The Association of MHC Class I with the 2B4 Receptor Inhibits Self-Killing of Human NK Cells

This information is current as Gili Betser-Cohen, Saar Mizrahi, Moran Elboim, Osnat of September 27, 2021. Alsheich-Bartok and Ofer Mandelboim J Immunol 2010; 184:2761-2768; Prepublished online 17 February 2010; doi: 10.4049/jimmunol.0901572

http://www.jimmunol.org/content/184/6/2761 Downloaded from

References This article cites 39 articles, 14 of which you can access for free at: http://www.jimmunol.org/content/184/6/2761.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

• Fast Publication! 4 weeks from acceptance to publication by guest on September 27, 2021

*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 © 2010 by The American Association of Immunologists, Inc. All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

The Association of MHC Class I Proteins with the 2B4 Receptor Inhibits Self-Killing of Human NK Cells

Gili Betser-Cohen, Saar Mizrahi, Moran Elboim, Osnat Alsheich-Bartok, and Ofer Mandelboim

The killing activity of NK cells is carried out by several activating NK receptors, which includes NKp46, NKp44, NKp30, NKp80, NKG2D, and 2B4. The ligands of these receptors are either self-derived, pathogen-derived, stress-induced ligands or tumor ligands. Importantly, none of these killer ligands are expressed on NK cells and thus self-killing of NK cells is prevented. A notable exception with this regard, is the ligand of the 2B4 receptor. This unusual receptor can exert both activating and inhibiting signals; however, in human NK cells, it serves mainly as an activating receptor. The ligand of 2B4 is CD48 and in contrast to the ligands of all the other NK activating receptors, CD48 is also present on NK cells. Thus, NK cells might be at risk for self-killing that is mediated via the

2B4-CD48 interaction. In this study, we identify a novel mechanism that prevents this self-killing as we show that the association of Downloaded from the MHC class I proteins with the 2B4 receptor, both present on NK cells, results in the attenuation of the 2B4-mediated self-killing of NK cells. The Journal of Immunology, 2010, 184: 2761–2768.

atural killer cells are cytotoxic lymphocytes that belong to CD150 subfamily, within the CD2 family, includes several dual- the innate immune system. They were initially described functional receptors such as CD150, CD84, CD229, CD244 N as cells able to kill virally infected and tumor cells (1), (2B4), NTB-A, and CS1 (11) that can exert both activating and http://www.jimmunol.org/ although it is known today that they posses several additional inhibitory signals, depending on the adaptor proteins, which they functions such as production of growth factors during pregnancy (2) interact with (12). An example for such a receptor, is 2B4, the and even being APCs (3). The killing of NK cells is regulated by subject of this research, which contains within its cytoplasmic tail surface receptors that either induce or inhibit the cytotoxic re- four signaling ITSM (immunoreceptor tyrosine-based switch sponse. NK inhibition is mediated mainly via the recognition of motif) (13). MHC class I molecules by NK inhibitory receptors, such as the The ligands for the activating NK receptors that have been killer-Ig–like receptors (KIRs), CD94/NKG2A, and the leukocytes identified so far includes stress-induced ligands for NKG2D (14), Ig-like receptors (LIRs). Recently, we have shown that TIGIT could viral ligands for NKp44, NKp46, and NKp30 (15–17), a soluble also inhibit NK cells by binding to PVRL-2 and PVR (4). MHC ligand for NKp30 (18), a tumor ligand for NKp30 (19), and self- by guest on September 27, 2021 class I proteins are also needed for proper NK cell education (5, 6) ligands such as AICL and PVR/PVRL-2 for the NKp80 (20) and and in the absence of MHC class I proteins the killing activity of DNAM/CD96 (21) receptors, respectively. Finally, the CD48 NK cells is impaired. , has been identified as a ligand for 2B4 (22) and CD2 (10). The killing of NK cells is extracted by a limited number of In humans, 2B4 almost exclusively functions as an activating re- activating and coactivating NK receptors, including the natural ceptor and its inhibitory properties are mostly evident in genetic cytotoxicity receptors NKp46, NKp44, NKp30 (known as the deficiencies in which the signaling lymphocyte activation mole- NCRs), as well as by NKp80, DNAM, CD96, and the NKG2D cule-associated protein (SAP) is absent. In such a deficiency receptor (7, 8). Another family of killer receptors that is involved named X-linked lymphoproliferative disease (XLP) (23), different in the killing mediated by NK cells is the CD2 family (9, 10). The phosphatases such as Src homology region 2 domain-containing phosphatase 1 (SHP-1), SHP-2, and SHIP are recruited, instead of SAP, and 2B4 becomes inhibitory (24). The Lautenberg Center for General and Tumor Immunology, The Institute for Med- Among all the killer receptor ligands, CD48 is the only protein ical Research Israel Canada, Hebrew University-Hadassah Medical School, Jerusa- expressed on NK cells (25). Therefore, we wondered why NK cells lem, Israel do not kill each other via the 2B4-CD48 interactions, and hy- Received for publication June 26, 2009. Accepted for publication December 23, 2009. pothesized that NK cells probably possess a mechanism that This work was supported by grants from the US–Israel Bi-National Science Foun- prevent this 2B4-mediated self-killing. dation, the Israeli Cancer Research Foundation, The Israeli Science Foundation, the legacy heritage Bio-Medical program of the Israel Science Foundation (1838/08), The European Consortium (MRTN-CT-2005 and LSCH-CT-2005-518178), and by the Association for International Cancer Research. Materials and Methods Address correspondence and reprint requests to Prof. Ofer Mandelboim, The Lauten- Cell, transfectants, mAb, and proteins berg Center for General and Tumor Immunology, The Institute for Medical Research Israel Canada, Hadassah Medical School, 91120 Jerusalem, Israel. E-mail address: The cell lines used in this work were the MHC class I-negative human B [email protected] lymphoblastoid cell line 721.221, various MHC class I transfectants, the human NK cell line YTS eco, and the Fc receptor positive murine mas- Abbreviations used in this paper: BG, background; ITSM, immunoreceptor tyrosine- tocytoma cell line P815. Primary human NK cells were isolated from PBLs based switch motif; KIR, killer-Ig–like receptor; LIR, leukocytes Ig-like receptor; Med, medium only; PI, propidium iodide; SAP, signaling lymphocyte activation by using a human NK cell isolation and an auto-MACS instrument molecule-associated protein; SHP, Src homology region 2 domain-containing phos- (Miltenyi Biotec, Auburn, CA). phatase 1; XLP, X-linked lymphoproliferative disease. For the generation of YTS cells expressing the chimeric protein con- taining the extracellular portion of NKp44 fused to the transmembrane and Copyright Ó 2010 by The American Association of Immunologists, Inc. 0022-1767/10/$16.00 cytoplasmic tail portion of 2B4, plasmid constructs were prepared in EcoRI- www.jimmunol.org/cgi/doi/10.4049/jimmunol.0901572 2762 MHC CLASS I ON NK CELLS PREVENTS SELF-KILLING

