Cutting Edge: Allele-Specific and Peptide-Dependent Interactions between KIR3DL1 and HLA-A and HLA-B

This information is current as Hathairat Thananchai, Geraldine Gillespie, Maureen P. of September 29, 2021. Martin, Arman Bashirova, Nobuyo Yawata, Makoto Yawata, Philippa Easterbrook, Daniel W. McVicar, Katsumi Maenaka, Peter Parham, Mary Carrington, Tao Dong and Sarah Rowland-Jones

J Immunol 2007; 178:33-37; ; Downloaded from doi: 10.4049/jimmunol.178.1.33 http://www.jimmunol.org/content/178/1/33

References This article cites 28 articles, 12 of which you can access for free at: http://www.jimmunol.org/ http://www.jimmunol.org/content/178/1/33.full#ref-list-1

Why The JI? Submit online.

• Rapid Reviews! 30 days* from submission to initial decision

• No Triage! Every submission reviewed by practicing scientists by guest on September 29, 2021

• Fast Publication! 4 weeks from acceptance to publication

*average

Subscription Information about subscribing to The Journal of is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2007 by The American Association of Immunologists All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. THE

JOURNAL OF IMMUNOLOGY CUTTING EDGE

Cutting Edge: Allele-Specific and Peptide-Dependent Interactions between KIR3DL1 and HLA-A and HLA-B1,2 † ‡ Hathairat Thananchai,* Geraldine Gillespie,* Maureen P. Martin, Arman Bashirova,ʈ Nobuyo Yawata,§ Makoto Yawata,§ Philippa Easterbrook,¶ Daniel W. McVicar, Katsumi Maenaka,# Peter Parham,§ Mary Carrington,† Tao Dong,3,4* and Sarah Rowland-Jones3,4* ؉ Although it is clear that KIR3DL1 recognizes Bw4 transmembrane comprising two (KIR2D) or three ؉ HLA-B, the role of Bw4 HLA-A allotypes as KIR3DL1 (KIR3D) extracellular Ig-like domains. In general, KIRs with long cytoplasmic tails (KIR2DL and KIR3DL) transduce an in- ligands is controversial. We therefore examined the bind- Downloaded from ing of tetrameric HLA-A and –B complexes, including hibitory signal to inhibit NK cell lysis, whereas KIRs with short ؉ HLA*2402, a common Bw4 HLA-A allotype, to cytoplasmic tails (KIR2DS and KIR3DS) activate NK cell KIR3DL1*001, *005, *007, and *1502 allotypes. Only function (5, 6). An exception is KIR2DL4, the HLA-G recep- ؉ Bw4 tetramers bound KIR3DL1. Three of four HLA- tor, which combines a long cytoplasmic tail with activating A*2402 tetramers bound one or more KIR3DL1 allotypes function (7). and all four KIR3DL1 allotypes bound to one or more KIR2D recognizes HLA-C determinants, whereas KIR3D http://www.jimmunol.org/ HLA-A*2402 tetramers, but with different binding spec- receptors recognize HLA-A and -B alleles (5). Polymorphisms ␣ ificities. Only KIR3DL1*005 bound both HLA-A*2402 in the C terminus of the HLA class I 1 helix (residues 77–83) and HLA-B*5703 tetramers. HLA-A*2402-expressing strongly influence KIR interactions. For HLA-A and -B this is target cells were resistant to lysis by NK cells expressing also the site of the serological Bw4 and Bw6 epitopes and their KIR3DL1*001 or *005. This study shows that HLA- corresponding sequence motifs (8). KIR3DL1 specificity was A*2402 is a ligand for KIR3DL1 and demonstrates how originally defined in cellular assays that examined the ability of ؉ different HLA class I molecules to protect targets from NK cell the binding of KIR3DL1 to Bw4 ligands depends upon lysis. In one such analysis of NK clones Cella et al. (9) found by guest on September 29, 2021 the bound peptide as well as HLA and KIR3DL1 ϩ that both Bw4 HLA-A and -B allotypes were inhibitory; they polymorphism. The Journal of Immunology, 2007, 178: emphasized the strong, but incomplete, correlation with the 33–37. presence of isoleucine 80 (I80). In contrast, the analysis of NK cell clones from other donors by Gumperz et al. (10) correlated ϩ atural killer cells represent a major component of in- inhibition with Bw4 HLA-B allotypes irrespective of the po- ϩ nate immunity; they can kill tumor or virus-infected sition 80 residue and found no interaction with two Bw4 cells without prior sensitization and play an impor- HLA-A allotypes, A*2501 and A*2403. Subsequent compari- N ϩ tant role in the control of virus infections (1, 2). NK lysis is son of five HLA-B27 subtypes (all Bw4 ) showed that the four inhibited when their inhibitory receptors interact with class I subtypes with threonine 80 (B*2701, *2703, *2704, *2705, HLA molecules on target cells. Receptors for the classical and *2706) were strong inhibitors of NK cells, whereas the one HLA-A, -B, and -C molecules are encoded by in the killer subtype with I80 (B*2702) did not inhibit (11). The molecular cell Ig-like (KIR)5 complex (3, 4); like HLA-A, B basis for these functional differences was not determined and and C, the KIR genes are highly polymorphic. KIRs are type I while there has been increasing evidence for KIR3DL1 interactions

