The Complementarity-Determining Region-Like Loops of CD8α Interact Differently with β2-Microglobulin of the Class I Molecules H-2Kb and Thymic Leukemia Antigen, While Similarly with Their α3 Domains

Lesley Devine, Linda Rogozinski, Olga V. Naidenko, Hilde Cheroutre and Paula B. Kavathas
J Immunol April 15, 2002, 168 (8) 3881-3886; DOI: https://doi.org/10.4049/jimmunol.168.8.3881

Abstract

The murine CD8 glycoprotein interacts with both classical MHC class I molecules and some nonclassical molecules, including the thymic leukemia Ag (TL). TL binds preferentially to CD8αα homodimers with a 10-fold higher affinity than H-2Kb class I molecules. To understand the molecular basis for this difference, we created a panel of CD8α mutants and tested the ability of the CD8αα homodimers to bind to H-2Kb tetramers and TL tetramers. Mutations in three CD8 residues located on the complementarity-determining region-like loops contacting the negatively charged loop in the α3 domain of MHC class I greatly reduced binding to both tetramers. Because TL and H-2Kb class I sequences are highly conserved in the α3 domain of MHC class I, this suggests that CD8 contacts the α3 domain of TL and H-2Kb in a similar manner. In contrast, mutations in residues on the A and B β strands of CD8 that are involved in contact with β2-microglobulin affected interaction with the H-2Kb tetramer, but not the TL tetramer. Therefore, the orientation of interaction of TL with CD8 appears to be different from that of H-2Kb. The unique high affinity binding of TL with CD8αα is most likely a result of amino acid differences in the α3 domain between TL and H-2Kb, particularly at positions 198 (K to D) and 228 (M to T), which are contact residues in the CD8αα-H-2Kb cocrystal.

The T cell surface glycoprotein CD8 functions either through a direct interaction with MHC class I or as a coreceptor forming a complex between TCR and MHC class I. CD8 is expressed either as a homodimer (CD8αα) or a heterodimer (CD8αβ), and in humans can also be expressed as a CD8ββ homodimer on transfectants or transgenic animals (1). CD8 signals inside the cell through its association with the tyrosine kinase p56lck (2, 3) and linker of activation in T cells (4). The β-chain facilitates interaction with rafts via palmitoylation of its cytoplasmic domain (5). CD8 also has a number of other ligands, including T lymphocyte-triggering factor (6), gp180 (7), and nonclassical class I MHC molecules (class Ib) such as HLA-G (8) and murine thymic leukemia Ag (TL)4 (9).

Mutational analyses (10, 11, 12, 13, 14) and x-ray crystallography (15, 16) of the CD8αα-MHC class I interaction show that the complementarity-determining region (CDR) loops on CD8αα make contact with a negatively charged loop on the α3 domain of MHC class I, and that CD8 residues on the A and B β-strands of one of the Ig domains (α1 subunit) interact with residues of β2-microglobulin (β2m) and the α2 domain of MHC class I. Although the major contact is with the α3 domain of MHC class I, the interaction is different between murine and human CD8 and MHC class I. Human CD8 has more contacts with the α2 domain of MHC class I and does not contact the AB loop at the base of the α3 domain. Furthermore, murine CD8 has four residues at the N terminus that make additional contacts with β2m. Currently, it is not known whether CD8 interacts with other ligands in a similar manner.

Class Ib MHC molecules are similar in structure to the classical class I molecules; they have α1, α2, and α3 domains with varying homology to MHC class I and are associated with β2m. This study focuses on a comparison of CD8 interaction with the class Ib MHC molecule TL and the classical MHC class I molecule H-2Kb using tetramers of both molecules. TL has a unique pattern of expression. It is most abundantly expressed on the intestinal epithelial cells in all mouse strains (17). It has an overall homology to MHC class I of 70%, with the greatest homology (86%) being in the α3 domain that contains the CD8-binding motif (18). Of the 21 aa in the α3 domain of H-2Kb shown to contact CD8 in the cocrystal (16), TL only varies from H-2Kb at two positions (198 and 228). In addition, TL has been shown to bind to both murine and human CD8 (9).