BAMH1 pBABE vector. Transfection with the encoding plasmid for Class I MHC tetramers-SA-PE that are used in this work are composed of the chimeric NKp44/2B4 protein or for the chimeric NKp44/Cw7 was a complex offour HLA-A2MHC class I molecules,each bound to the specific performed in three PCR steps: first, amplification of the extracellular do- peptide and conjugated with a fluorescent protein (Beckman Coulter). main of NKp44 containing 3 aa from the beginning of the transmembrane BrdU and propidium iodide assays domain of 2B4 or HLA-Cw7, using the same 59 primer for both proteins GGAATTCCGCCGCCACCATGGCCTGGCGAGCCCTACAC (including For BrdU incorporation, cells (1 3 106) were incubated with 20 mM BrdU for the EcoRI restriction site) and 39 primer for 2B4 CACCAAAAAGG- 30 min, washed three times with PBS, fixed with 5 ml ice-cold 70% ethanol CAATGGGGGCTGCAGGG or 39 primer for HLA-Cw7 GATGCCCATGG- and kept over night at 220˚C. On the next day, cells underwent centrifu- CAATGGGGGCTGCAGGG. Second, amplification includes 3 aa derived gation and washed three times to remove ethanol residues, and emulsified from the end of the extracellular domain of NKp44 with the transmembrane with 0.3 ml PBS, 30 min on ice (rehydration). DNA denaturation was per- domain and the tail of 2B4 receptor or HLA-Cw7, for the 2B4 using 59 primer formed with 1 ml fresh HCL solution (0.5% Triton X-100, 2 N HCL) for 30 CCCATTGCCTTTTTGGTGATCATCGTGATTCTA and for the HLA-Cw7 min at room temperature. After the incubation, cells were washed twice, using 59 primer CCCATTGCCATGGGCATCGTTGCTGGCGGCCTG. The resuspended in 1 ml basic solution (0.1 M Na2B4O7), and immediately 39 primer for both proteins CGGGATCCCGCTAGGAATTAAACATCAAA- washed with incubation solution (1% BSA, 0.5% Triton X-100). For BrdU GTTCTC (including the BamHI restriction site). The final PCR step included staining, the cells were resuspend in 50 ml incubation solution containing mixing of the two PCR products that were used as templates and PCR ampli- 1:10 anti–BrdU-FITC for 30 min on ice. Cells were washed twice and re- fication using the 59 primer of the first PCR and the 39 primerof the second PCR. suspended with PBS containing 50 mg/ml RNAse for 15 min at room tem- All transfectants were periodically monitored for expression by staining with perature. Before FACS analysis, 5 mg/ml propidium iodide (PI) was added. the appropriate mAb. The CellQuest software program was used for data acquisition and analysis. The mAbs used in this work are C17 mAb (Beckman Coulter, Fullerton, Cytotoxic assays CA) directedagainst2B4; anti-CD48mAb;GB7(IgG1),W632,and mem147 (IgG1) directed against MHC class I molecules; anti-BBM1 directed against The cytotoxic activity of YTS cells was assessed in 5 h [35S]-Met release the L chain (b-2-microglobulib) of the MHC complex; B1.23.2 directed assays, as previously described (26). In the redirected lysis experiment Downloaded from against HLA-C; GL183 (CD158b) mAb directed against KIR2DL3 (Im- target p815 cells were preincubated for 1 h, on ice with 0.1 mg/well of the munotech, Westbrook, ME); biotynlated anti-human CD107a (LAMP-1) indicated Abs. Percent-specific lysis was calculated as 100% 3 [cpm mAb (BD Biosciences Pharmingen, San Jose, CA); anti–BrdU-FITC (experimental well) – cpm (spontaneous release)/[cpm (maximum release (Becton Dickinson, San Jose, CA), and anti-NKp44 mAb IgG2A (R&D – cpm (spontaneous release)]. Maximum release was determined by adding Systems, Minneapolis, MN); BBM.1 is an anti–b-2-microglobulin mAb; 0.1 M NaOH. The spontaneous release was ,25% of maximal release. For HC10 (IgG2a) is directed against HLA-B and -C free heavy chains. the statistical analysis, we used t test. http://www.jimmunol.org/ by guest on September 27, 2021

