Murine CD160, Ig-Like Receptor on NK Cells and NKT Cells, Recognizes Classical and Nonclassical MHC Class I and Regulates NK Cell Activation This information is current as of October 2, 2021. Motoi Maeda, Carmine Carpenito, Ryan C. Russell, Jyoti Dasanjh, Linnea L. Veinotte, Hideaki Ohta, Takashi Yamamura, Rusung Tan and Fumio Takei J Immunol 2005; 175:4426-4432; ;

doi: 10.4049/jimmunol.175.7.4426 Downloaded from http://www.jimmunol.org/content/175/7/4426

References This article cites 43 articles, 25 of which you can access for free at:

http://www.jimmunol.org/content/175/7/4426.full#ref-list-1 http://www.jimmunol.org/

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 October 2, 2021 • Fast Publication! 4 weeks from acceptance to publication

*average

Subscription Information about subscribing to The Journal of Immunology 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 © 2005 by The American Association of Immunologists All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

Murine CD160, Ig-Like Receptor on NK Cells and NKT Cells, Recognizes Classical and Nonclassical MHC Class I and Regulates NK Cell Activation1

Motoi Maeda,2* Carmine Carpenito,3* Ryan C. Russell,4* Jyoti Dasanjh,* Linnea L. Veinotte,* Hideaki Ohta,5* Takashi Yamamura,† Rusung Tan,‡ and Fumio Takei6*‡

Human and mouse NK cells use different families of receptors to recognize MHC class I (MHC I) on target cells. Although human NK cells express both Ig-like receptors and lectin-like receptors specific for MHC I, all the MHC I-specific receptors identified on mouse NK cells to date are lectin-like receptors, and no Ig-like receptors recognizing MHC I have been identified on mouse NK cells. In this study we report the first MHC I-specific Ig-like receptor on mouse NK cells, namely, murine CD160 (mCD160). The expression of mCD160 is restricted to a subset of NK cells, NK1.1؉ T cells, and activated CD8؉ T cells. The mCD160-Ig fusion binds to rat cell lines transfected with classical and nonclassical mouse MHC I, including CD1d. Furthermore, the level Downloaded from of mCD160 on NK1.1؉ T cells is modulated by MHC I of the host. Overexpression of mCD160 in the mouse NK cell line KY-2 inhibits IFN-␥ production induced by phorbol ester plus ionomycin, whereas it enhances IFN-␥ production induced by NK1.1 cross-linking or incubation with dendritic cells. Cross-linking of mCD160 also inhibits anti-NK1.1-mediated stimulation of KY-2 cells. Anti-mCD160 mAb alone has no effect. Thus, mCD160, the first MHC I-specific Ig-like receptor on mouse NK cells, regulates NK cell activation both positively and negatively, depending on the stimulus. The Journal of Immunology, 2005, 175: 4426–4432. http://www.jimmunol.org/

atural killer cell functions are regulated by MHC class I receptors found on mouse NK cells, including gp49B1 and 2B4, (MHC I)7-specific receptors. Human NK cells express have been shown to interact with MHC I (3, 4). Although paired N two families of MHC I-specific receptors (1): killer Ig- Ig-like receptors belong to the Ig superfamily and recognize MHC like receptors (KIRs) and the lectin like receptors CD94/NKG2. I, they are not expressed on NK cells (5). Recently, KIR-like Two families of lectin-like receptors, namely, Ly49 and CD94/ have been found on mouse X (6, 7). However, the NKG2, have been identified as MHC I-specific receptors on mouse ligand specificity and functions of the protein products of the NK cells (2). Although human KIR and murine Ly49 are struc- mouse KIR-like genes are currently unknown. Many MHC I-spe-

turally divergent, they share the same functions of recognizing cific NK cell receptors are also expressed on subsets of T cells, by guest on October 2, 2021 specific repertoires of classical MHC I. CD94/NKG2 heterodimers including CD1d-restricted NKT cells that recognize ␣-galactosyl b recognize the nonclassical MHC I HLA-E in humans and Qa-1 in ceramide bound to CD1d by invariant TCR (8) and CD1d-inde- mice. Thus, all the MHC I-specific receptors found on mouse NK pendent NK1.1ϩ T cells (9). NK cell receptors expressed on these cells to date belong to the lectin family, and none of the Ig-like T cells regulate their activation (9, 10). Although the importance of Ly49 and CD94/NKG2 in NK and NKT cell functions has been *Terry Fox Laboratory, British Columbia Cancer Agency, Vancouver, British Co- well established, recent reports suggest that mouse NK cells may lumbia, Canada; †Department of Immunology, National Institute of Neuroscience, express additional MHC I-specific receptors as yet to be identified. Tokyo, Japan; and ‡Department of Pathology and Laboratory Medicine, University of British Columbia, Vancouver, British Columbia, Canada For example, it has been reported that CD1d on target cells inhibits Received for publication April 6, 2005. Accepted for publication July 18, 2005. the cytotoxicity of mouse and human NK cells (11–13). However, the receptor for CD1d on NK cells has not been identified. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance CD160 is a GPI-anchored membrane protein containing a single with 18 U.S.C. Section 1734 solely to indicate this fact. Ig domain. Human CD160 has been identified by the mAb BY55 1 This work was supported by a grant from the National Cancer Institute of Canada. (14, 15). It is expressed on cytotoxic cells, including NK cells, M.M. is the recipient of fellowships from the Leukemia Research Fund of Canada and ␥␦ the Michael Smith Foundation for Health Research. CD8 T cells, T cells, and intraepithelial lymphocytes. It seems 2 Current address: Department of Immunology, National Institute of Infectious Dis- to bind several classical MHC I, including HLA-A2, -B7, and ease, Toyama 1-23-1, Shinjuku-ku, Tokyo 162-8640, Japan. -Cw3 and the nonclassical MHC I HLA-E and -G (16). Binding of 3 Current address: Abramson Family Cancer Research Institute, University of Penn- CD160 on human NK cells to HLA-Cw3 on target cells enhances sylvania, BRB II/III, 421 Curie Boulevard, Philadelphia, PA 19104. cytotoxicity, suggesting that it functions as an activating receptor. 4 Current address: Department of Laboratory Medicine and Pathobiology, University Binding of the BY55 mAb also augments the proliferation of dou- of Toronto, 1 King’s College Circle, Medical Sciences Building, Toronto, Ontario, Canada M5S 1A8. ble-negative T cells (17–19). These results suggest that human 5 Current address: Department of Developmental Medicine (Pediatrics), Osaka Uni- CD160 recognizes a wide range of MHC I and acts as a costimu- versity Graduate School of Medicine, 2-2 Yamadaoka, Suita, Osaka 565-0871, Japan. latory receptor on NK cells. Although murine CD160 (mCD160) 6 Address correspondence and reprint requests to Dr. Fumio Takei, Terry Fox Lab- has been cloned, its expression pattern and function are unknown. oratory, British Columbia Cancer Research Center, 675 West 10th Avenue, Vancou- In this study we have characterized mCD160. It is expressed on ver, British Columbia, Canada V5Z 1L3. E-mail address: [email protected] subsets of NK cells, NKT cells, and activated CD8 T cells. It binds 7 Abbreviations used in this paper: MHC I, MHC class I; BM, bone marrow; DC, ␤ ␤ to a broad range of MHC I, including CD1d, and appears to reg- dendritic cell; KIR, killer Ig-like receptor; 2m, 2-microglobulin; m, murine; MIG, MSCV IRES GFP; PIPLC, phosphatidylinositol-specific phospholipase C. ulate the cytokine production of NK cells. To our knowledge,