In contrast to the interaction of CD8 with classical MHC class I molecules, TL was recently shown to bind preferentially to murine CD8αα homodimers in comparison with CD8αβ heterodimers (19). In addition, the affinity for the interaction with CD8αα was 10-fold greater than that of classical MHC class I (19). TL tetramers have been shown to bind to a TL-specific CTL line via a TCR/CD3-dependent mechanism (20), and to intraepithelial lymphocytes and thymocytes via a CD8-dependent/CD3-independent interaction (19).

In this study, we confirm that CD8αα alone is sufficient to bind TL tetramers and compare the ability of H-2Kb and TL tetramers to bind to CD8 with mutations of solvent-exposed residues. Our mutational analysis demonstrates that both H-2Kb and TL make contact with CD8 residues located on the CDR-like loops that contact the negatively charged loop in the α3 domain of MHC class I. An anti-CD8α Ab recognizing an epitope on the α3 domain contact region also blocked the interaction of both proteins. However, mutations in residues located on the side of CD8 (A and B β strands) involved in contact with β2m had a major effect on binding of the H-2Kb tetramer, but not the TL tetramer. Based on these results, we suggest that the main contact between CD8αα and TL is with the α3 domain of TL and the face of CD8 containing the CDR-like loops. However, the orientation is different so that CD8 has either no or little contact with β2m.

Materials and Methods

Mutagenesis

Mutagenesis of solvent-exposed residues on mouse CD8α (Lyt-2) was performed using the GeneEditor Mutagenesis kit (Promega, Madison, WI). Oligonucleotides were designed to change the designated amino acid and to create a restriction site to identify mutants (Table I⇓). Mutagenesis was conducted in the expression vector (pCDL-SRα296), and each construct was sequenced to confirm the presence of the mutation.

View this table:
Table I.

List of CD8α mutants and the oligonucleotides used to make mutations using the GeneEditor mutagenesis kit from Promegaa

Mutants are described using the amino acid single letter code, identifying the original amino acid, residue number, and what amino acid was substituted, e.g., R8A is arginine at position 8 changed to an alanine.

CD8 expression on COS-7 cells

COS-7 fibroblasts were cultured and maintained in RPMI 1640 with 10% FBS (HyClone, Logan, UT). Eighteen to 24 h before transfection, cells were plated out at 2 × 105 cells per 60-mm plate. Liposome-mediated transfection was performed, as previously described (13). For expression of the CD8αα homodimer, 4 μg of wild-type (WT) or mutant CD8α cDNA in 200 μl of Optimem was added to 200 μl of Optimem plus 16 μl of Lipofectamine (both Life Technologies, Grand Island, NY). For experiments requiring expression of the CD8αβ heterodimer, 2 μg of CD8α and 2 μg of CD8β cDNA were used for the transfection mix. Reagent mixtures were kept in the dark at room temperature for 30 min, after which 2 ml of Optimem was added to each tube, followed by plating onto the COS-7 cells. After 5 h, the reagent mix was removed, and 4 ml of culture medium was added to each plate. Cells were given fresh media every 24 h and harvested for phenotypic analysis at 48 or 72 h.