FIGURE 1. Identification of a mAb that inhibits the 2B4-mediated killing. A, Staining of NK cells or YTS eco cells with the GB7 mAb (gray histogram). Background (BG) is the staining of the same cells with secondary FITC-conjugated goat anti-mouse Abs (black histogram). B, GB7 inhibits the 2B4-mediated killing. YTS eco, or NK cells were tested in a redirected killing assay against P815 target cells coated with an anti-2B4 Ab (black bars) or with an anti-2B4 mAb combined with the GB7 mAb (gray bars). pp # 0.002; ppp # 0.001. C, The GB7-mediated inhibition is specific to 2B4. Redirected killing assay mediated by NK cells was performed using anti-CD16, NKp30, NKp44, NKp46, and NKG2D Abs incubated with P815 cells in the presence (gray bars) or absence (black bar) of the GB7 mAb. Med is for medium only. D, Titration of GB7 mAb. Redirected killing assay mediated by YTS eco cells (upper) or primary bulk NK cells (lower) was performed using different mAb concentrations (0.004–0.5 mg/well) of GB7 mAb or anti-CD99 mAb with and without 0.1 mg/well of the 2B4 mAb. The percentage of specific lysis is shown for an E:T ratio of 5:1. Shown is one representative experiment of 10 performed. ppp # 0.001. The Journal of Immunology 2763

FIGURE 2. GB7 mAb recognizes MHC class I proteins. A, Sequencing the protein band recognized by GB7. Immunoprecip- itation was performed using the GB7 mAb and analyzed on SDS-PAGE gels. The protein bands were sent for sequencing and the identified peptides are shown. B, The GB7 mAb recognizes HLA-A, -G, and -B, but not -C proteins. Staining of dif- ferent 721.221 transfectants (indicated in the x-axis) was performed either with the GB7 mAb (black bar) or with W6/32 mAb (gray bar). Downloaded from http://www.jimmunol.org/ Protein band sequencing ona capillaryHPLC system (CapLC Waters,Milford,MA) coupledtotheMS. Mass spectrometry was carried out with Q-TOF2 (Micromass, Manchester, The peptide mixtures was either solid phase extracted with C18 resin filled tip England) using nanospray attachment (27). Data analysis was performed (ZipTip Milipore, Billerica, MA) and nanosprayed into the Q-TOF2 MS using the biolynx package (Micromass), and database searches were per- systemin50%CH3CN1%CHOOHsolutionorinjectedto0.75UlC18column formed with the Mascot package (Matrix Science, London, England). by guest on September 27, 2021

FIGURE 3. Inhibition of the 2B4-mediated killing by various anti-MHC class I mAbs. A, FACS staining of YTS eco and NK cells. Staining (indicated by arrows, gray histogram) was performed with the following Abs: GB7, W632, mem147 (all directed against MHC class I), and the BBM1-anti–b-2-mi- croglobulin. BG is the staining of the cells with secondary FITC-conjugated goat anti-mouse Abs (black histogram). B, Redirected killing assay performed with bulk NK cell cultures (black bar) and YTS eco cells (gray bars) by using P815 cells, precoated either with the anti-2B4 mAb alone or with the 2B4 mAb, together with the different anti MHC class I mAbs (indicated in the x-axis). The percentage of specific lysis is shown for a E:T ratio of 5:1. Shown is one representative experiment of five performed. pp # 0.002; ppp # 0.001. 2764 MHC CLASS I ON NK CELLS PREVENTS SELF-KILLING shRNA cules (Fig. 2B). As expected, the control mAb W632, a pan MHC The shRNA constructs directed against 2B4 (Open Biosystems, Huntsville, class I mAb, recognized all MHC class I proteins (Fig. 2B). AL) were designed to include a hairpin of 21-bp sense and antisense stem To further demonstrate that indeed MHC class I molecules are and a 6-bp loop. Each hairpin sequence was cloned into lentiviral vector capable of inhibiting the 2B4-mediated killing, we used several (pLKO.1). This vector allows for production of viral particles using len- anti-MHC class I mAbs and performed redirected killing experi- tiviral packaging cell lines (293-T). Two days after transfection, the soups ments. As expected, all four mAbs (GB7, W6/32, mem147, and containing the viruses were collected and filtered. These viruses were used to transduce YTS eco cells in the presence of polybrene (5 mg/ml). BBM.1) recognized MHC class I proteins on YTS and NK cells (Fig. 3A). Strikingly and in agreement with the inhibition of 2B4- Immunoprecipitation mediated killing observed with GB7 mAb, all anti-MHC class I YTS eco cells (15 3 106–20 3 106/ml) or YTS/NKp44-2B4 were washed mAb inhibited the 2B4-mediated killing of both primary NK and four times with cold PBS containing 1 mM MgCl2 and 0.1 mM CaCl2. For YTS cells (Fig. 3B). immunoprecipitation with GB7, the cells were biotinylated with EZ-Link Sulfo-NHS-LC-Biotin (Pierce, Rockford, IL) for 30 min at 4˚C and Engagement of MHC class I proteins by their appropriate washed four times with cold PBS. For immunoprecipitation with anti-2B4 inhibitory receptors inhibits self-killing of YTS cells (C17), or anti-CD99 (12E7), or anti–b-2-microglobulin (BBM.1), the cells were first incubated with 0.1 mM of vanadate for 10 min at 37˚C, washed The above results show that cross-linking of MHC class I proteins four times with RPMI 1640. on NK cells inhibits the 2B4-mediated killing. Under physiological After washing, cells were lysed with lysis buffer (PBS containing 150 conditions, such cross-linking could result from the cis or trans mM NaCl, 50 mM TRIS [pH 7.6], 0.5% Nonidet P-40, 9 mM iodoaceta- interactions formed between the NK inhibitory receptors and their mide, 5 mM EDTA, 1 mM PMSF, 1:100 aprotinin). Cell lysates were then immunoprecipitated for 3 h at 4˚C with 20 mg of the necessary Ab followed cognate MHC class I proteins that are expressed on the NK cells. Downloaded from by Protein-G plus beads (Pierce). The immunoprecipitates were washed To test whether indeed cross-linking of the MHC class I proteins with lysis buffer, and the proteins were eluted by using the IgG elution buffer (Pierce). All the samples were loaded on SDS-PAGE and then transfer to nitro- cellulose membrane. The blotted proteins were visualized by antistrepavidin HRP or by anti-free H chain (HC10) or anti-NKp44 or anti-2B4, followed by

anti-mouse HRP (DakoCytomation, Carpinteria, CA), using ECL detection http://www.jimmunol.org/ (Biological Industries, Bet-haemek, Israel). A sample from the cell lysate was taken before immunoprecipitation for the detection of protein amount.