*Medical Research Council Human Immunology Unit, Weatherall Institute of Molecular tute, National Institutes of Health under Contract N01-CO-12400. H.T. was funded by Medicine, Oxford, United Kingdom; †Laboratory of Genomic Diversity, Science Appli- the Royal Thai Government. cations International Corporation-Frederick, National Cancer Institute, Frederick, MD 2 The content of this publication does not necessarily reflect the views or policies of the 21702; ‡Johns Hopkins University School of Medicine, Baltimore, MD 21205; §Depart- Department of Health and Human Services, nor does mention of trade names, commercial ment of Structural Biology and Department of Microbiology and Immunology, Stanford products, or organizations imply endorsements by the U.S. government. University School of Medicine, Stanford, CA 94305; ¶Department of HIV/Genitourinary Medicine, The Guy’s, Kings’, and St. Thomas’ School of Medicine, London, United King- 3 T.D. and S.R.-J. contributed equally to this work. dom; ʈLaboratory of Experimental Immunology, National Cancer Institute, Frederick, 4 Address correspondence and reprint requests to Dr. Tao Dong and Dr. Sarah Rowland- MD 21702; and #Division of Structural Biology, Medical Institute of Bioregulation, Jones, Medical Research Council Human Immunology Unit, Weatherall Institute of Mo- Kyushu University, Fukuoka, Japan lecular Medicine, John Radcliffe Hospital, Oxford OX3 9DS, United Kingdom. E-mail Received for publication May 19, 2006. Accepted for publication October 17, 2006. addresses: [email protected] and [email protected] The costs of publication of this article were defrayed in part by the payment of page charges. 5 Abbreviations used in this paper: KIR, killer Ig-like receptor; LTNP, long-term nonpro- This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. gressor; rh, recombinant human. Section 1734 solely to indicate this fact. 1 This work was funded by the Medical Research Council, United Kingdom and in part with federal funds under the Intramural Research Program of the National Cancer Insti-

www.jimmunol.org 34 CUTTING EDGE: SPECIFIC BINDING OF HLA CLASS I MOLECULE TO KIR3DL1