Preparation of tetramers

OVAp or SIYp H-2Kb tetramers were prepared with peptides SIINFEKL or SIYRYYGL, respectively, using the previously described method (21). Briefly, human β2m and a truncated form of H-2Kb H chain, in which the transmembrane and cytosolic domain had been removed and a specific biotinylation site added to the C terminus (22), were expressed in Escherichia coli strain BL21 (DE3) LysS. Inclusion bodies were purified and the proteins were refolded, as described previously (23). Refolding was performed at 10°C in the presence of 25 μg/ml peptide (Research Genetics, Huntsville, AL) and protease inhibitors: pepstatin A (1 μg/ml), leupeptin (1 μg/ml), and PMSF (0.4 mM). Soluble monomeric complexes were purified by gel filtration over a Superdex 200HR column (Amersham Pharmacia Biotech, Piscataway, NJ) and enzymatically biotinylated by overnight incubation with purified biotin protein ligase (BirA) at room temperature with components, as follows: 5 μM HLA-A2/peptide, BirA enzyme (1.5 × 106 U; Avidity, Denver, CO), 80 μM biotin, 10 mM ATP, 10 mM MgOAc, and 20 mM bicine. Unbound biotin was removed by gel filtration, and the purified monomers were tetramerized by incubation with PE-labeled streptavidin (Molecular Probes, Eugene, OR) at a molar ratio of 4:1.

TL tetramers were prepared as described previously (19). Basically, the extracellular domain of TL terminating at the tryptophan residue at the end of the α3 domain was linked to a BirA tag-coding sequence. This was expressed in baculovirus with murine β2m. The protein was biotinylated, and tetramers were made with streptavidin-PE (BD PharMingen, San Diego, CA).

FACS analysis of CD8 expression and tetramer binding

CD8αα expression was determined by staining with mAbs CT-CD8α-FITC (Caltag Laboratories, Burlingame, CA) and 53.6.7-FITC (BD PharMingen) before tetramer-binding assays. In addition, CD8αβ expression was confirmed by staining with CT-CD8β-PE (Caltag Laboratories). For analysis of MHC tetramer binding, 1 μl of H-2Kb tetramer or 0.5 μl of TL tetramer was added to a minimum of 105 cells and incubated for 1 h at 4°C on a rocker platform. Following incubation, cells were washed twice in buffer (PBS with 1% FCS and 0.1 mM sodium azide), then analyzed on a FACScan or FACSCalibur flow cytometer. For Ab inhibition experiments, 10 μl of unconjugated 53.6.7 (or 4 μg of purified Ab), H59 supernatant (S. Jameson, University of Minnesota Medical School, Minneapolis, MN), or 6 μg of CT-CD8α was added to cells for 30 min, washed once, then assayed for tetramer binding, as described.

Results were calculated using both the percentage of positive cells and the mean fluorescence of the positive population relative to the expression level, using the following formulas: (mean fluorescence intensity tetramer)/(mean fluorescence intensity Ab) × (% positive tetramer − vector)/(% positive Ab − vector) = binding index (BI).

The BI of tetramer binding to CD8α WT was taken as 100%, and results were expressed relative to WT: (BI mutant)/(BI WT) × 100 = % binding relative to WT.

Results

Comparison of tetramer binding to CD8

The ability of H-2Kb and TL tetramers (T18d allele) to bind to COS-7 transfectants expressing either CD8α alone or CD8α and β was compared. The best binding was observed with the TL tetramer binding to CD8αα homodimers, with the histogram profile being similar to that of anti-CD8α Ab staining (Fig. 1⇓). In this experiment, the mean fluorescence intensity of the positive population was 1013 for TL binding to CD8αα compared with 633 for H2-Kb tetramers binding to the same cells. Additional comparisons of TL and H2-Kb tetramers binding to WT CD8αα and CD8αβ can be seen in Figs. 6⇓ and 7⇓. The H-2Kb tetramers bound to both CD8αα and CD8αβ transfectants, but the binding was never as good as the binding of TL, consistent with previous findings on the much higher affinity of TL for CD8αα.

FIGURE 1.
FIGURE 1.

Representative flow cytometric analyses of COS-7 cells transfected with vector, CD8α, or CD8α and CD8β cDNAs. Cells were stained with mAbs against mouse CD8α and CD8β, as well as tetramers of H-2Kb and TL. Mean fluorescence intensities of the positive population are shown for tetramer binding to COS-7 cells expressing CD8αα and CD8αβ.

