single-chain Fv fragments of anti-neuraminidase antibody NC10 containing five- and ten-residue linkers form dimers and with zero- residue linker a trimer.
Protein Eng.
10
:
423