Copyright © 2005 by The American Association of Immunologists, Inc. 0022-1767/05/$02.00 The Journal of Immunology 4427 mCD160 is the first MHC I-specific Ig-like receptor found on Phosphatidylinositol-specific phospholipase C (PIPLC) treatment mouse NK cells or NKT cells. Cells were washed three times with PBS, incubated with PIPLC (5 U/ml; Invitrogen Life Technologies) for 1 h at 37°C, and washed again. Control Materials and Methods cells were incubated without PIPLC. Cells were then stained with mAbs Animals and were analyzed by flow cytometry. ␤ ␤ C57BL/6 (B6), C57BL/10 (B10), B10D2, 2-microglobulin ( 2m)-defi- Immunoprecipitation cient mice of B6 background, BALB/c, CD1d-deficient mice of BALB/c background, and C3H/HeJ mice were purchased from The Jackson Labo- The mCD160-transfected K562 cells were incubated with purified ratory and bred in our animal facility. Fisher344 rats were purchased from CNX46-3 mAb and lysed with lysis buffer (1% Triton X-100, 1% BSA, Taconic Farms. Male mice (6–12 wk old) were used in this study. The use 150 mM NaCl, and 0.1% NaN3 in 10 mM Tris-HCl, pH 7.5). The cell of animals for this study was approved by the animal care committee of lysates were centrifuged at 14,000 rpm for 15 min at 4°C, and the cleared University of British Columbia, and animals were maintained in accor- cell lysates were incubated overnight at 4°C with protein G beads (Amer- dance with the guidelines of the Canadian Council on Animal Care. sham Biosciences). The beads were then washed four times with lysis buffer (without BSA and NaN3), and bound were eluted by boiling RT-PCR, protein expression, and purification for 5 min in 100 ␮lof2ϫ SDS sample buffer (reducing). The eluted proteins were separated by SDS-PAGE and transferred onto nitrocellulose. RNA was isolated from spleen cells using an RNA extraction (Qiagen). After incubation with biotinylated CNX46-3 mAb and peroxidase-conju- RNA was reverse transcribed using an RT kit (Qiagen). For the isolation of gated streptavidin, proteins on the blots were detected by the ECL (Am- full-length mCD160 cDNA, PCR was performed with an initial denatur- ersham Biosciences) method as previously described (26). ation at 94°C, followed by 40 cycles of 94°C for 30 s, 65°C for 30 s, and 68°C for 60 s, with a final 7-min extension at 68°C using the primers: N-glycosidase F treatment 5Ј-TGGAATTCTCAACATTTCCGTGAAATTCCTGGG-3Ј and 5Ј-TG AGATCTCTCGCTGTACCTTCCTCTGAAG-3Ј (EcoRI and BglII restric- After immunoprecipitation as described above, mCD160 cells bound to Downloaded from tion enzyme sites used for cloning are underlined) and HIFI Taq PCR protein G beads were washed and resuspended in PBS, treated with 500 U enzyme (Invitrogen Life Technologies). The PCR product was sequenced. of peptide:N-glycosidase F (New England Biolabs) for6hat37°C, eluted A cDNA fragment corresponding to the extracellular domain of mCD160 with 100 ␮lof2ϫ SDS sample buffer, and analyzed as described above. was amplified by PCR using mCD160 cDNA as a template and the primers 5Ј-TGGAATTCTCAACATTTCCGTGAAATTCCTGGG-3Јand5Ј-TGAG Transfection Ј ATCTACTTACCTGTGCTGAAGTCAGGGTGTGACGTTT-3 (EcoRI Full-length mCD160 was subcloned into the pBCMGSNeo (27) and and BglII restriction enzyme sites are underlined). This PCR was per- pCEP4 (Invitrogen Life Technologies). The pH␤AprmCD1.1neo was pro- http://www.jimmunol.org/ formed with an initial denaturation at 94°C, followed by 30 cycles of 94°C vided by Dr. M. Kronenberg (La Jolla Institute for Allergy and Immunol- for 45 s, 55°C for 30 s, and 68°C for 60 s, with a final 7-min extension at ogy, San Diego, CA) (28). Cell lines were electroporated with 10 ␮gof 68°C. The PCR product was digested with EcoRI and BamHI and cloned plasmid DNA by Pulser (Bio-Rad). Transfectants were selected un- into the pIG vector (20). COS cells were transiently transfected with the der optimal concentrations of G418 or hygromycin (Invitrogen Life Tech- plasmid using FuGene 6 (Roche) according to the manufacturer’s proto- nologies), and cells expressing the transfected genes were isolated by cell cols, and mCD160-Fc in the culture supernatants was purified by affinity sorting. For KY-2, the MSCV IRES GFP (MIG) retrovirus vector (29), chromatography using a protein A column (Amersham Biosciences). In provided by Dr. R. K. Humphries (Terry Fox Laboratory, British Columbia some experiments mCD160-Fc was biotinylated using EZ-Link Sulfo- Cancer Agency, Vancouver, British Columbia, Canada), was used. Full- NHS-LC-Biotin (Pierce). A fusion protein consisting of the extracellular length mCD160 cDNA was subcloned into the MIG vector, and retrovi- domain of H60 and human IgG (H60-Fc) was generated in the same man-