ϩ ϩ with Bw4 HLA-B, the significance of interactions with Bw4 lysis was calculated using the following formula: percentage specific lysis ϭ ϫ Ϫ Ϫ HLA-A (which all have I80) remains uncertain. 100% [(experimental lysis spontaneous lysis)/(maximum lysis sponta- neous lysis)]. A possible cause of the differences observed in the HLA-A and -B specificity of KIR3DL1 is genetic polymorphism of KIR3DL1 genotyping KIR3DL1, which we now know is extensive (12–14) but was Genomic DNA was isolated from PBMCs using the PureGene DNA isolation not appreciated at the time of the earlier studies (9–11). Thus, (GentraSystems). Following PCR, KIR3DL1 was sequenced using the fol- Ј Ј the NK cell clones may well have expressed different forms of lowing primers: exon 3, 5 -TTCTTGGTCCAGAGGGCCGGT-3 (forward) and 5Ј-CTGTGACCATGATCACCAC-3Ј (reverse); exon 4, 5Ј-GAAACCA KIR3DL1, potentially with different HLA class I specificities. CAGAAAACCTTCCC-3Ј (forward) and 5Ј-AGAGAGAAGGTTTCTCA To address this question, we used well-defined peptide-HLA TATG-3Ј (reverse); exon 5, 5Ј-GCCTCTTCTCCTTCCAGGTCC-3Ј (for- class I tetrameric complexes (“tetramers”) to dissect the inter- ward) and 5Ј-CACCTGTGACAGAAACAAG-3Ј (reverse); exons 7–9, 5Ј- AGTGGTCATCATCCTCTTCATC-3Ј (forward) and 5Ј-GTGTACAAGA actions between polymorphic variants of KIR3DL1 and TGG TATCTGTA-3Ј (reverse). Cycle sequencing was performed using the HLA-A and -B. ABI BigDye terminator cycle sequencing ready reaction kit (Applied Biosys- tems) and samples were run on an ABI 3730xl sequencer. Materials and Methods Generation of NK clones Results and Discussion We previously used HLA class I tetramers to show that PBMCs were obtained from laboratory workers and members of a previously KIR3DL2 specificity depends upon both the MHC class I mol- described HIV-1-infected, long-term nonprogressor (LTNP) cohort (ethically ecule, either HLA-A3 or –A11, and a specific bound peptide de- approved by the Research Ethics Committee of King’s College Hospital, Lon- Downloaded from don, U.K.) (15). NK cells were negatively selected using anti-CD3, anti-CD14, rived from the EBV (17). We used a similar strategy to investigate and anti-CD19-coated magnetic beads (Dynal Biotech) and then cocultured the binding of HLA class I ligands to four KIR3DL1 allotypes. with irradiated allogeneic PBMCs in RPMI 1640 medium supplemented with 10% heat-inactivated human AB serum (H10) and recombinant human (rh) KIR3DL1 binds to HLA-A*2402 tetramers IL-2 at 100U/ml for 2 wk. NK cell clones were generated by limiting dilution and cultured in H10 with rhIL-2 at 200U/ml and rhIL-15 at 10 ␮g/ml with Four NK clones from three donors either expressed 3DL1*001 irradiated feeder cells. or 3DL1*005 or lacked 3DL1 (Table I); these were stained with a panel of tetramers containing different antigenic peptides. http://www.jimmunol.org/ KIR3DL1 transfectant cell lines Correct tetramer folding was confirmed by staining well-char- cDNA clones encoding KIR3DL1 alleles were isolated from NK cells as de- acterized specific cytotoxic T cell lines (18, 19). Three of the scribed (16). KIR3DL1*005, *007, and *1502 cDNA were cloned into the pEF6/V5-His-TOPO vector (Invitrogen Life Technologies). The correspond- four NK clones bound to HLA-A*2402 tetramers containing ing plasmids (30 ␮g each) were electroporated into Jurkat cells using a BTX different epitope peptides from HIV-1 nef, HIV-1 gag p17, and square wave electroporator. Transfected cells were selected in 20 ␮g/ml blasti- human cytomegalovirus (Fig. 1A and Table I). The ability to cidin S (Invitrogen Life Technologies) and then dilution cloned. Individual bind A*2402 tetramers correlated with KIR3DL1 expression as clones were screened by flow cytometry using anti-KIR3DL1 mAb (DX9) and maintained in 10 ␮g/ml blasticidin S. defined by staining with the anti-KIR3DL1 mAb, DX9. These data show that A*2402 binds to KIR3DL1; this was confirmed by guest on September 29, 2021 Tetramer staining using a 3DL1*001 transfectant of the mouse Baf3 cell line that ϩ NK clones or KIR3DL1 transfectants cell were stained with a panel of class I bound the same three A*2402 tetramers as the KIR3DL1 NK “tetramers.” Briefly, 2 ϫ 105 cells were washed and resuspended with wash clones. Abrogation of tetramer binding in the presence of DX9 buffer (PBS containing 0.1% BSA and 0.1% azide). Two micrograms of tet- Ab, but not by isotype control Ab, showed that binding was ramer was added followed by incubation at 37°C for 15 min. Cells were washed and fixed with wash buffer. For blocking experiments, cells were incubated on specific for KIR3DL1 (Fig. 1B and data not shown). These re- ice with DX9 or isotype control Ab for 30 min before adding a tetramer. sults demonstrate that the complexes of HLA-A*2402 with sev- eral virus-derived peptides are ligands for KIR3DL1. Cytotoxicity assays The killing activity of NK clones was assessed using 51Cr release assays. Class Interaction of HLA-A*2402 with KIR3DL1 inhibits NK cell lysis I-deficient 221 lymphoblastoid cells were used as target cells, either without To investigate the function of the HLA-A24/KIR3DL1 inter- treatment or following infection with recombinant vaccinia virus containing action, we determined whether A*2402 expression protected A*2402, A*0201, or no insert. A*2402 and A*0201 expressions were compa- ϩ rable as determined by anti-MHC class I Ab. 221 cells expressing A*2402 and target cells from lysis by KIR3DL1 NK clones. A*2402 ex- A*0201 were pulsed with 100 ␮mol of peptides and washed before use in the pression protected 221 cells from lysis by NK cell clones assay. expressing either 3DL1*005 (H002-2) or 3DL1*001 For blocking experiments, effector cells were preincubated with 10 ␮g/ml DX9 or IgG1 isotype control at 4°C for 30 min. The ratio of NK cells to target (LTNP208-43), and lysis was restored in the presence of DX9 cells varied from 5:1 to 10:1, and they were incubated at 37°C for 4 h. Specific mAb (Fig. 2, A–D). In contrast, the expression of HLA-A*0201