Results Inhibition of the 2B4-mediated killing of NK cells by a novel mAb Because both 2B4 and its ligand CD48 are expressed on NK cells, we reasoned that a mechanism aiming at preventing the 2B4-CD48 by guest on September 27, 2021 self-killing should exist. As the 2B4 killing machinery is intact in YTS cells and YTS also express CD48 and because we have previously generated mAbs against YTS to identify the function of CD100 (28), we screened 200 hybridomas that recognized YTS and NK cells (example is shown in Fig. 1A) for their ability to inhibit the 2B4-mediated killing of YTS and NK cells. In these assays, we triggered 2B4 either on NK cells or on YTS cells by using an anti-2B4 mAb and included also the 200 hybridomas that we have generated. As can be seen in Fig. 1B, one of all hy- bridomas tested named GB7 inhibited the 2B4 mediated killing of both YTS and bulk NK cell cultures. The other hybridomas tested had no effect (data not shown). We next tested whether the killing extracted by NK receptors, other than 2B4, such as CD16, NKp30, NKp44, NKp46, and NKG2D will also be inhibited by GB7 mAb and observed that the GB7-mediated inhibition is restricted to 2B4 only (Fig. 1C). The effect is specific because the titration of the FIGURE 4. MHC class I is involved in the self-killing of YTS cells. A, GB7 mAb resulted in a gradually reduced inhibition in both YTS FACS analysis. GL183 mAb staining (open black histogram) of YTS ex- eco and primary bulk NK cell cultures (Fig. 1D). pressing KIR2DL3 (right panel) and YTS expressing the mutated KIR2DL3 in which the two tyrosines at the ITIM were mutated to phenylalanine, MHC class I proteins inhibit the 2B4-mediated killing KIR2DL3(2YF) (left panel). Staining of YTS eco cells with the GL183 To identify the protein recognized by the GB7 mAb, we performed mAb is shown as control (filled gray histogram). B, The killing properties of immunoprecipitation experiments using YTS cells and the GB7 the various YTS transfectants. Redirected killing assay were performed with mAb. We obtained two bands in the size of ∼45 kDa and a slightly YTS eco (black bar), YTS-KIR2DL3(2YF) (white bar) and YTS-KIR2DL3 larger band of around 75 kDa (Fig. 2A). All three bands were sent (gray bar) using the indicated Abs. The last three right columns show the killing of 721.221 cells. Med is for medium only. The percentage of specific for sequencing and, surprisingly, all peptides derived from all bands lysis is shown for a E:T ratio of 5:1. Shown is one representative experiment corresponded to peptides derived from MHC class I proteins (Fig. of three performed. ppp # 0.001. C, Blocking the KIR2DL3(2YF) resulted 2A). To further confirm that indeed GB7 recognizes MHC class I in increased killing. Killing assay of 721.221 cells was performed with molecules, we stained different 721.221 transfectants expressing YTS- KIR2DL3(2YF) that were preincubated with B1.23.2 or GB7 or various MHC class I proteins with the GB7 mAb and observed that GL183 mAbs. The lysis was performed at different E:T ratios as indicated. GB7 recognizes HLA-G, HLA-A, HLA-B, but not HLA-C mole- Shown is one representative experiment of three performed. The Journal of Immunology 2765 by the inhibitory receptors would inhibit the 2B4-mediated YTS To verify that indeed the ITIM mutations in YTS-KIR2DL3 self-killing, we initially expressed in YTS cells the LIR1 in- (2YF) render this receptor nonfunctional, we performed a redir- hibitory receptor that recognizes a large group of MHC class I ected killing assay with P815 cells coated with anti-KIR2DL3 mAb proteins (29, 30). LIR1-positive YTS transfectants were obtained, combined with anti-2B4 mAb. All three cell lines; YTS eco, YTS- however, they stopped proliferating and died within a short period KIR2DL3, and YTS-KIR2DL3(2YF) showed a similar killing (data not shown). We assumed that the reason for this phenome- profile against P815 cells coated with anti-2B4 only or against non is because the LIR1 interaction with MHC class I proteins on 721.221 cells (Fig. 4B), indicating that the killing machinery in all YTS cells results in an inhibitory signal that prevented YTS YTS transfectant is intact and equal among all various trans- proliferation. This observation might explain why the YTS tumor fectants. As expected, when redirected killing experiments were cell line, which is derived from a cancer patient (31), lacks the performed with anti-2B4 and anti-KIR2DL3 mAb, inhibition was expression of all known inhibitory receptors. We next thought to observed only in the wild-type YTS-KIR2DL3 cells (Fig. 4B). express in YTS cells a KIR receptor that will interact with a spe- Thus, the mutated KIR2DL3 protein is, indeed, nonfunctional and cific MHC class I protein and for that purpose we determined the can be used to test whether the engagement of MHC class I MHC haplotype of YTS cells to be HLA-A-26, A-31, Cw-01, Cw- proteins (in this particular case HLA-Cw-08) will affect the 2B4 08, B-48, B-55. We previously successfully expressed the killing. We next assumed that if indeed the interaction between the KIR2DL1 receptor in YTS cells (32) and we think that this effi- mutated KIR2DL3 and the MHC class I proteins on the NK cells cient expression was due to the absence of the appropriate MHC inhibit the self- killing of NK cells, blocking of such interactions class I ligand for KIR2DL1. YTS cells express HLA-Cw-08 that is will result in increased killing of target cells. We used the YTS- a ligand for KIR2DL3 and therefore we expressed in YTS cells KIR2DL3(2YF) cells to avoid inhibition that is mediated via Downloaded from a wild-type KIR2DL3 receptor and a mutated KIR2DL3 protein, KIR2DL3, and as can be seen in Fig. 4C, indeed blocking of MHC in which the two tyrosine residues in its ITIM motif were mutated class I with the B1.23.2 mAb, or blocking the receptor with the to phenylalanin (KIR2DL3[2YF]) (Fig. 4A). In agreement with GLI83 mAb resulted in increased killing, whereas the GB7 mAb our hypothesis, expression of KIR2DL3 was observed on the cell had no effect (probably because it is a nonblocking mAb). To surface; however, the transfectant grew for about 3 wk and then further confirm that the MHC class I proteins prevent the 2B4-