ruses were generated as previously described (29). KY-2 was cocultured by guest on October 2, 2021 ner and was used as a control. with an irradiated mCD160MIG virus-producing cell line in the presence Generation of anti-mCD160 mAb of 200 U/ml IL-2 (PeproTech). After 2 days of infection, nonadherent cells were harvested, and GFPhigh cells were isolated by cell sorting. Fisher 344 rats were immunized by i.p. injection of 50–100 ␮gof mCD160-Fc emulsified in IFA (Sigma-Aldrich) three times at 4-wk inter- Cell culture and cytokine ELISA vals. After a final i.v. injection of mCD160-Fc, immune spleen cells were ϩ The cell culture methods for IL-2-stimulated NK and NK1.1 T cells and fused with the mouse myeloma line SP2/0-Ag14 using the standard pro- splenocytes stimulated with immobilized CD3 mAb were described pre- tocol. Hybridoma culture supernatants were screened by ELISA using viously (9). Bone marrow (BM)-derived dendritic cells (BM-DC) were mCD160-Fc-coated plates. Positive hybridomas were subcloned three cultured as previously described (30). Retrovirus-transduced KY-2 lines times. This resulted in one clone producing anti-mCD160 mAb, termed (1 ϫ 105 cells/wells) were cocultured with BM-DC (1 ϫ 105 cells/wells) CNX46-3 (rat IgG2b, ␬). or stimulated with PMA (50 ng/ml) and ionomycin (1 ␮M; Sigma-Aldrich) Cells, reagents, and flow cytometry in 96-well, round-bottom plates for 24 h at 37°C. In some experiments, retrovirus-transduced KY-2 lines (2 ϫ 105 cells/well) were incubated with COS-7, K562, RBL-1, Sp2/0-Ag14, M1/42 (anti-pan H-2), 2.4G2 (anti- combinations of mAbs (50 ␮g/ml) immobilized in 96-well, flat-bottom FcR␥), and HO-13-14 (anti-Thy1.2) hybridomas were obtained from plates for 24 h as previously described (31). The amounts of IFN-␥ in the American Type Culture Collection. RBL-1 cells transfected with mouse supernatants were measured using Quantikine kits (R&D Systems). MHC I have been described previously (21). The murine T cell lines RMA, EL4, and RMA-S have been described previously (9). The murine NK cell Results clone, KY-2 (22), was a gift from Dr. W. Yokoyama (Howard Hughes Medical Institute, Rheumatology Division, Washington University School Analysis of mCD160 cDNA and protein of Medicine, St. Louis, MO). Anti-murine MHC-class I mAbs have been The mCD160 cDNA clones were isolated by RT-PCR from described previously (21). FITC-conjugated anti-CD45.2 mAb AL14A2, anti-Mac-1 mAb M1/70, and anti-Gr-1 mAb RB6-8C5 have also been de- splenocytes of B6, BALB/c, and C3H/HeJ mice and sequenced. As scribed (23). All other mAbs used in this study were purchased from BD reported, the mCD160 cDNA sequence predicts a polypeptide of Biosciences. 185 aa with the characteristic signal sequence at the N-terminal, a Cell surface staining, flow cytometric analysis, and cell sorting have GPI membrane anchor signal sequence at the C-terminal, single Ig been described previously (10). For analysis of mCD160-Fc binding, cells domain, and three potential sites for N-linked glycosylation. The were first incubated with the 2.4G2 mAb or human ␥ globulin (Sigma- Aldrich) to block FcRs, then incubated with 40 ␮g/ml biotinylated mature protein is predicted to be a GPI-anchored polypeptide with mCD160-Fc or H60-Fc in HBSS with 2% FBS and 0.09% (w/v) sodium approximate Mr of 14,900. A shorter splice variant of mCD160 azide at 37°C for 60 min. After washing, the cells were stained with cDNA, lacking exon 4 that encodes the GPI anchor motif (Fig. 1A) streptavidin-PE on ice for 30 min and analyzed by flow cytometry as de- and potentially encoding a secreted mCD160, was also isolated. A scribed above. In some experiments CD1d-transfected RBL-1 cells were pulsed with ␣-galactosylceramide for2hat37°C as previously described comparison of our mCD160 cDNA sequences with those in Gen- b Bank (an original report in which mouse strain is not indicated: (24). PE-conjugated K -OVA254–267 tetramer was synthesized and used for staining as previously described (25). accession no. AF060982; FVB/N: accession no. NM018767; B6: 4428 MURINE CD160 Downloaded from