Table I. Phenotype and genotype of NK clones and their ability to bind HLA-A*2402 tetramers

Binding of HLA-A*2402 Tetramers

NK Clonea DX31 (KIR3DL2) DX9 KIR3DL1 Genotype A*2402 nef A*2402 p17

LTNP025-50 ϩϪ3DL1 negative ϪϪ LTNP025-115 ϩϩ3DL1*001 ϩϩ LTNP208-16 ϪϪ3DL1 negative ND ND LTNP208-43 Ϫϩ3DL1*001 ϩϩ H002-2 ϩϩ3DL1*005 ϩϩ

a All NK clones were positive for CD94 and negative for ILT2. The Journal of Immunology 35 Downloaded from http://www.jimmunol.org/ FIGURE 1. A and B, HLA-A*2402 tetramers can bind specifically to NK clones and transfectants expressing KIR3DL1*001. A, DX9ϩ NK clone LTNP208-43 was stained with A*2402 tetramers. B, Baf3 cells transfected with 3DL1*001 were stained with an A*2402 tetramer refolded with HIV-1 nef pep- FIGURE 2. The interactions between KIR3DL1*001 and 005 and HLA- tide alone and with the addition of isotype Ig control or DX9 mAb. C and D, A*2402 are peptide specific. A and B, NK clones were stained with a panel of Expression of KIR3DL1 on transfectant cell lines. 3DL1*001-transfected Baf3 HLA-A*2402 tetramers folded with different viral epitope peptides. NK clones (C)or3DL1*005-transfected Jurkat cells (D) were stained with DX9 mAb. LTNP208-43 (3DL1*001) (A) and H002-2 (3DL1*005) (B) could interact Gray shaded histograms show staining of untransfected cells and black lines with an A*2402 tetramer refolded with HIV-1 nef and p17 peptides. C–E, Kill- show staining of 3DL1-transfected cells. E and F, KIR3DL1*005 binds both ing of 221 cells expressing HLA-A*2402 and HLA-A*0201, pulsed or un- by guest on September 29, 2021 HLA-A*2402 and HLA-B*5703 tetramers. 3DL1*005-transfected Jurkat cells pulsed, with HIV peptides by DX9ϩ NK clones LTNP208-43 (C) and H002-2 were stained with HLA class I tetramers in the presence of isotype control Ab (D). E shows representative DX9Ϫ NK clone LTNP208-16. Gray bars show (black line) or DX9 Ab (dashed line). Transfectant cells were stained with killing without DX9 mAb. Filled (black) bars show killing in the presence of A*2402 nef (E) or B*5703 KAF (F). These data are representative of three rep- isotype Ig control. Open bars show killing in the presence of DX9 mAb. The licate experiments. E:T ratio varied from 5:1 to10:1. These data are representative of two replicate experiments. wt, Wild type.