Effect of CD8 Abs on H-2Kb and TL tetramer binding

The ability of Abs against both CD8α and CD8β to block the interaction of tetramers was examined. The two anti-CD8α Abs CT-CD8a and H59.101 bind to regions of CD8 that are involved in contact with MHC class I (α3 domain or β2m, respectively, unpublished epitope-mapping data). They blocked both H-2Kb (Fig. 2⇓A) and TL tetramer binding (Fig. 2⇓B) to COS-7 cells expressing either CD8α alone or CD8α and β. The anti-CD8β-specific Ab, H35.17, only reduced the binding of H-2Kb tetramers, not TLtetramers, to cells transfected with CD8α and CD8β cDNAs, indicating that CD8β does not appear to play a role in TL tetramer binding in contrast to H-2Kb.

FIGURE 2.
FIGURE 2.

Effect of anti-CD8α Abs (CT-CD8a and H59.101) and the CD8β Ab (H35.17) on a, H-2Kb tetramer staining, and b, TL tetramer staining to CD8αα (▪) and CD8αβ transfectants (▨). Cells were pretreated with Abs against CD8α (CT-CD8a or H59.101) or CD8β (H35.17) for 30 min before addition of tetramers. Tetramer staining was assessed using flow cytometry and results expressed relative to WT.

Mutations affecting H-2Kb and TL tetramer binding to CD8

Given that the anti-CD8α Abs that interact with CD8-MHC class I contact residues blocked both H-2Kb and TL tetramer binding and given the strong homology (86%) between H-2Kb and TL in the α3 domain (Fig. 3⇓), we hypothesized that the α3 domain of TL would contact CD8. To test this hypothesis, we created a panel of CD8α mutants and tested the ability of the mutant forms of CD8αα to bind to either the TL or H-2Kb tetramer by flow cytometry (Fig. 4⇓). All mutants shown were expressed equivalent to or better than WT CD8. We found that mutations in residues S31A/L, K62E, and N107A had strong effects on both H-2Kb and TL binding, reducing tetramer binding to 20% or less relative to WT CD8 (Fig. 4⇓). These three residues are located on the CD8 CDR1, 2, and 3 loops, respectively (Fig. 5⇓); they contact the negatively charged loop of the α3 domain of MHC class I based on the murine CD8-MHC class I cocrystal (16). Substituting S31 with leucine, which has a much bulkier side chain, completely knocked out binding of both tetramers (S31L) (Fig. 6⇓). Therefore, it appears that CD8 contacts the α3 domain of TL using similar contact residues as for H-2Kb.

FIGURE 3.
FIGURE 3.

Homology in the α3 domains of H-2Kb and TL. Residues in bold are known contact residues for murine CD8αα and MHC class I. Differences in these residues between H-2Kb and TL are identified with ○.

FIGURE 4.
FIGURE 4.

Effect of mutations in CD8α on binding of TL tetramer (A) or H-2Kb tetramer (B) binding to CD8αα homodimers. Percentage of binding shown is expressed relative to tetramer binding (TL or H-2Kb) to CD8αα WT. The results are an average of at least three experiments, and error bars represent the SE from each group. Known contacts with MHC class I from the cocrystal structure of CD8 with H2-Kb (16 ) are indicated by an asterisk.

FIGURE 5.
FIGURE 5.

Ribbon diagram of the cocrystal of murine CD8αα with MHC class I. The location of CD8 residues that when mutated affected interaction of CD8 with H-2Kb tetramer and/or TL tetramer is indicated. If residues on both CD8 Ig domain subunits do not both make contact, then only the contact residue is indicated, except for K12. The location on both CD8 subunits is indicated. Ribbon diagram was generated using Rasmol with coordinates from the Protein Data Bank (accession no. 1bqh; Ref. 16 ).

FIGURE 6.
FIGURE 6.

Representative flow cytometric analyses of H-2Kb and TL tetramer binding to CD8αα WT and two CD8α mutants. The mutations were of R8, which interacts with β2m, and S31, which makes contact with the negatively charged loop on the α3 domain of MHC class I in the CD8-MHC class I cocrystal. The left panel shows the level of expression of each construct as determined by an Ab against CD8α.