FIGURE 1. Analysis of mCD160 cDNA and protein. A, The mCD160 mRNA was detected by RT-PCR. RNA was isolated from spleen cells from B6 (lane 1) and C3H/HeJ (lane 2) mice and subjected to RT-PCR using specific primers, and PCR products were analyzed by agarose gel electrophoresis. Lane 3, DNA size standard. B, K562 cells transfected with empty pCEP4 vector (control/K562) or with mCD160 cDNA in pCEP4 (mCD160/K562) were stained with either biotinylated CNX46-3 mAb (shaded histograms) or biotinylated isotype-matched control mAb (open histograms), followed by streptavidin-PE, and were analyzed by flow cytometry. C, The mCD160 was immunoprecipitated from mCD160/K562 with CNX46-3 mAb and protein G-coupled beads. http://www.jimmunol.org/ The beads were incubated with PBS (lane 1)orN-glycosidase F (lane 2), and the immunoprecipitated proteins were eluted from the beads. They were separated by 10% SDS-PAGE, blotted, probed with biotinylated CNX46-3 mAb plus HRP-conjugated streptavidin, and visualized by chemiluminescence. D, The mCD160-transfected EL4 cells were either treated with PIPLC or left untreated, then stained with anti-CD45 mAb, anti-Thy1, and anti-mCD160 mAb (CNX46-3; shaded histograms) or isotype-matched control (open histograms) plus FITC-conjugated secondary Abs, and analyzed by flow cytometry. accession no. AK042093) identified two single nucleotide poly- NK1.1ϩCD3ϩ cells and NK1.1ϩTCR␣␤ϩ cells (74.4 Ϯ 3.7%; not morphisms that result in amino acid changes. In B6 and BALB/c shown), whereas other spleen cells were mostly negative (Fig. 2). mice, mCD160 has an asparagine in position 76, whereas FVB/N Murine CD160 was also undetectable on Mac1ϩ cells and Gr1ϩ by guest on October 2, 2021 and C3H/HeJ mice have a serine residue. The originally reported cells in BM. The proportion of mCD160-positive NK cells in- mCD160 sequence (15) has a proline residue in position 45, creased after culture with high dose IL-2 (1000 U/ml; 20.4 Ϯ 4.7 whereas other strains (FVB, B6, BALB/c, and C3H/HeJ) have a to 36 Ϯ 2.4%; n ϭ 4). In contrast, the proportion of mCD160-positive leucine residue in its place. NK1.1ϩ T cells declined after stimulation with 200 U/ml IL-2 We generated a fusion protein, mCD160-Fc, that consists of the (79.1 Ϯ 5.4 to 40.1 Ϯ 10%; n ϭ 4). Stimulation of T cells with extracellular domain of mCD160 derived from BALB/c mice and immobilized anti-CD3 mAb induced the expression of mCD160 on the Fc portion of human IgG1 (see Materials and Methods). By Ͼ80% of CD8 T cells, but not on CD4 T cells (Fig. 2). immunizing rats with purified mCD160-Fc, we generated a novel mAb, CNX46-3, that specifically recognizes mCD160 on the hu- Ligand specificity of mCD160 man erythroleukemia cell line K562 transfected with mCD160 Binding of mCD160-Fc to the murine T cell lymphoma line RMA (Fig. 1B). SDS-PAGE and Western blot analysis of mCD160 im- and the TAP-deficient variant line RMA-S was compared. The munoprecipitated from mCD160-transfected K562 identified an fusion protein bound to RMA cells that express a high level of ϳ20-kDa band and a faint ϳ40-kDa band under reducing condi- MHC I, whereas it bound poorly to RMA-S cells, which express a tions (Fig. 1C, lane 1). After N-glycosidase F treatment, an ϳ15- much lower level of MHC I (Fig. 3A). This suggested that kDa band was identified, as expected for nonglycosylated mCD160 might bind to mouse MHC I. Therefore, we next tested mCD160 (Fig. 1C, lane 2). To confirm that mCD160 is GPI an- the binding of mCD160-Fc to a panel of rat basophilic leukemia chored, mCD160-transfected EL4 cells were incubated with phos- RBL-1 cell lines transfected with individual murine MHC I. phatidyl inositol-specific phospholipase C, and mCD160 expres- mCD160-Fc bound to both classical and nonclassical MHC I, in- sion on the cell surface was analyzed by flow cytometry. The cluding Dd,Db,Kb, Qa-1b, and CD1d to varying degrees (Fig. 3B). enzyme treatment significantly reduced the level of mCD160 as The interaction between mCD160 and MHC I was also tested by well as the control GPI-anchored protein, CD90 (Thy1), but not binding tetrameric MHC I H-2Kb folded with OVA-derived pep- that of CD45, which is not GPI anchored (Fig. 1D). These results tide to the human erythroleukemia cell line K562 transfected with show that mCD160 is a GPI-anchored glycoprotein and seems to mCD160. The tetramer very weakly bound to mCD160-transfected exist on the cell surface as a homodimer of ϳ20-kDa chains. K562 cells, but not to control K562 cells transfected with empty vector (Fig. 3C). Distribution of mCD160 on leukocytes The anti-mCD160 mAb CNX46-3 stained NK1.1ϩCD3ϩ cells, Modulation of mCD160 expression by MHC I in vivo NK1.1ϩTCR␣␤ϩ cells (40.9 Ϯ 7%; not shown), and ϳ5% of The level of Ly49 expression on NK cells is known to be cali- double-negative cells in thymus, but not other thymocytes (Fig. 2). brated by the host MHC I (32). To test whether mCD160 is sim- In spleen, the mAb stained ϳ20% of NK cells and ϳ80% of both ilarly modulated in vivo, the levels of mCD160 expression on The Journal of Immunology 4429

FIGURE 2. Distributions of mCD160 on leukocytes. Thymocytes, spleen cells, and BM cells were stained with various combinations of mAbs and an- alyzed by multicolor flow cytometry for the expression of mCD160. The proportions of mCD160-positive cells among NK1.1ϩCD3ϩ cells, CD8 sin- gle-positive T cells (CD8), CD4 single- positive T cells (CD4), CD4ϪCD8Ϫ T cells (DN), CD4ϩCD8ϩ cells (DP), NK1.1ϩCD3Ϫ cells (NK), CD19ϩ cells (B), Mac-1ϩ cells, and Gr-1ϩ cells are shown. For activated T cells, spleen cells were stimulated with immobilized anti-CD3⑀ mAb for 48 h. Spleen cells were also cultured with 2000 U/ml (for NK cells) or 200 U/ml (for NK1.1ϩ T cells) IL-2 for 8 days; stained for NK1.1, CD3⑀, and mCD160; and analyzed. Downloaded from Shaded histograms show staining with anti-mCD160 mAb (CNX46-3); open histograms show staining with isotype- matched control mAb. The percentages are the mean Ϯ SD of at least four individual experiments. http://www.jimmunol.org/