(either in the presence or absence of the HIV-1 p17 peptide) did ϩ not protect 221 target cells from killing by KIR3DL1 NK Ϫ clones (Fig. 2, C and D). The KIR3DL1 NK clone transfected cell lines were comparable (Fig. 1, C and D, and data (LTNP208-16) lysed 221 cells expressing A*2402, further not shown). demonstrating the specificity of the inhibitory interaction (Fig. The four KIR3DL1 allotypes exhibited three different pat- 2E). Additionally, the expression of other receptors on NK terns of tetramer binding (Table II). The 3DL1*001 transfec- clones was determined by cell surface staining with anti- tant cell line bound the A24 p17, A24 nef, and A24 CMV tet- KIR3DL2 (DX31), anti-ILT2, and anti-CD94 (Table I). Al- ramers, but not the A24 D2 tetramer or any HLA-B tetramers. ϩ though some KIR3DL1 NK clones expressed KIR3DL2, it is In contrast, the 3DL1*005 transfectant cell line bound to A24 clear that the inhibition is due solely to interaction between p17, A24 nef (Fig. 1E and data not shown), and the two B57 ϩ HLA-A24 and KIR3DL1, because DX9 NK cell clones are no tetramers (Fig. 1F and data not shown). Only the A24 nef tet- longer inhibited in the presence of the anti-KIR3DL1 Ab. ramer bound to the 3DL1*007 and 3DL1*1502 transfectant Taken together, these data show that KIR3DL1 recognition of cell lines. All of the positive reactions were with HLA allotypes HLA-A*2402 inhibits NK cell effector function. having the Bw4 motif; no binding was detected with the HLA- B7, B8, and B*3501 tetramers with the alternative Bw6 motif. KIR3DL1 allotypes differ in their ability to bind HLA class I tetramers These data show first that KIR3DL1 allotypes can have differ- Previous studies showed that heterogeneous levels of KIR3DL1 ent specificities of HLA binding and second that binding can expression on NK and T cells reflect KIR3DL1 gene polymor- depend upon the bound peptide as well as the HLA class I al- phism (16). In this study we compared the binding of four dif- lotype. Consequently, the potential variability in KIR3DL1 ferent KIR3DL1 allotypes to our panel of HLA class I tetramers recognition of HLA class I is much greater than previously ap- by using Baf3 or Jurkat cell lines transfected with the preciated, which clearly warrants further studies to dissect the 3DL1*001, 3DL1*005, 3DL1*007, and 3DL1*01502 alleles pattern of HLA class I and peptide recognition for all common (Table II). The levels of KIR3DL1 allotype expression on the KIR3DL1 allotypes. 36 CUTTING EDGE: SPECIFIC BINDING OF HLA CLASS I MOLECULE TO KIR3DL1

Table II. HLA class I tetramers refolded with various antigenic peptides were tested for binding to different KIR3DL1 transfected cell lines

Transfectant Cell Lines

Baf3 Jurkat

HLA Tetramer Pathogen Epitope Peptide Sequence 3DL1*001 3DL1*005 3DL1*007 3DL1*01502

A*2402 (Bw4ϩ) A24 p17 HIV p17 (28–36) KYKLKHIVW ϩϩϪ Ϫ A24 nef HIV nef2 (134–141) RYPLTFGW ϩϩϩ ϩ A24 CMV CMV (pp65) QVDPVAALF ϩϪϪ Ϫ A24 D2 Dengue NS3 (D2) INYADRRWCF ϪϪϪ Ϫ

B*5703 (Bw4ϩ) B57 KAF HIV p24 (30–40) KAFSPEVIPMF ϪϩϪ Ϫ B57 A2G HIV p24 (A2G,S4N) KGFNPEVIPMF ϪϩϪ Ϫ

B*0702 (Bw6ϩ) B7 nef HIV nef (128–137) TPGPGVRYPL ϪϪϪ Ϫ B*0801 (Bw6ϩ) B8 EBV EBV (BZLF1) RAKFKQLL ϪϪϪ Ϫ B8 nef Q5 HIV nef 5Q (90–97) FLKEQGGL ϪϪϪ Ϫ