Nevertheless, significant differences in binding of CD8 mutants to TL vs H-2Kb tetramer were also noted. One difference was noted for residues known to make contact with β2m from the CD8αα/H-2Kb cocrystal. For example, L29A had no effect on TL tetramer binding, while binding of the H-2Kb tetramer was reduced to ≈30% (Fig. 4⇑). In addition, R8A completely abrogated H-2Kb tetramer binding, but only reduced TL tetramer binding to 70% of WT (Fig. 5⇑). Making a more drastic change of this arginine to aspartic acid did not dramatically alter the effect of this mutation on TL tetramer binding as compared with the alanine substitution (Fig. 4⇑).

Additionally, mutation of a residue that is not known to be a contact residue with MHC class I also showed differences. Mutant K12E had a dramatic effect on H-2Kb tetramer binding, but had no effect on TL binding (K12E, or K12E,K13E) (Figs. 4⇑ and 7⇓). Mutant K13E alone also caused a small decrease in H2-Kb tetramer binding (60% of WT), but not as dramatic as with K12E. This mutation also had no effect on TL tetramer binding. Staining with several anti-CD8α Abs binding to different regions of the CD8 molecule was unaltered, indicating that no gross conformational changes had occurred that were responsible for the reduction in binding of H-2Kb tetramers. The molecular basis for this effect is not known; however, indirect effects on interaction between CD8 and the α2 domain of MHC class I are possible.

FIGURE 7.
FIGURE 7.

Representative flow cytometric analyses of H-2Kb and TL tetramer binding to CD8αα WT and K12E, K13E mutants. The left panel shows the level of expression of each construct as determined by an Ab against CD8α.

Discussion

The nonclassical MHC class I molecule TL binds CD8αα with a higher affinity than classical MHC class I molecules (19) and demonstrates preferential binding to CD8αα. Our results support this finding in that the TL tetramer bound CD8αα transfectants similarly to an anti-CD8 Ab. In contrast, a smaller number of cells were positive with the H-2Kb tetramer, these presumably being the higher and intermediate CD8-expressing cells.

To understand the molecular basis for this difference, we used a mutational analysis to compare the interaction between CD8 with TL vs MHC class I. We conclude that CD8 binds similarly to the α3 domain of both TL and H-2Kb. This is based on the fact that: 1) mutations in CD8 residues S31, K62, and N107 that contact a negatively charged loop on the α3 domains of H-2Kb eliminate or dramatically reduced the ability of both TL and H-2Kb tetramers to bind to CD8; 2) an anti-CD8α Ab, which binds to that contact region, blocks CD8 binding to both tetramers; and 3) sequence similarity (11 of 12 residues identical between TL and H-2Kb) in the negatively charged loop on the α3 domains that contact CD8 (Fig. 3⇑).

The residues on the side of one of the Ig domain subunits of CD8 containing the A and B β strands make contact with β2m or the α2 domain of MHC class I in both the murine and human CD8 cocrystals. We previously reported that mutations in human CD8 residues L25 and R4 located on those β strands greatly reduced or abolished the interaction of human CD8 with MHC class I (13). This finding was reproduced in this study in that mutations in L29 and R8 (the murine homologous residues) reduced binding of the H-2Kb tetramer to either 30% or 5% of WT. In contrast, those mutations bound TL tetramers 70% or better compared with WT.

The H-2Kb tetramers were constructed with a human β2m because of reported greater stability, while the TL tetramer was constructed with murine β2m. We do not believe that the differences can be attributed to this species difference for the following reason: 1) the contact residues for R8 on β2m (K58, D59) are conserved between human and mouse β2m, and 2) using a cell-cell adhesion assay between COS-7 cells expressing murine CD8 and a cell line expressing H-2Kb proteins associated with murine β2m, the R8 mutation also ablated binding of the H-2Kb-expressing cells (unpublished). In addition, TL molecules refolded with human β2m or mouse β2m bind with identical high affinity to CD8αα when measured using surface plasmon resonance (H. Cheroutre and O. V. Naidenko, unpublished observations).