ϩ ␤ NK1.1 T cells in wild-type and 2m-deficient B6 mice were ilarly, upon coculture with BM-DC cells, mCD160/KY-2 cells se- compared. The level of mCD160 expression in wild-type mice was creted more IFN-␥ than MIG/KY-2 cells (Fig. 5C). It should be noted ␤ ␥ slightly lower than that in 2m-deficient mice (Fig. 4A). The level that these cell lines produced the same amount of IFN- without stim- of mCD160 expression on NK1.1ϩ T cells also differed between ulation. IFN-␥ was not detected in the culture of DC alone. Thus, MHC-congenic B10 (H-2b) and B10.D2 (H-2d) mice (Fig. 4A), and overexpression of mCD160 seems to have both costimulatory and the level of mCD160 on NK cells differed between wild-type and inhibitory effects on KY-2 cells, depending on the stimulus. ␤ by guest on October 2, 2021 2m-deficient mice as well as between B10 and B10.D2 mice (Fig. To further investigate the role of mCD160 in NK cell activation, 4B). Although these differences were small, they were statistically the effects of mCD160 cross-linking on NK1.1-mediated stimula- significant (Fig. 4C). Therefore, mCD160 expression seems to be tion of mCD160/KY-2 cells were tested. Cross-linking of mCD160 calibrated by host MHC I in vivo. by immobilized anti-mCD160 mAb alone had no effect on Because mCD160 binds CD1d in vitro, the effects of CD1d mCD160/KY-2 cells. However, mCD160/KY-2 cells incubated deficiency on mCD160 expression were examined. We compared with immobilized NK1.1 mAb plus mCD160 mAb secreted sig- NKT cells from wild-type and CD1d-deficient BALB/c mice. Be- nificantly lower amounts of IFN-␥ than those stimulated with im- cause BALB/c mice do not express NK1.1, NKT cells were iden- mobilized NK1.1 mAb plus isotype control mAb (Fig. 5D). The tified as DX5ϩ T cells (33) in this experiment. Very few DX5ϩ T treatments did not affect the viability of the cells (data not shown). cells expressed mCD160 in CD1d-deficient BALB/c mice (Fig. These results indicate that mCD160 cross-linking inhibits NK1.1- 4D), whereas ϳ25% of NKT cells in wild-type BALB/c mice were mediated activation of KY-2 cells. mCD160 positive. The expression of mCD160 on DX5ϩTCR␣␤ϩ We also isolated mCD160ϩ and mCD160Ϫ primary NK cells cells also significantly differed between wild-type (48.6 Ϯ 4.5%) and tested cytotoxicity against RMA-S cells. As reported for hu- ␤ Ϯ ϩ and 2m-deficient (22.9 7.0%) B6 mice (data not shown). It man CD160 (14, 15), mCD160 NK cells were more cytotoxic should be noted that NK1.1ϩ and DX5ϩ subsets among T cells than mCD160Ϫ NK cells (data not shown). only partially overlap (33). Because the mCD160 cDNA sequences from BALB/c and B6 mice were identical (see above), the results Discussion suggest that the genetic background of the mice, in addition to the We have characterized mCD160 using newly generated mAb and MHC, influences the expression of mCD160. mCD160-Fc fusion protein. Similar to the human counterpart, mCD160 is a GPI-anchored protein with a single Ig domain. It Functions of mCD160 interacts with a broad range of MHC I, including CD1d. Among By retrovirus-mediated gene transfer into the murine NK cell clone resting lymphocytes, mCD160 is predominantly expressed on KY-2, we established the KY-2 cell line overexpressing mCD160 NK1.1ϩ T cells and a subset of NK cells. Thus, mCD160 is the (mCD160/KY-2) as well as control virus-transduced KY-2 (MIG/ first Ig-like receptor for MHC I identified on mouse NK cells. KY-2). NK1.1 expression on these two KY-2 lines was the same Although the precise function of mCD160 is still unclear, overex- (data not shown). PMA plus ionomycin potently stimulated both pression of mCD160 in the murine NK cell line KY-2 both en- mCD160/KY-2 and MIG/KY-2 cells. However, the former se- hances and inhibits KY-2 cell activation, depending on the stim- creted significantly less IFN-␥ than the latter (Fig. 5A). In contrast, ulus. Overexpression of mCD160 in KY-2 inhibits stimulation by upon NK1.1 cross-linking by immobilized mAb, mCD160/KY-2 PMA plus ionomycin, but augments stimulation with immobilized secreted significantly more IFN-␥ than MIG/KY-2 (Fig. 5B). Sim- anti-NK1.1 mAb or BM-DC. Cross-linking of mCD160 on KY-2 4430 MURINE CD160

cells with immobilized mAb inhibits NK1.1-mediated stimulation of KY-2 cells. It should be noted that the anti-mCD160 mAb, CNX46-3, used in this study is a nonblocking Ab, because it does not inhibit the binding of mCD160-Fc to MHC I-transfected cell lines. Incubation of primary NK cells, NKT cells, or mCD160- transduced KY-2 cells with the CNX46-3 mAb, either in solution or immobilized on plastic surface, did not have any detectable effect. The Ab also did not induce redirected killing of P815 target cells (data not shown). Therefore, mCD160 does not seem to act as a stimulatory receptor on its own. Rather, it seems to act as a coreceptor and regulates, either positively or negatively, activation of NK cells, depending on the primary activation signals. Because mCD160 is GPI anchored and does not have a cyto- plasmic tail, it seems likely that the regulatory effects of mCD160 are mediated by its association with lipid rafts. As expected, mCD160 is found in detergent-insoluble, low density fractions in sucrose gradient centrifugation of nonionic detergent-solubilized cells, indicating that it localizes in lipid rafts (data not shown). Transmembrane adaptor proteins associated with lipid rafts are thought to be key mediators of receptor-mediated signaling (34). It Downloaded from is particularly interesting that mCD160 cross-linking inhibits NK1.1-mediated activation of KY-2 cells. NK1.1 is associated with ITAM-bearing Fc⑀RI␥ (35). Fc⑀RI␥ also functions as an adaptor protein for the high affinity IgE receptor Fc⑀RI and is recruited into lipid rafts by Fc⑀RI aggregation. Src family protein tyrosine kinases localized in lipid rafts are thought to initiate Fc⑀RI-mediated signaling http://www.jimmunol.org/ (36). In contrast, C-terminal Src kinase and Src homology 2 domain- containing phosphatase 1, which are thought to transduce inhibitory signals, have been reported to localize to lipid rafts (37, 38). There- fore, lipid rafts seem to act as a platform for the assembly of both stimulatory and inhibitory signaling proteins. It is tempting to spec- ulate that mCD160 on NK or NKT cells may translocate to the site of cell contact by its interaction with MHC I on interacting cells, recruit- ing lipid rafts and raft-associated molecules to the contact site, thus by guest on October 2, 2021 regulating the activation of NK and NKT cells. However, the precise mechanisms by which NK cell activation is regulated by mCD160 remain to be elucidated. The broad recognition of MHC I by mCD160 suggests that mCD160 may bind to the nonpolymorphic ␣3 domain of the H ␤ chain or 2m associated with the H chain. Because mCD160-Fc binds to mouse MHC I-transfected human and rat cell lines re- ␤ gardless of whether mouse 2m is cotransfected (data not shown), ␤ mCD160 does not seem to directly recognize 2m. It is likely that mCD160 is similar to CD8 and recognizes the ␣3 domain of the MHC I H chain. However, it is unclear how CD1d binds to mCD160, because the ␣3 domain of CD1d is quite different from those of other MHC I molecules. It should also be noted that the interaction between mCD160 and MHC I seems weaker than those between MHC I and other NK cell receptors. Although binding of MHC I tetramers to specific Ly49 and CD94/NKG2 is readily detectable by flow cytometry (39–41), their binding to mCD160 is much weaker and barely detectable. This is similar to binding of CD8 to MHC I. Although binding of soluble CD8 dimer to MHC FIGURE 3. Binding of mCD160 to MHC I. A, TAP-sufficient RMA or I has been reported (42), MHC I tetramers do not bind to CD8 on TAP-mutated RMA-S cells were incubated with 40 ␮g/ml biotinylated mature T cells (43). It is also possible that the conformation of mCD160-Fc (shaded histograms) or biotinylated H60-Fc (open histo- grams) plus streptavidin-PE and analyzed by flow cytometry. The expres- sion levels of MHC I were determined by anti-pan MHC I mAb M1/42 (shaded histograms) and compared with isotype-matched control Ab stain- I mAbs (shaded histograms) and compared with isotype-matched controls ing (open histograms). Representative data from at least four independent (open histograms). Representative data from at least four independent ex- experiments are shown. B, A panel of mouse MHC I-transfected RBL-1 periments are shown. C, The mCD160-transfected K562 (shaded histo- cell lines was stained with biotinylated mCD160-Fc (shaded histograms) or grams) or control vector-transfected K562 (open histograms) cells were biotinylated H60-Fc (open histograms) plus streptavidin-PE as described in stained with 10 ␮g/ml OVA/Kb tetramer for1hat4°Candanalyzed by A and analyzed by flow cytometry. The expression levels of individual flow cytometry. Representative data from three independent experiments MHC I on the transfected RBL-1 cells were detected by specific anti-MHC are shown. The Journal of Immunology 4431