B*3501 (Bw6ϩ) B35 nef HIV B35 nef (75–82) VPLRPMTY ϪϪϪ Ϫ Downloaded from Structural models for the distinctive HLA class I specificity of Concluding remarks KIR3DL1*005 This study examined the direct interaction of soluble tetrameric The tetramer binding studies indicate that 3DL1*005 has a complexes of a defined peptide and HLA class I allotype with broader specificity for HLA class I than 3DL1*001, 007, and ϩ 1502, which includes both Bw4 HLA-A and -B allotypes. An- other distinguishing feature of 3DL1*005 is that it combines http://www.jimmunol.org/ low cell surface expression with a high inhibitory capacity (20). From amino acid sequence comparisons, the structural basis for these functional differences appears to be two substitutions, serine for proline at position 182 in the D1 domain and leucine for tryptophan at position 183 in the D0 domain. Mutagenesis experiments (21) are consistent with a model in which the D1 and D2 domains of KIR3DL1 interact with HLA class I in a similar manner to that seen in the crystallographic structure of by guest on September 29, 2021 the KIR2DL1-HLA-Cw4 complex (22). Mutagenesis has also shown that the substitution of serine for leucine at position 182 decreases cell surface expression of KIR3DL1 (23). Computer modeling of the KIR3DL1 structure (24) indi- cates that residue 283 is located in the hydrophobic core of the interface between the D1 and D2 domains. Consequently, the substitution of leucine for tryptophan at position 283 in 3DL1*005 is likely to change the relative orientation of the D1 and D2 domains. A computer-generated model of 3DL1*005 bound to B*5703 was made from the crystallographic structure of B*5703 bound to the KAF peptide and the structure of the KIR2DL1-HLA-Cw4 complex (Fig. 3A). In this model, the el- bow surface of D1-D2 domains interacts with a region that in- cludes the Bw4 epitope and the C-terminal part of the bound peptide. Two features can be identified that could contribute to the specificity of the 3DL1*005-B*5703 interaction. One is that changes in the orientation of D1 and D2 caused by leucine 283 alters the elbow surface to favor interaction with B*5703. FIGURE 3. Computer models of the interaction of KIR3DL1 with peptide The other is that the unusual central bulge of the 11-mer KAF HLA class I complexes. A, Model of the bulged-out region of the 11-residue peptide when bound to B*5703 (19) favors binding to the AЈB peptide in the B*5703–3DL1 complex. The B*5703 sequence is superimposed loop of the D1 domain of 3DL1*005, providing additional in- onto the structure of the KIR2DL1-HLA-Cw4 complex (19, 22). Yellow, KIR2DL1 (modeling for KIR3DL1); dark blue, B*5703; white, bound pep- teractions that could increase the binding affinity. Further tide. The AЈB loop of the KIR (red dotted circle) is close enough to interact with study is needed to determine the contribution of these two po- the bulged-out region of the 11-residue peptide (green dotted circle) bound to tential mechanisms. Crystal structures for A*2402 (25) and B*5703. B, The three-dimensional structural of A*2402 is compatible with B*5101 (24), a well-characterized ligand for KIR3DL1 (24), KIR interaction. The structure in the Bw4 epitope region of A*2402 is com- are similar in both the C-terminal part of the bound peptide pared with the corresponding region of B*5101, a known KIR3DL1 ligand. ␣ Residues 77–83, comprising the Bw4 motif, are indicated by the dotted circle. and the 1 domain helix (Fig. 3B). Thus, the A*2402 structure The bound peptides of A*2402 and B*5101 are shown in pink and dark blue, is compatible with it being a ligand for KIR3DL1, as we have respectively. Because the structures of the two molecules are sufficiently similar, demonstrated. A*2402 is predicted to bind KIR3DL1 in a manner like that of B*5101. The Journal of Immunology 37 cells expressing four KIR3DL1 allotypes. Three of four HLA- 6. Vilches, C., and P. Parham. 2002. KIR: diverse, rapidly evolving receptors of innate A*2402 tetramers bound to one or more KIR3DL1 allotypes, and adaptive immunity. Annu. Rev. Immunol. 