stopped proliferating and died (data not shown). In contrast, the mediated self-killing of NK cells, we initially demonstrated that http://www.jimmunol.org/ mutated KIR2DL3 protein was stably expressed on YTS cells for YTS eco and YTS/KIR2DL3(2YF) cells express similar levels of a long period. the killer receptor 2B4 and its ligands CD48 (Fig. 5A). Next, we by guest on September 27, 2021

FIGURE 5. The self-killing mediated by the 2B4 receptor is decreased after KIR2DL3(2YF) expression. A, FACS analysis. Expression levels of the 2B4 receptor and its ligand CD48 on YTS eco, and YTS-KR2DL3(2YF) cells (open black histogram). The BG is staining with secondary FITC-conjugated goat anti-mouse Abs (filled gray histogram). B, Self-killing of YTS cells. Staining of CD107a on the surface of YTS eco cells and on YTS-KIR2DL3(2YF) cells. BG is the staining of the cells with secondary FITC-conjugated goat anti-mouse Abs (filled gray histogram). C, BrdU labeling of the YTS eco and YTS- KIR2DL3(2YF). The density plots show (upper panel) the percentages of the proliferated cells (S-phase). The histograms (lower panel) present cell cycle analysis by PI staining. 2766 MHC CLASS I ON NK CELLS PREVENTS SELF-KILLING used two types of assays, CD107a mobilization and BrdU in- 2B4 interactions with CD48 by using a specific mAb against 2B4 corporation. The CD107a measure NK cell degranulation and we and demonstrated that such blocking of resulted in reduced CD107a reasoned that if the interactions between the inhibitory receptor mobilization (Fig. 6D), whereas, the control CD99 had little or no KIR2DL3(2YF) and the MHC class I proteins inhibit self-killing, effect (data not shown). These three assays indicate that 2B4 is these YTS cells should be less degranulated. Indeed, as can be directly involved in the self-killing of YTS cells by directly inter- seen in Fig. 5B YTS cells expressing the inhibitory-signal de- acting with MHC class I proteins. ficient KIR2DL3(2YF) express less CD107a as compared with The interaction of 2B4 and the MHC molecules is crucial for YTS eco cells. In the second assay we used BrdU, a synthetic the inhibition of self-killing analog of thymidine, which is incorporated into the newly syn- thesized DNA of replicating cells (during the S-phase of the cell To investigate the mechanism responsible for the MHC class cycle) and again predicted that YTS/KIR2DL3(2YF) cells will be I-mediated inhibition of 2B4 killing, we generated two types of more proliferating as compared with YTS cells because self- chimeric molecules in which the extracellular portion of the NKp44 killing is prevented. As expected, the percentage of the pro- receptor was fused either to the transmembrane and cytoplasmic liferating cells (S-phase) was enhanced in KIR2DL3(2YF) cells tail of 2B4 (YTS 44-2B4, Fig. 7A) or to the transmembrane and (Fig. 5C, upper panel). Furthermore, cell cycle analysis using PI cytoplasmic tail of HLA-Cw7 (YTS 44-Cw7, Fig. 7B), and ex- staining detected an apoptotic sub-G1 population that is absent in pressed them in YTS eco cells. YTS-KIR2DL3(2YF) cells (Fig. 5C, lower panel). Importantly, cross-linking of YTS 44-2B4 with anti-NKp44 mAb, resulted in efficient killing, which was equivalent to that of MHC class I proteins directly bind to 2B4 and the self-killing of 2B4 and the NKp44-2B4–mediated killing was inhibited when Downloaded from YTS cells is mediated via the 2B4 receptor MHC class I molecules were engaged (Fig. 7A, lower panel). The above results demonstrate that the interaction between the Thus, the transmembrane and cytoplasmic tail of 2B4 are re- inhibitory receptor and MHC class I on NK cells inhibit the self- sponsible for the MHC class I proteins-mediated inhibition. As killing of NK cells in a 2B4-dependent manner. To demonstrate that expected, cross-linking of YTS 44-Cw7 with anti-NKp44 mAb MHC class I proteins interact directly with 2B4, we initially re- did not result in killing, as the tail and transmembrane portions of

pressed the 2B4 expression using shRNA against the 2B4 receptor the MHC class I proteins do not contain any activating signal http://www.jimmunol.org/ (Fig. 6A) and then examine the binding of HLA-A2 tetramer to the needed for killing (Fig. 7B, lower panel). Furthermore, cross- various cells. As can be seen in Fig. 6B, efficient binding of the linking of YTS 44-Cw7 cells with anti-NKp44 mAb together with HLA-A2 tetramer was observed to YTS cells expressing 2B4 and the cross-linking of 2B4, did not inhibit the 2B4-mediated killing when 2B4 expression was reduced, the binding of the HLA-A2 (Fig. 7B, lower panel). tetramer was also much reduced (Fig. 6B). We next tested whether These results indicate that the transmembrane and the cyto- the reduced 2B4 expression will also lead to a reduction in the self- plasmic tail of 2B4 are involved in the MHC class I-mediated killing of YTS cells. As can be seen in Fig. 6C the reduction in 2B4 inhibition and that such inhibition does not require the tail and the expression was accompanied with reduced CD107a mobilization to transmembrane portions of the MHC class I. Thus, we hypothe- the cells surface, which suggest that 2B4 is the receptor responsible sized that the extracellular portion of MHC class I is involved in the by guest on September 27, 2021 for the self-killing of YTS cells. Finally, to directly demonstrate 2B4 inhibition and that this inhibition probably occurs through the that 2B4 is involved in the self-killing of YTS cells, we blocked the association of MHC class I proteins with 2B4. To test this