FIGURE 5. Murine CD160 regulates stimulation of the NK cell line KY-2. A, A bulk population of mCD160-transduced KY-2 (mCD160/ KY-2; f) or control virus-transduced KY-2 (MIG/KY-2; Ⅺ) cells was isolated by cell sorting of GFP-positive cells. The sorted cells were ex- Downloaded from panded in cultures, 105 cells/well were stimulated with PMA (50 ng/ml) and ionomycin (1 ␮M) for 24 h, and the amount of IFN-␥ secreted in the culture supernatants was determined by ELISA. Data are the mean Ϯ SD of triplicate cultures and are representative of three independent experi- ments. B, The mCD160/KY-2 (f) and MIG/KY-2 (Ⅺ) cells (2 ϫ 105 cells/well) were stimulated with immobilized NK1.1 mAb or isotype con- http://www.jimmunol.org/ trol mAb for 24 h, and the amount of IFN-␥ in the culture supernatants was determined by ELISA. Data are the mean Ϯ SD of triplicate cultures and are representative of three independent experiments. C, The mCD160/ KY-2 (f) and MIG/KY-2 (Ⅺ) cells (105 cells/well) were cultured with (DC) or without (control) BM-DC (105 cells/well) in microculture wells for 24 h, and the amount of IFN-␥ secreted in the culture supernatants was determined by ELISA. Data are the mean Ϯ SD of triplicate cultures and are representative of three independent experiments. D, The mCD160/KY-2 cells (2 ϫ 105 cells/well) were stimulated with various combinations of immobi-

lized mAb in microculture wells. The total concentrations of mAbs to be im- by guest on October 2, 2021 mobilized were kept at 50 ␮g/ml in all wells, and two mAbs were mixed in equal proportions. The amount of IFN-␥ secreted in the culture supernatants was determined by ELISA. Data are the mean Ϯ SD of triplicate cultures and are representative of three independent experiments.