20: 217–251. ϩ 7. Rajagopalan, S., Y. T. Bryceson, S. P. Kuppusamy, D. E. Geraghty, A. van der Meer, indicating that this common Bw4 HLA-A allotype is probably I. Joosten, and E. O. Long. 2006. Activation of NK cells by an endocytosed receptor for soluble HLA-G. PLoS Biol. 4: e9. an important KIR3DL1 ligand. The four KIR3DL1 allotypes 8. Natarajan, K., N. Dimasi, J. Wang, R. A. Mariuzza, and D. H. Margulies. 2002. exhibited three different specificities for the four A*2402 tet- Structure and function of natural killer cell receptors: multiple molecular solutions to ramers, showing that the binding is dependent upon the self, nonself discrimination. Annu. Rev. Immunol. 20: 853–885. 9. Cella, M., A. Longo, G. B. Ferrara, J. L. Strominger, and M. Colonna. 1994. NK3- KIR3DL1 allele and the bound peptide as well as the HLA class specific natural killer cells are selectively inhibited by Bw4-positive HLA alleles with I allotype. isoleucine 80. J. Exp. Med. 180: 1235–1242. 10. Gumperz, J. E., V. Litwin, J. H. Phillips, L. L. Lanier, and P. Parham. 1995. The Bw4 Such molecular heterogeneity implies that KIR3DL1 allo- public epitope of HLA-B molecules confers reactivity with natural killer cell clones types will be differentially inhibited by HLA-A*2402, as has that express NKB1, a putative HLA receptor. J. Exp. Med. 181: 1133–1144. been reported for B*5101, B*2705, and B*5801 (20, 26). Of 11. Luque, I., R. Solana, M. D. Galiani, R. Gonzalez, F. Garcia, J. A. Lopez de Castro, and ϩ J. Pena. 1996. Threonine 80 on HLA-B27 confers protection against lysis by a group the six Bw4 tetramers tested, four bound 3DL1*005, three of natural killer clones. Eur. J. Immunol. 26: 1974–1977. bound 3DL1*001, and only one bound 3DL1*007 and 12. Selvakumar, A., U. Steffens, and B. Dupont. 1997. Polymorphism and domain vari- ability of human killer cell inhibitory receptors. Immunol. Rev. 155: 183–196. 3DL1*1502. If this small sample of pathogen-derived peptides 13. Shilling, H. G., L. A. Guethlein, N. W. Cheng, C. M. Gardiner, R. Rodriguez, ϩ extends to self-peptides, then the proportion of Bw4 HLA D. Tyan, and P. Parham. 2002. Allelic polymorphism synergizes with variable gene content to individualize human KIR genotype. J. Immunol. 168: 2307–2315. class I molecules at the cell surface that actually serve as inhib- 14. Hughes, A. L. 2002. Natural selection and the diversification of vertebrate immune itory ligands will vary with the KIR3DL1 allotype. The number effectors. Immunol. Rev. 190: 161–168. 15. Easterbrook, P. J., T. Rostron, N. Ives, M. Troop, B. G. Gazzard, and of KIR3DL1 allotypes and the differences in their specificity for Downloaded from ϩ S. L. Rowland-Jones. 1999. Chemokine receptor polymorphisms and human immu- Bw4 peptide-HLA class I complexes can readily explain why nodeficiency virus disease progression. J. Infect. Dis. 180: 1096–1105. ϩ KIR3DL1 NK cells from some donors are susceptible to in- 16. Gardiner, C. M., L. A. Guethlein, H. G. Shilling, M. Pando, W. H. Carr, ϩ R. Rajalingam, C. Vilches, and P. Parham. 2001. Different NK cell surface pheno- hibition by Bw4 HLA-A (9) whereas others are not (10). types defined by the DX9 antibody are due to KIR3DL1 gene polymorphism. J. Im- KIR3DS1 is an activating receptor with ligand-binding ex- munol. 166: 2992–3001. 17. Hansasuta, P., T. Dong, H. Thananchai, M. Weekes, C. Willberg, H. Aldemir, tracellular domains similar to those of KIR3DL1. The combi- S. Rowland-Jones, and V. M. Braud. 2004. Recognition of HLA-A3 and HLA-A11 by nation of KIR3DS1 and an HLA-B allotype with I80 was asso- KIR3DL2 is peptide-specific. Eur. J. Immunol. 34: 1673–1679. http://www.jimmunol.org/ 18. Dorrell, L., H. Yang, B. Ondondo, T. Dong, K. di Gleria, A. Suttill, C. Conlon, ciated with delayed progression to AIDS in HIV-1 infection D. Brown, P. Williams, P. Bowness, et al. 2006. Expansion and diversification of vi- (27). Although the activating (KIR2DS1) and inhibitory rus-specific T cells following immunization of human immunodeficiency virus type 1 (KIR2DL1) forms of KIR2D recognize a similar repertoire of (HIV-1)-infected individuals with a recombinant modified vaccinia virus Ankara/ HIV-1 Gag vaccine. J. Virol. 