FIGURE 6. The self-killing of YTS cells is mediated via the 2B4 receptor. A, ShRNA-mediated downregulation of the 2B4 receptor. FACS analysis of the 2B4 expression postinfection of YTS eco cells with shRNA directed against 2B4 receptor. B, Decrease binding of the MHC tetramer resulting from the reduction of 2B4 expression. C, Reduced CD107a expression after 2B4 downregulation. D, Blocking the 2B4 receptor results in reduced self-killing. Staining YTS eco cells with anti-CD107a mAb after blocking of the 2B4 receptor with a specific mAb. For A–D, ”After” indicates after shRNA infection, “Before” indicates before shRNA infection. BG is background with second mAb only. The Journal of Immunology 2767 Downloaded from http://www.jimmunol.org/

FIGURE 7. Mechanism of MHC class I mediated inhibition of self-killing. A, NKp44-2B4 expression and killing assay. Expression of the NKp44 (extracellular domain) fused to 2B4 (transmembrane and tail) visualized with anti-NKp44 mAb (black histogram). Control is the YTS eco cells that were stained with the same anti-NKp44 mAb (filled gray histogram). Lower part, YTS 44-2B4 were tested in redirected killing assay against P815 target cells in the presence of an anti-NKp44 or anti-2B4 mAbs with or without GB7 Ab. pp # 0.002. B, NKp44-Cw7 expression and killing assay. Expression of the NKp44 (extracellular domain) fused to HLA-Cw7 (transmembrane and tail) visualized with anti-NKp44 mAb (black histogram). Control is the YTS eco cells that were stained with the same anti-NKp44 mAb (filled gray histogram). YTS 44-Cw7 cells were tested in redirected killing assay against P815 target cells in the presence of an anti-NKp44 with or without 2B4 mAb. Med is medium only. The percentage of specific lysis is shown for an E:T ratio of 5:1. by guest on September 27, 2021 Shown is one representative experiment of three performed. ppp # 0.001. C, Coimmunoprecipitation of 2B4 or NKp44-2B4 proteins. Immunoprecipitations were performed with anti-NKp44 mAb for the YTS 44-2B4 cells or with anti-2B4 for the YTS eco. Western blot analysis was performed with anti-free H chain (HC10) mAb (left blot). YTS eco cells and YTS 44-2B4 cells were immunoprecipitated with anti–b-2-microglobulin mAb (BBM.1). Western blot analysis was performed with anti-2B4 for YTS eco or with anti-NKp44 for YTS 44-2B4 (right blot). Anti-CD99 mAb was used as a control. To demonstrate that equal levels of proteins were present in the lysates before precipitations, the amounts of the indicated proteins was evaluated. Shown is one repre- sentative experiment of three performed. hypothesis, we immunoprecipitated either the 2B4 protein using expression and we could precipitate MHC class I with anti-2B4 and specific anti-2B4 mAb or the NKp44-2B4 protein using anti- vice versa. The mechanism by which MHC class I affects 2B4 NKp44 mAb and performed Western blot analysis with anti-MHC killing is unknown and all signaling molecules we have tested, class I mAb HC10. As predicted, MHC class I proteins were including SAP, SHP1, SHIP1, FYN kinase, and ERK/pERK were coimmunoprecipitated with both anti-2B4 and anti-NKp44, sug- not altered by the MHC class I cross-linking (data not shown). gesting that MHC class I proteins are probably associated with both We also show in this study that in addition to 2B4, MHC class I receptors (Fig. 7C, left blot). Finally, to further strengthen these proteins also interact with NKp44, but not with NKp30 (data not results we also performed the reciprocal experiments and dem- shown). However, the association of MHC class I molecules with onstrated that both the 2B4 or NKp44 receptors were coimmu- NKp44 did not lead to the inhibition of NKp44-mediated killing, noprecipitated when the anti–b-2-microblobulin mAb was used but only affects 2B4 killing. (Fig. 7C, right blot). If 2B4 is such a dangerous, potentially self-destructing receptor why do all NK cells express this receptor? One possibility is that 2B4 Discussion is important for killing of hazardous pathogens/tumors. Indeed, it In contrast to all other NK receptors, whose ligands are not was shown that 2B4 is an important modulator of immune cells expressed on NK cells, the CD48 protein and its interacting receptor activity and that overexpression of CD48 enhanced the NK-medi- 2B4 are both expressed on normal hematopoietic cells, including ated killing (25, 33). It seems, however, that the most