MHC tetramers may not be suitable for the binding of mCD160 or CD8. In addition to the binding of mCD160-Fc to MHC I-trans- fected cells in vitro, small, but significant, differences in the ex- pression levels of mCD160 on the surface of NKT cells between MHC congenic mouse strains as well as between wild-type and ␤ 2m-deficient mice suggest that MHC I of the host interacts with mCD160 and modulates its level of expression on the cell surface in vivo. Such modulations of NK cell receptors are most evident with Ly49. The Ly49 expression level is highest in MHC I-defi- ␤ cient mice such as 2m-deficient mice, and it is down-modulated by its interaction with the host MHC I (32). Strong interaction between Ly49 and host MHC I results in low expression of Ly49. The FIGURE 4. Down-modulation of mCD160 in vivo. A,Intheleft panel, ␤ age- and sex-matched, wild-type (WT; thin histogram) and 2m knockout (bold histogram) B6 mouse spleen cells were stained for NK1.1, CD3⑀, and intensity of mCD160. C, The mean fluorescence intensities (MFI) of mCD160, and mCD160 expression levels on NK1.1ϩCD3ϩ cells were de- mCD160 staining of NK1.1ϩCD3ϩ cells and NK cells (NK1.1ϩCD3Ϫ) termined by flow cytometry. Dot-line histograms show staining with iso- from various strains of mice, as described in A and B, were compared. The type-matched control mAb. In the right panel, mCD160 expression levels numbers show the mean fluorescence intensity Ϯ SD of four individual on NK1.1ϩCD3ϩ spleen cells from age- and sex-matched B10D2 (thin animals. Paired Student’s t test showed statistically significant (p Ͻ 0.05) D, The expression of mCD160 on .(ء) histogram) and B10 (bold histogram) mice were determined by flow cy- differences between the datasets tometry as in the left panel. Dot-line histograms show isotype control mAb DX5ϩCD3ϩ cells from wild-type and CD1d knockout BALB/c mice was staining. B, The expression of mCD160 on NK cells from the same mice analyzed by flow cytometry. Shaded histograms show staining with anti- as those in A was analyzed by flow cytometry. Open histograms show mCD160 mAb, and open histograms show staining with isotype control isotype control Ab staining, and shaded histograms show staining of mAb. The numbers show average percentage (ϮSD) of cells stained with mCD160. The bars indicate the gates to calculate the mean fluorescence anti-m160 mAb obtained from four individual animals. 4432 MURINE CD160 modulation of mCD160 by host MHC I is less pronounced than that cytotoxic effector T lymphocyte subset lacking CD28 expression. Int. Immunol. of Ly49, perhaps reflecting the weak interaction of the former. 14: 445–451. 18. Le Bouteiller, P., A. Barakonyi, J. Giustiniani, F. Lenfant, A. Marie-Cardine, The expression of mCD160 among resting lymphocytes is most M. Aguerre-Girr, M. Rabot, I. Hilgert, F. Mami-Chouaib, J. Tabiasco, et al. 2002. prominent on NK1.1ϩ T cells in B6 mice. Because NK1.1ϩ T cells Engagement of CD160 receptor by HLA-C is a triggering mechanism used by ␤ circulating natural killer (NK) cells to mediate cytotoxicity. Proc. Natl. Acad. Sci. in 2m-deficient B6 mice also express mCD160, its expression is USA 99: 16963–16968. not limited to CD1d-restricted NKT cells. In contrast, mCD160 is 19. Barakonyi, A., M. Rabot, A. Marie-Cardine, M. Aguerre-Girr, B. Polgar, barely detectable on DX5ϩ T cells in CD1d-deficient BALB/c V. Schiavon, A. Bensussan, and P. Le Bouteiller. 2004. Cutting edge: engage- ment of CD160 by its HLA-C physiological ligand triggers a unique cytokine mice, whereas those in wild-type BALB/c mice express mCD160, profile secretion in the cytotoxic peripheral blood NK cell subset. J. Immunol. suggesting that only CD1d-restricted NKT cells may express 173: 5349–5354. mCD160 in this strain. Because mCD160 recognizes CD1d, it may 20. Fawcett, J., C. L. Holness, L. A. Needham, H. Turley, K. C. Gatter, D. Y. Mason, and D. L. Simmons. 1992. Molecular cloning of ICAM-3, a third ligand for play an important role in the development or functions of CD1d- LFA-1, constitutively expressed on resting leukocytes. Nature 360: 481–484. restricted NKT cells. In addition to NK1.1ϩ T cells, a subset of NK 21. Lian, R. H., Y. Li, S. Kubota, D. L. Mager, and F. Takei. 1999. Recognition of class I MHC by NK receptor Ly-49C: identification of critical residues. J. Im- cells expresses mCD160. Although our results with the mouse NK munol. 162: 7271–7276. cells line KY-2 suggest a regulatory function of mCD160 in cy- 22. Karlhofer, F. M., M. M. Orihuela, and W. M. Yokoyama. 1995. Ly-49-indepen- tokine production, the role of mCD160 in NK cell-mediated cy- dent natural killer (NK) cell specificity revealed by NK cell clones derived from p53-deficient mice. J. Exp. Med. 181: 1785–1795. totoxicity is still unknown. It has been reported that CD1d on 23. Maeda, M., N. Uchida, C. J. Eaves, and F. Takei. 2003. Slow receptor acquisition target cells inhibits the cytotoxicity of a subset of NK cells (11– by NK cells regenerated in vivo from transplanted fetal liver or adult bone mar- 13). It remains to be determined whether mCD160 is involved in row stem cells. Exp. Hematol. 31: 1015–1018. 24. Pal, E., T. Tabira, T. Kawano, M. Taniguchi, S. Miyake, and T. Yamamura. 2001. the inhibition of NK cytotoxicity by target cell CD1d. Future stud- Costimulation-dependent modulation of experimental autoimmune encephalomy- ies, including generation of mCD160-deficient mice, will provide elitis by ligand stimulation of V␣14 NK T cells. J. Immunol. 166: 662–668. Downloaded from 25. Chung, B., A. Aoukaty, J. Dutz, C. Terhorst, and R. Tan. 2005. Cutting edge: more information on the role of mCD160 in immune responses signaling lymphocytic activation molecule-associated protein controls NKT cell mediated by NK and NKT cells. functions. J. Immunol. 174: 3153–3157. 26. Lian, R. H., J. D. Freeman, D. L. Mager, and F. Takei. 1998. Role of conserved glycosylation site unique to murine class I MHC in recognition by Ly-49 NK cell Disclosures receptor. J. Immunol. 161: 2301–2306. The authors have no financial conflict of interest. 27. Karasuyama, H., A. Kudo, and F. Melchers. 1990. The proteins encoded by the