80: 4705–4716. peptides bound to HLA-Cw4 (28), similar studies have not been 19. Stewart-Jones, G. B., G. Gillespie, I. M. Overton, R. Kaul, P. Roche, A. J. McMichael, performed for KIR3DS1. Our results demonstrate that KIR3DL1 S. Rowland-Jones, and E. Y. Jones. 2005. Structures of three HIV-1 HLA-B*5703- peptide complexes and identification of related HLAs potentially associated with long- allotypes interact with HLA-A24 and HLA-B57 bound to HIV-1 term nonprogression. J. Immunol. 175: 2459–2468. peptides in an allele-specific manner. This raises two possibilities: 20. Yawata, M., N. Yawata, M. Draghi, A. M. Little, F. Partheniou, and P. Parham. 2006. Roles for HLA and KIR polymorphisms in natural killer cell repertoire selection and 1) KIR3DS1 has its own unique preferences for HLA-peptide modulation of effector function. J. Exp. Med. 203: 633–645. by guest on September 29, 2021 complexes; and 2) KIR3DL1 polymorphism might also influence 21. Khakoo, S. I., R. Geller, S. Shin, J. A. Jenkins, and P. Parham. 2002. The D0 domain disease progression in HIV-1 infected individuals. of KIR3D acts as a major histocompatibility complex class I binding enhancer. J. Exp. Med. 196: 911–921. 22. Fan, Q. R., E. O. Long, and D. C. Wiley. 2001. Crystal structure of the human natural Acknowledgments killer cell inhibitory receptor KIR2DL1-HLA-Cw4 complex. Nat. Immunol. 2: 452–460. We are very grateful to the donors who gave blood for this study. We also thank 23. Pando, M. J., C. M. Gardiner, M. Gleimer, K. L. McQueen, and P. Parham. 2003. Laura Quigley for providing KIR3DL1 transfected Jurkat cell lines. The made from a common allele of KIR3DL1 (3DL1*004) is poorly expressed at cell surfaces due to substitution at positions 86 in Ig domain 0 and 182 in Ig domain 1. J. Immunol. 171: 6640–6649. Disclosures 24. Maenaka, K., T. Maenaka, H. Tomiyama, M. Takiguchi, D. I. Stuart, and E. Y. Jones. The authors have no financial conflict of interest. 2000. Nonstandard peptide binding revealed by crystal structures of HLA-B*5101 complexed with HIV immunodominant epitopes. J. Immunol. 165: 3260–3267. 25. Cole, D. K., P. J. Rizkallah, F. Gao, N. I. Watson, J. M. Boulter, J. I. Bell, M. Sami, References G. F. Gao, and B. K. Jakobsen. 2006. Crystal structure of HLA-A*2402 complexed 1. Biron, C. A., K. B. Nguyen, G. C. Pien, L. P. Cousens, and T. P. Salazar-Mather. with a telomerase peptide. Eur. J. Immunol. 36: 170–179. 1999. Natural killer cells in antiviral defense: function and regulation by innate cyto- 26. Carr, W. H., M. J. Pando, and P. Parham. 2005. KIR3DL1 polymorphisms that affect kines. Annu. Rev. Immunol. 17: 189–220. NK cell inhibition by HLA-Bw4 ligand. J. Immunol. 175: 5222–5229. 2. Cooper, M. A., T. A. Fehniger, and M. A. Caligiuri. 2001. The biology of human 27. Martin, M. P., X. Gao, J. H. Lee, G. W. Nelson, R. Detels, J. J. Goedert, natural killer-cell subsets. Trends Immunol. 22: 633–640. S. Buchbinder, K. Hoots, D. Vlahov, J. Trowsdale, et al. 2002. Epistatic interaction 3. Lanier, L. L. 1998. NK cell receptors. Annu. Rev. Immunol. 16: 359–393. between KIR3DS1 and HLA-B delays the progression to AIDS. Nat. Genet. 31: 4. Long, E. O., D. N. Burshtyn, W. P. Clark, M. Peruzzi, S. Rajagopalan, S. Rojo, 429–434. N. Wagtmann, and C. C. Winter. 1997. Killer cell inhibitory receptors: diversity, 28. Stewart, C. A., F. Laugier-Anfossi, F. Vely, X. Saulquin, J. Riedmuller, A. Tisserant, specificity, and function. Immunol. Rev. 155: 135–144. L. Gauthier, F. Romagne, G. Ferracci, F. A. Arosa, et al. 2005. Recognition of peptide- 5. Boyington, J. C., and P. D. Sun. 2002. A structural perspective on MHC class I rec- MHC class I complexes by activating killer immunoglobulin-like receptors. Proc. ognition by killer cell immunoglobulin-like receptors. Mol. Immunol. 38: 1007–1021. Natl. Acad. Sci. USA 102: 13224–13229. .