Another argument for a difference in interaction is our results with mutant K12E either alone or in combination with K13E. This mutation had a dramatic effect on H-2Kb tetramer binding to CD8αα transfectants (reducing binding to ≈10% of WT), but had no effect on TL tetramer binding. K12 is not a known contact with MHC class I and is located on the A β strand of CD8, either facing the α2 domain of MHC class I if it is the α1 subunit of CD8 or out in space if it is the α2 subunit. Because Ab and TL binding to this mutant were not affected, mutation of this solvent-exposed residue presumably has no major effect on the conformation of the protein. It could be indirectly affecting the interaction with the α1α2 domains of MHC class I. Regardless of the molecular basis for this effect, it indicates a major difference in the interaction between H2-Kb and TL.

We also observed differences in the contribution of CD8β to the interaction of tetramers with CD8. Although TL tetramers could bind to COS-7 cells expressing CD8αβ, blocking CD8β with an Ab that reduced H-2Kb tetramer binding had no effect on TL tetramer binding. Therefore, although CD8β is contributing to the binding with H-2Kb, it seems unlikely that CD8β is involved in TL binding. There are two possible reasons that TL binds similarly to CD8αα- and CD8αβ-transfected cells, despite the low binding affinity of TL for CD8αβ (19). Since cotransfecting COS-7 cells with CD8α and CD8β results in some CD8αα homodimer expression, the TL tetramer could be binding to CD8αα alone. Alternatively, the high levels of CD8α expressed as a heterodimer may be sufficient for TL tetramer binding, such that a single CD8α Ig domain would be contacting TL and not the β-chain.

Since soluble TL was shown to bind CD8αα with a 10 times higher affinity than H-2Kb, it was striking that we did not find any unique contact residues on CD8 for TL. In fact, fewer residues on CD8 appeared to be involved in the interaction as compared with H-2Kb. One possible explanation is that some residues may interfere with CD8/MHC class I binding, thus reducing the overall affinity as compared with TL. The two residues in the α3 domain that are different between TL and H-2Kb and are known contact residues are at position 198, a charge substitution (K to D in TL), and at 228, a nonpolar to a polar substitution (M to T in TL). When comparing the affinity of CD8αα for several class I molecules and to HLA-E and HLA-G, Gao et al. (24) found that a single amino acid change in the α3 domain of MHC class I could lead to significant differences in binding affinities. Therefore, one or both of these amino acid changes may significantly contribute to changes in affinity.

In this study, we have shown that MHC class I tetramers, while a valuable tool for monitoring Ag-specific T cell responses, can also be used to study protein-protein interactions.

Acknowledgments

We thank Nick Surh and Usha Soundararajan for technical assistance in creating the CD8 mutants. We also thank Dr. Pamela Bjorkman (California Institute of Technology, Pasadena, CA) and Dr. Michael Hodsdon (Yale University) for their helpful discussions.

Footnotes

  • ↵1 This work was funded by National Institutes of Health Grants CA48115 (to P.B.K.) and DK54451 and AI50263 (to H.C.).

  • ↵2 Current address: Department of Pathology and Immunology, Washington University School of Medicine, St. Louis, MO 63110.

  • ↵3 Address correspondence and reprint requests to Dr. Paula B. Kavathas, Department of Laboratory Medicine, Yale University School of Medicine, 333 Cedar Street, P.O. Box 208035, New Haven, CT 06520-8035. E-mail address: paula.kavathas{at}yale.edu

  • ↵4 Abbreviations used in this paper: TL, thymic leukemia Ag; β2m, β2-microglobulin; CDR, complementarity-determining region; WT, wild type; BI, binding index; BirA, biotin protein ligase.

  • Received November 9, 2001.
  • Accepted January 31, 2002.