VpreB and ␭5 pre-B cell-specific genes can associate with each other and with ␮ http://www.jimmunol.org/ heavy chain. J. Exp. Med. 172: 969–972. References 28. Teitell, M., H. R. Holcombe, L. Brossay, A. Hagenbaugh, M. J. Jackson, L. Pond, 1. Parham, P. 2005. MHC class I molecules and kirs in human history, health and S. P. Balk, C. Terhorst, P. A. Peterson, and M. Kronenberg. 1997. Nonclassical survival. Nat. Rev. Immunol. 5: 201–214. behavior of the mouse CD1 class I-like molecule. J. Immunol. 158: 2143–2149. 2. Takei, F., K. L. McQueen, M. Maeda, B. T. Wilhelm, S. Lohwasser, R. H. Lian, 29. Antonchuk, J., G. Sauvageau, and R. K. Humphries. 2001. HOXB4 overexpres- and D. L. Mager. 2001. Ly49 and CD94/NKG2: developmentally regulated ex- sion mediates very rapid stem cell regeneration and competitive hematopoietic pression and evolution. Immunol. Rev. 181: 90–103. repopulation. Exp. Hematol. 29: 1125–1134. 3. Castells, M. C., L. B. Klickstein, K. Hassani, J. A. Cumplido, M. E. Lacouture, 30. Lutz, M. B., N. Kukutsch, A. L. Ogilvie, S. Rossner, F. Koch, N. Romani, and ␣ ␤ K. F. Austen, and H. R. Katz. 2001. gp49B1- v 3 interaction inhibits antigen- G. Schuler. 1999. An advanced culture method for generating large quantities of induced mast cell activation. Nat. Immunol. 2: 436–442. highly pure dendritic cells from mouse bone marrow. J. Immunol. Methods 223: 4. Brown, M. H., K. Boles, P. A. van der Merwe, V. Kumar, P. A. Mathew, and 77–92. A. N. Barclay. 1998. 2B4, the natural killer and T cell immunoglobulin super- 31. Arase, H., N. Arase, and T. Saito. 1996. Interferon ␥ production by natural killer by guest on October 2, 2021 ϩ family surface protein, is a ligand for CD48. J. Exp. Med. 188: 2083–2090. (NK) cells and NK1.1 T cells upon NKR-P1 cross-linking. J. Exp. Med. 183: 5. Nakamura, A., E. Kobayashi, and T. Takai. 2004. Exacerbated graft-versus-host 2391–2396. disease in PirbϪ/Ϫ mice. Nat. Immunol. 5: 623–629. 32. Hoglund, P., J. Sundback, M. Y. Olsson-Alheim, M. Johansson, M. Salcedo, 6. Hoelsbrekken, S. E., O. Nylenna, P. C. Saether, I. O. Slettedal, J. C. Ryan, C. Ohlen, H. G. Ljunggren, C. L. Sentman, and K. Karre. 1997. Host MHC class S. Fossum, and E. Dissen. 2003. Cutting edge: molecular cloning of a killer cell I gene control of NK-cell specificity in the mouse. Immunol. Rev. 155: 11–28. Ig-like receptor in the mouse and rat. J. Immunol. 170: 2259–2263. 33. Yang, Y., A. Ueno, M. Bao, Z. Wang, J. S. Im, S. Porcelli, and J. W. Yoon. 2003. 7. Welch, A. Y., M. Kasahara, and L. M. Spain. 2003. Identification of the mouse Control of NKT cell differentiation by tissue-specific microenvironments. J. Im- killer immunoglobulin-like receptor-like (Kirl) gene family mapping to chromo- munol. 171: 5913–5920. some X. Immunogenetics 54: 782–790. 34. Horejsi, V., W. Zhang, and B. Schraven. 2004. Transmembrane adaptor proteins: 8. Godfrey, D. I., H. R. MacDonald, M. Kronenberg, M. J. Smyth, and L. Van Kaer. organizers of immunoreceptor signalling. Nat. Rev. Immunol. 4: 603–616. 2004. NKT cells: what’s in a name? Nat. Rev. Immunol. 4: 231–237. 35. Arase, N., H. Arase, S. Y. Park, H. Ohno, C. Ra, and T. Saito. 1997. Association 9. Maeda, M., A. Shadeo, A. M. MacFadyen, and F. Takei. 2004. CD1d-indepen- with FcR␥ is essential for activation signal through NKR-P1 (CD161) in natural ␤ ϩ dent NKT cells in 2-microglobulin-deficient mice have hybrid phenotype and killer (NK) cells and NK1.1 T cells. J. Exp. Med. 186: 1957–1963. function of NK and T cells. J. Immunol. 172: 6115–6122. 36. Holowka, D., and B. Baird. 2001. Fc⑀RI as a paradigm for a lipid raft-dependent 10. Maeda, M., S. Lohwasser, T. Yamamura, and F. Takei. 2001. Regulation of NKT receptor in hematopoietic cells. Semin. Immunol. 13: 99–105. cells by Ly49: analysis of primary NKT cells and generation of NKT cell line. 37. Davidson, D., M. Bakinowski, M. L. Thomas, V. Horejsi, and A. Veillette. 2003. J. Immunol. 167: 4180–4186. Phosphorylation-dependent regulation of T-cell activation by PAG/Cbp, a lipid 11. Chang, C. S., L. Brossay, M. Kronenberg, and K. P. Kane. 1999. The murine raft-associated transmembrane adaptor. Mol. Cell Biol. 23: 2017–2028. nonclassical class I major histocompatibility complex-like CD1.1 molecule pro- 38. Fawcett, V. C., and U. Lorenz. 2005. Localization of Src homology 2 domain- tects target cells from lymphokine-activated killer cell cytolysis. J. Exp. Med. containing phosphatase 1 (SHP-1) to lipid rafts in T lymphocytes: functional 189: 483–491. implications and a role for the SHP-1 carboxyl terminus. J. Immunol. 174: 12. Huang, M. M., P. Borszcz, S. Sidobre, M. Kronenberg, and K. P. Kane. 2004. 2849–2859. CD1d1 displayed on cell size beads identifies and enriches an NK cell population 39. Hanke, T., H. Takizawa, C. W. McMahon, D. H. Busch, E. G. Pamer, negatively regulated by CD1d1. J. Immunol. 172: 5304–5312. J. D. Miller, J. D. Altman, Y. Liu, D. Cado, F. A. Lemonnier, et al. 1999. Direct 13. Campos-Martin, Y., d. M. Gomez, B. Gozalbo-Lopez, J. Suela, and assessment of MHC class I binding by seven Ly49 inhibitory NK cell receptors. E. Martinez-Naves. 2004. Expression of human CD1d molecules protects target Immunity 11: 67–77. cells from NK cell-mediated cytolysis. J. Immunol. 172: 7297–7305. 40. Michaelsson, J., A. Achour, M. Salcedo, A. Kase-Sjostrom, J. Sundback, 14. Maiza, H., G. Leca, I. G. Mansur, V. Schiavon, L. Boumsell, and A. Bensussan. R. A. Harris, and K. Karre. 2000. Visualization of inhibitory Ly49 receptor spec- 1993. A novel 80-kD cell surface structure identifies human circulating lympho- ificity with soluble major histocompatibility complex class I tetramers. Eur. cytes with natural killer activity. J. Exp. Med. 178: 1121–1126. J. Immunol. 30: 300–307. 15. Anumanthan, A., A. Bensussan, L. Boumsell, A. D. Christ, R. S. Blumberg, 41. Salcedo, M., F. Colucci, P. J. Dyson, L. A. Cotterill, F. A. Lemonnier, P. Kourilsky, S. D. Voss, A. T. Patel, M. J. Robertson, L. M. Nadler, and G. J. Freeman. 1998. J. P. Di Santo, H. G. Ljunggren, and J. P. Abastado. 2000. Role of Qa-1b-binding Cloning of BY55, a novel Ig superfamily member expressed on NK cells, CTL, receptors in the specificity of developing NK cells. Eur. J. Immunol. 30: 1094–1101. and intestinal intraepithelial lymphocytes. J. Immunol. 161: 2780–2790. 42. Meyerson, H. J., J. H. Huang, J. D. Fayen, H. M. Tsao, R. R. Getty, 16. Agrawal, S., J. Marquet, G. J. Freeman, A. Tawab, P. L. Bouteiller, P. Roth, N. S. Greenspan, and M. L. Tykocinski. 1996. Functional dissociation of CD8␣’s W. Bolton, G. Ogg, L. Boumsell, and A. Bensussan. 1999. Cutting edge: MHC Ig homologue and connecting peptide domains. J. Immunol. 156: 574–584. class I triggering by a novel cell surface ligand costimulates proliferation of 43. Daniels, M. A., L. Devine, J. D. Miller, J. M. Moser, A. E. Lukacher, activated human T cells. J. Immunol. 162: 1223–1226. J. D. Altman, P. Kavathas, K. A. Hogquist, and S. C. Jameson. 2001. CD8 17. Nikolova, M., A. Marie-Cardine, L. Boumsell, and A. Bensussan. 2002. BY55/ binding to MHC class I molecules is influenced by T cell maturation and glyco- CD160 acts as a co-receptor in TCR signal transduction of a human circulating sylation. Immunity 15: 1051–1061.