crucial NK cells. This dangerous situation might potentially lead to the self- function of 2B4 is during viral infections (34) and the most con- killing of NK cells. In this study, we demonstrate that the MHC vincing evidence for this comes from XLP patients. In these pa- class I proteins expressed on NK cells inhibit the 2B4 self-killing not tients, SAP is absent, 2B4 can no longer function as an activating only by serving as an inhibitory ligand for inhibitory NK receptors receptor, and the patients suffer from fatal immune dysfunction, but also through the association with 2B4. We demonstrate a direct associated with the inability to control EBV infections (35). association between MHC and 2B4 as the binding of MHC class I Another difficult question is why should NK cells express CD48, tetramer was reduced concomitantly with the reduction of 2B4 the ligand of 2B4, and be susceptible to self-killing. One possible 2768 MHC CLASS I ON NK CELLS PREVENTS SELF-KILLING explanation is that other proteins also interact with CD48 and that 16. Cerwenka, A., and L. L. Lanier. 2001. Ligands for receptors: redundancy or specificity. Immunol. Rev. 181: 158–169. such an interaction would be beneficial for NK cell activity. Indeed, 17. Mandelboim, O., N. Lieberman, M. Lev, L. Paul, T. I. Arnon, Y. Bushkin, CD48 is also the ligand for CD2, an important costimulatory D. M. Davis, J. L. Strominger, J. W. Yewdell, and A. Porgador. 2001. Recog- molecule of lymphocyte activation, including NK cells (36). Trig- nition of haemagglutinins on virus-infected cells by NKp46 activates lysis by human NK cells. Nature 409: 1055–1060. gering of CD2 induces calcium flux and tyrosine phosphorylation 18. Pogge von Strandmann, E., V. R. Simhadri, B. von Tresckow, S. Sasse, that leads to increased proliferation, cytotoxicity, IFN-g, and IL-2 K. S. Reiners, H. P. Hansen, A. Rothe, B. Bo¨ll, V. L. Simhadri, P. Borchmann, et release (37). In addition, NK cells are involved in regulating the al. 2007. Human leukocyte antigen-B-associated transcript 3 is released from tumor cells and engages the NKp30 receptor on natural killer cells. Immunity 27: adaptive responses not only indirectly, by secreting cyto- 965–974. kines, but also through the direct killing of immature dendritic cells 19. Brandt, C. S., M. Baratin, E. C. Yi, J. Kennedy, Z. Gao, B. Fox, B. Haldeman, (38, 39). NK cells could also serve as APCs, which express MHC C. D. Ostrander, T. Kaifu, C. Chabannon, et al. 2009. The B7 family member B7-H6 is a tumor cell ligand for the activating natural killer cell receptor NKp30 class II proteins and other costimulatory molecules after activation in humans. J. Exp. Med. 206: 1495–1503. and thus could directly prime T cell responses (3). Because it was 20. Welte, S., S. Kuttruff, I. Waldhauer, and A. Steinle. 2006. Mutual activation of natural killer cells and mediated by NKp80-AICL interaction. Nat. demonstrated that adhesion of T cells to APCs, or NK cells is in- Immunol. 7: 1334–1342. creased after CD2 cross-linking (12, 37), the CD48 expression on 21. Bottino, C., R. Castriconi, D. Pende, P. Rivera, M. Nanni, B. Carnemolla, NK cells might function as another costimulatory molecule when C. Cantoni, J. Grassi, S. Marcenaro, N. Reymond, et al. 2003. Identification of PVR (CD155) and -2 (CD112) as cell surface ligands for the human NK cells serve as APCs for priming of T cell responses. DNAM-1 (CD226) activating molecule. J. Exp. Med. 198: 557–567. 22. Brown, M. H., K. Boles, P. A. van der Merwe, V. Kumar, P. A. Mathew, and A. N. Barclay. 1998. 2B4, the natural killer and T cell immunoglobulin super- Acknowledgments family surface protein, is a ligand for CD48. J. Exp. Med. 188: 2083–2090. We thank Dr. Eric O. Long for providing plasmids for KIR2DL3 and 23. Engel, P., M. J. Eck, and C. Terhorst. 2003. The SAP and SLAM families in Downloaded from KIR2DL3(2YF) receptors. immune responses and X-linked lymphoproliferative disease. Nat. Rev. Immunol. 3: 813–821. 24. Eissmann, P., L. Beauchamp, J. Wooters, J. C. Tilton, E. O. Long, and C. Watzl. Disclosures 2005. Molecular basis for positive and negative signaling by the natural killer cell receptor 2B4 (CD244). Blood 105: 4722–4729. The authors have no financial conflicts of interest. 25. Nakajima, H., and M. Colonna. 2000. 2B4: an NK cell activating receptor with unique specificity and signal transduction mechanism. Hum. Immunol. 61: 39–