References

  1. ↵
    Devine, L., L. J. Kieffer, V. Aitken, P. B. Kavathas. 2000. Human CD8β, but not mouse CD8β, can be expressed in the absence of CD8α as a ββ homodimer. J. Immunol. 164: 833
    OpenUrlAbstract/FREE Full Text
  2. ↵
    Barber, E. K., J. D. Dasgupta, S. F. Schlossman, J. M. Trevillyan, C. E. Rudd. 1989. The CD4 and CD8 antigens are coupled to a protein-tyrosine kinase (p56lck) that phosphorylates the CD3 complex. Proc. Natl. Acad. Sci. USA 86: 3277
    OpenUrlAbstract/FREE Full Text
  3. ↵
    Veillette, A., M. A. Bookman, E. M. Horak, J. B. Bolen. 1988. The CD4 and CD8 T cell surface antigens are associated with the internal membrane tyrosine-protein kinase p56lck. Cell 55: 301
    OpenUrlCrossRefPubMed
  4. ↵
    Bosselut, R., W. Zhang, J. M. Ashe, J. L. Kopacz, L. E. Samelson, A. Singer. 1999. Association of the adaptor molecule LAT with CD4 and CD8 coreceptors identifies a new coreceptor function in T cell receptor signal transduction. J. Exp. Med. 190: 1517
    OpenUrlAbstract/FREE Full Text
  5. ↵
    Arcaro, A., C. Gregoire, N. Boucheron, S. Stotz, E. Palmer, B. Malissen, I. F. Luescher. 2000. Essential role of CD8 palmitoylation in CD8 coreceptor function. J. Immunol. 165: 2068
    OpenUrlAbstract/FREE Full Text
  6. ↵
    Olsson, T., M. Bakhiet, B. Hojeberg, A. Ljungdahl, C. Edlund, G. Andersson, H.-P. Ekre, W.-P. Fung-Leung, T. Mak, H. Wigzell, et al 1993. CD8 is critically involved in lymphocyte activation by a T. brucei-released molecule. Cell 72: 715
    OpenUrlCrossRefPubMed
  7. ↵
    Yio, X. Y., L. Mayer. 1997. Characterization of a gp180-kDa intestinal epithelial cell membrane glycoprotein, gp180. J. Biol. Chem. 272: 12786
    OpenUrlAbstract/FREE Full Text
  8. ↵
    Sanders, S., P. A. Giblin, P. Kavathas. 1991. Cell-cell adhesion mediated by CD8 and human histocompatibility leukocyte antigen G, a nonclassical major histocompatibility complex class I molecule on cytotrophoblasts. J. Exp. Med. 174: 737
    OpenUrlAbstract/FREE Full Text
  9. ↵
    Teitell, M., M. F. Mescher, C. A. Olson, D. R. Littman, M. Kronenberg. 1991. The thymus leukemia antigen binds human and mouse CD8. J. Exp. Med. 174: 1131
    OpenUrlAbstract/FREE Full Text
  10. ↵
    Salter, R. D., R. J. Benjamin, P. K. Wesley, S. E. Buxton, T. P. J. Garrett, C. Clayberger, A. M. Krensky, A. M. Norment, D. R. Littman, P. Parham. 1990. A binding site for the T-cell co-receptor CD8 on the α3 domain of HLA-A2. Nature 345: 41
    OpenUrlCrossRefPubMed
  11. ↵
    Sun, J., D. J. Leahy, P. B. Kavathas. 1995. Interaction between CD8 and MHC class I mediated by multiple contact surfaces that include the α2 and α3 domains of MHC class I. J. Exp. Med. 182: 1275
    OpenUrlAbstract/FREE Full Text
  12. ↵
    Sanders, S. K., R. O. Fox, P. Kavathas. 1991. Mutations in CD8 that affect interactions with HLA class I and monclonal anti-CD8 antibodies. J. Exp. Med. 174: 371
    OpenUrlAbstract/FREE Full Text
  13. ↵