43. http://www.jimmunol.org/ References 26. Mandelboim, O., H. T. Reyburn, M. Vale´s-Go´mez, L. Pazmany, M. Colonna, 1. Trinchieri, G. 1989. Biology of natural killer cells. Adv. Immunol. 47: 187–376. G. Borsellino, and J. L. Strominger. 1996. Protection from lysis by natural killer 2. Hanna, J., D. Goldman-Wohl, Y. Hamani, I. Avraham, C. Greenfield, cells of group 1 and 2 specificity is mediated by residue 80 in human histo- S. Natanson-Yaron, D. Prus, L. Cohen-Daniel, T. I. Arnon, I. Manaster, et al. compatibility leukocyte antigen C alleles and also occurs with empty major 2006. Decidual NK cells regulate key developmental processes at the human histocompatibility complex molecules. J. Exp. Med. 184: 913–922. fetal-maternal interface. Nat. Med. 12: 1065–1074. 27. Wilm, M., and M. Mann. 1996. Analytical properties of the nanoelectrospray ion 3. Hanna, J., T. Gonen-Gross, J. Fitchett, T. Rowe, M. Daniels, T. I. Arnon, source. Anal. Chem. 68: 1–8. R. Gazit, A. Joseph, K. W. Schjetne, A. Steinle, et al. 2004. Novel APC-like 28. Mizrahi, S., G. Markel, A. Porgador, Y. Bushkin, and O. Mandelboim. 2007. properties of human NK cells directly regulate T cell activation. J. Clin. Invest. CD100 on NK cells enhance IFNgamma secretion and killing of target cells expressing CD72. PLoS One 2: e818. 114: 1612–1623. 29. Gruda, R., H. Achdout, N. Stern-Ginossar, R. Gazit, G. Betser-Cohen, 4. Stanietsky, N., H. Simic, J. Arapovic, A. Toporik, O. Levy, A. Novik, Z. Levine, I. Manaster, G. Katz, T. Gonen-Gross, B. Tirosh, and O. Mandelboim. 2007. by guest on September 27, 2021 M. Beiman, L. Dassa, H. Achdout, et al. 2009. The interaction of TIGIT with Intracellular cysteine residues in the tail of MHC class I proteins are crucial for PVR and PVRL2 inhibits human NK cell cytotoxicity. Proc. Natl. Acad. Sci. extracellular recognition by leukocyte Ig-like receptor 1. J. Immunol. 179: 3655– USA 106: 17858–17863. 3661. 5. Yokoyama, W. M., and S. Kim. 2006. Licensing of natural killer cells by self- 30. Colonna, M., F. Navarro, T. Bello´n, M. Llano, P. Garcı´a, J. Samaridis, major histocompatibility complex class I. Immunol. Rev. 214: 143–154. L. Angman, M. Cella, and M. Lo´pez-Botet. 1997. A common inhibitory receptor 6. Raulet, D. H. 2006. Missing self recognition and self tolerance of natural killer for major histocompatibility complex class I molecules on human lymphoid and (NK) cells. Semin. Immunol. 18: 145–150. myelomonocytic cells. J. Exp. Med. 186: 1809–1818. 7. Moretta, A., C. Bottino, M. Vitale, D. Pende, C. Cantoni, M. C. Mingari, 31. Yodoi, J., K. Teshigawara, T. Nikaido, K. Fukui, T. Noma, T. Honjo, R. Biassoni, and L. Moretta. 2001. Activating receptors and coreceptors involved M. Takigawa, M. Sasaki, N. Minato, M. Tsudo, et al. 1985. TCGF (IL 2)-re- in human natural killer cell-mediated cytolysis. Annu. Rev. Immunol. 19: 197–223. ceptor inducing factor(s). I. Regulation of IL 2 receptor on a natural killer-like 8. Thomas, M., J. M. Boname, S. Field, S. Nejentsev, M. Salio, V. Cerundolo, cell line (YT cells). J. Immunol. 134: 1623–1630. M. Wills, and P. J. Lehner. 2008. Down-regulation of NKG2D and NKp80 li- 32. Katz, G., G. Markel, S. Mizrahi, T. I. Arnon, and O. Mandelboim. 2001. Rec- gands by Kaposi’s sarcoma-associated herpesvirus K5 protects against NK cell ognition of HLA-Cw4 but not HLA-Cw6 by the NK cell receptor killer cell Ig- cytotoxicity. Proc. Natl. Acad. Sci. USA 105: 1656–1661. like receptor two-domain short tail number 4. J. Immunol. 166: 7260–7267. 9. Tangye, S. G., H. Cherwinski, L. L. Lanier, and J. H. Phillips. 2000. 2B4-me- 33. Nakajima, H., M. Cella, H. Langen, A. Friedlein, and M. Colonna. 1999. Ac- diated activation of human natural killer cells. Mol. Immunol. 37: 493–501. tivating interactions in human NK cell recognition: the role of 2B4-CD48. Eur. J. 10. Sidorenko, S. P., and E. A. Clark. 2003. The dual-function CD150 receptor Immunol. 29: 1676–1683. subfamily: the viral attraction. Nat. Immunol. 4: 19–24. 34. Colonna, M. 2005. Fine-tuning NK cell responses: it’s a family affair. Nat. 11. Tangye, S. G., J. H. Phillips, and L. L. Lanier. 2000. The CD2-subset of the Ig Immunol. 6: 961–962. superfamily of cell surface molecules: receptor-ligand pairs expressed by NK 35. Nichols, K. E., C. S. Ma, J. L. Cannons, P. L. Schwartzberg, and S. G. Tangye. cells and other immune cells. Semin. Immunol. 12: 149–157. 2005. Molecular and cellular pathogenesis of X-linked lymphoproliferative 12. Boles, K. S., S. E. Stepp, M. Bennett, V. Kumar, and P. A. Mathew. 2001. 2B4 disease. Immunol. Rev. 203: 180–199. (CD244) and CS1: novel members of the CD2 subset of the immunoglobulin 36. Veillette, A. 2006. NK cell regulation by SLAM family receptors and SAP-re- superfamily molecules expressed on natural killer cells and other leukocytes. lated adapters. Immunol. Rev. 214: 22–34. Immunol. Rev. 181: 234–249. 37. Holter, W., M. Schwarz, A. Cerwenka, and W. Knapp. 1996. The role of CD2 as 13. Tangye, S. G., J. H. Phillips, L. L. Lanier, and K. E. Nichols. 2000. Functional a regulator of human T-cell cytokine production. Immunol. Rev. 153: 107–122. requirement for SAP in 2B4-mediated activation of human natural killer cells as 38. Ferlazzo, G., M. L. Tsang, L. Moretta, G. Melioli, R. M. Steinman, and C. Mu¨nz. revealed by the X-linked lymphoproliferative syndrome. J. Immunol. 165: 2932– 2002. Human dendritic cells activate resting natural killer (NK) cells and are 2936. recognized via the NKp30 receptor by activated NK cells. J. Exp. Med. 195: 14. Bauer, S., V. Groh, J. Wu, A. Steinle, J. H. Phillips, L. L. Lanier, and T. Spies. 343–351. 1999. Activation of NK cells and T cells by NKG2D, a receptor for stress-in- 39. Spaggiari, G. M., R. Carosio, D. Pende, S. Marcenaro, P. Rivera, M. R. Zocchi, ducible MICA. Science 285: 727–729. L. Moretta, and A. Poggi. 2001. NK cell-mediated lysis of autologous antigen- 15. Arnon, T. I., H. Achdout, O. Levi, G. Markel, N. Saleh, G. Katz, R. Gazit, presenting cells is triggered by the engagement of the phosphatidylinositol 3- T. Gonen-Gross, J. Hanna, E. Nahari, et al. 2005. Inhibition of the NKp30 ac- kinase upon ligation of the natural cytotoxicity receptors NKp30 and NKp46. tivating receptor by pp65 of human cytomegalovirus. Nat. Immunol. 6: 515–523. Eur. J. Immunol. 31: 1656–1665.