    Giblin, P. A., D. J. Leahy, J. Mennone, P. B. Kavathas. 1994. The role of charge and multiple faces of the CD8α/α homodimer in binding to major histocompatibility complex class I molecules: support for a bivalent model. Proc. Natl. Acad. Sci. USA 91: 1716
    OpenUrlAbstract/FREE Full Text
  14. ↵
    Devine, L., J. Sun, M. R. Barr, P. B. Kavathas. 1999. Orientation of the Ig domains of CD8αβ relative to MHC class I. J. Immunol. 162: 846
    OpenUrlAbstract/FREE Full Text
  15. ↵
    Gao, G. F., J. Tormo, U. C. Gerth, J. R. Wyer, A. J. McMichael, D. I. Stuart, J. I. Bell, E. Y. Jones, B. K. Jakobsen. 1997. Crystal structure of the complex between human CD8αα and HLA-A2. Nature 387: 630
    OpenUrlCrossRefPubMed
  16. ↵
    Kern, P. S., M.-K. Teng, A. Smolyar, J. Liu, R. E. Hussey, R. Spoerl, H.-C. Chang, E. L. Reinherz, J.-H. Wang. 1998. Structural basis of CD8 coreceptor function revealed by crystallographic analysis of a murine CD8αα ectodomain fragment in complex with H-2Kb. Immunity 9: 519
    OpenUrlCrossRefPubMed
  17. ↵
    Hershberg, R., P. Eghtesady, B. Sydora, K. Brorson, H. Cheroutre, R. Modlin, M. Kronenberg. 1990. Expression of the thymus leukemia antigen in mouse intestinal epithelium. Proc. Natl. Acad. Sci. USA 87: 9727
    OpenUrlAbstract/FREE Full Text
  18. ↵
    Teitell, M., H. Cheroutre, C. Panwala, H. Holcombe, P. Eghtesady, M. Kronenberg. 1994. Structure and function of H-2 T (Tla) region class I MHC molecules. Crit. Rev. Immunol. 14: 1
    OpenUrlPubMed
  19. ↵
    Leishman, A. J., O. V. Naidenko, A. Attinger, F. Konig, C. J. Lena, Y. Xiong, H.-C. Chang, E. L. Reinherz, M. Kronenberg, H. Cheroutre. 2001. A unique T cell receptor (TCR)-independent high affinity interaction of CD8αα with the non classical MHC class I molecule, TL, alters TCR-induced responses. Science 294: 1936
    OpenUrlAbstract/FREE Full Text
  20. ↵
    Tsujimura, K., Y. Obata, Y. Matsudaira, S. Ozeki, K. Yoshikawa, S. Saga, T. Takahashi. 2001. The binding of thymus leukemia (tl) antigen tetramers to normal intestinal intraepithelial lymphocytes and thymocytes. J. Immunol. 167: 759
    OpenUrlAbstract/FREE Full Text
  21. ↵
    Altman, J. D., P. A. Moss, P. J. 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. Science 274: 94
    OpenUrlAbstract/FREE Full Text
  22. ↵
    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. Bio/Technology 11: 1138
    OpenUrlCrossRefPubMed
  23. ↵
    Garboczi, D. N., D. T. Hung, D. C. Wiley. 1992. HLA-A2-peptide complexes: refolding and crystallization of molecules expressed in Escherichia coli and complexed with single antigenic peptides. Proc. Natl. Acad. Sci. USA 89: 3429
    OpenUrlAbstract/FREE Full Text
  24. ↵
    Gao, G. F., B. E. Willcox, J. R. Wyer, J. M. Boulter, C. A. O’Callaghan, K. Maenaka, D. I. Stuart, E. Y. Jones, P. A. Van Der Merwe, J. I. Bell, B. K. Jakobsen. 2000. Classical and nonclassical class I major histocompatibility complex molecules exhibit subtle conformational differences that affect binding to CD8αα. J. Biol. Chem. 275: 15232
    OpenUrlAbstract/FREE Full Text
Back to top

In this issue

Print
Download PDF
Article Alerts
Email Article
Citation Tools

Jump to section