Cell Adhesion Molecule 1 NK Killing Via Interaction with CEA-Related Carcinoembryonic Antigen

Home , Carcinoembryonic antigen, Cell adhesion molecule

Carcinoembryonic Antigen (CEA) Inhibits NK Killing via Interaction with CEA-Related Cell Adhesion Molecule 1

This information is current as Noam Stern, Gal Markel, Tal I. Arnon, Raizy Gruda, Henry of October 6, 2021. Wong, Scott D. Gray-Owen and Ofer Mandelboim J Immunol 2005; 174:6692-6701; ; doi: 10.4049/jimmunol.174.11.6692 http://www.jimmunol.org/content/174/11/6692 Downloaded from

References This article cites 30 articles, 15 of which you can access for free at: http://www.jimmunol.org/content/174/11/6692.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

*average by guest on October 6, 2021

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

Carcinoembryonic Antigen (CEA) Inhibits NK Killing via Interaction with CEA-Related Cell Adhesion Molecule 11

Noam Stern,* Gal Markel,* Tal I. Arnon,* Raizy Gruda,* Henry Wong,† Scott D. Gray-Owen,† and Ofer Mandelboim2*

The NK killing activity is regulated by activating and inhibitory NK receptors. All of the activating ligands identified so far are either viral or stress-induced proteins. The class I MHC proteins are the ligands for most of the inhibitory NK receptors. However, in the past few years, several receptors have been identified that are able to inhibit NK killing independently of class I MHC recognition. We have previously demonstrated the existence of a novel inhibitory mechanism of NK cell cytotoxicity mediated by the homophilic carcinoembryonic Ag (CEA)-related cell adhesion molecule 1 (CEACAM1) interactions. In this study, we demonstrate that CEACAM1 also interacts heterophilically with the CEA protein. Importantly, we show that these heterophilic interactions of CEA and CEACAM1 inhibit the killing by NK cells. Because CEA is expressed on a wide range of carcinomas and Downloaded from commonly used as tumor marker, these results represent a novel role for the CEA protein enabling the escape of tumor cells from NK-mediated killing. We further characterize, for the first time, the CEACAM1-CEA interactions. Using functional and binding assays, we demonstrate that the N domains of CEACAM1 and CEA are crucial but not sufficient for both the CEACAM1- CEACAM1 homophilic and CEACAM1-CEA heterophilic interactions. Finally, we suggest that the involvement of additional domains beside the N domain in the heterophilic and homophilic interactions is important for regulating the balance between cis and trans interactions. The Journal of Immunology, 2005, 174: 6692–6701. http://www.jimmunol.org/

he human carcinoembryonic Ag (CEA)3 family is com- cruitment of the Src homology 2 domain-containing tyrosine posed of 29 genes tandemly arranged on chromosome phosphatase-1 and -2 (5–7). The CEACAM1 protein is expressed T 19q13.2. Based on nucleotide homologies, these genes are on a variety of immune cells including monocytes, granulocytes, classiﬁed into two major subfamilies, the CEACAM and the preg- activated T cells, B cells, and NK cells (1, 2, 8). nancy-speciﬁc glycoprotein subgroups (1, 2). The CEACAM- Our laboratory has recently demonstrated that CEACAM1 ho- encoded proteins include CEA, CEA-related cell adhesion mole- mophilic interactions inhibit NK killing independently of MHC cule (CEACAM)1, CEACAM3, CEACAM4, CEACAM6, class I recognition (8). The CEACAM1 interactions are probably

CEACAM7, and CEACAM8 (2). The CEACAM family belongs important in some cases of metastatic melanoma, as increased by guest on October 6, 2021 to the Ig superfamily. Structurally, each of the human CEACAMs CEACAM1 expression was observed on NK cells derived from contain one N-terminal domain that includes 108–110 aa and is such patients compared with healthy donors (8). A major role for homologous to Ig variable domains, followed by a different num- CEACAM1 on NK cells was also demonstrated in TAP2-deficient ber (zero to six) of Ig C2-type constant-like domains (3). The patients and during pregnancy (9, 10). In addition it was demon- CEACAM proteins can interact homophilically and heterophili- strated that CEACAM1 engagement by the Opa proteins of Neis- cally with each other (4). seria gonorrhoeae inhibits T cell activation and proliferation (5). CEACAM1 is a unique protein within this family because it It is well established that the CEACAM1 homophilic interactions contains an ITIM motif in its cytoplasmic domain (2). A similar are mediated through the N domain, and several publications have ITIM is present in various inhibitory receptors that function to identified specific amino acids that are involved in these interac- antagonize kinase-dependent signaling cascades initiated by lym- tions (11, 12). It has also been previously reported that CEACAM1 phocyte activation. This inhibitory effect is triggered by phosphor- interacts with other CEACAM protein family members, such as ylation of tyrosine residues within the ITIM, which results in re- CEA (12, 13). Our laboratory has previously demonstrated the dichotomy of CEACAM family members by showing that the recognition of *Lautenberg Center for General and Tumor Immunology, Hebrew University Hadas- sah Medical School, Jerusalem, Israel; and †Department of Medical Genetics and CEACAM1 is determined by the presence of R and Q residues at Microbiology, University of Toronto, Toronto, Ontario, Canada positions 43 and 44, respectively (12). Importantly, the CEA pro- Received for publication September 14, 2004. Accepted for publication March tein contains these RQ residues, suggesting that these residues are 15, 2005. involved in binding to CEACAM1. The CEA protein is of great The costs of publication of this article were defrayed in part by the payment of page interest, because it is expressed on a wide range of carcinomas and charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. is commonly used as a tumor marker in the prognosis and man- 1 This work was supported by research grants from the Israel Science Foundation, the agement of many types of cancer (14, 15). CEA consists of an Ig Israel Cancer Association (20042031-B), the Israel Cancer Research Foundation, and V-like N-terminal domain followed by three pairs of Ig C2-like the United States-Israel Binational Science Foundation. domains (denoted AxBx) which are terminated by a hydrophobic 2 Address correspondence and reprint requests to Dr. Ofer Mandelboim, Lautenberg domain, that is processed to allow the addition of glycophosphati- Center for General and Tumor Immunology, Hebrew University Hadassah Medical School, Jerusalem, 91120, Israel. E-mail address: [email protected] dyl-inositol membrane anchor (2, 16). In addition to the hetero- 3 Abbreviations used in this paper: CEA, carcinoembryonic Ag; CEACAM, CEA- philic interaction of CEA with CEACAM1, CEA binds homophili- related cell adhesion molecule; mIL-2, murine IL-2. cally to itself. The CEA-CEA interactions were shown to be

Ј mediated by double reciprocal bonds between the N and A3B3 F(ab )2 of goat anti-human IgG (Jackson ImmunoResearch Laboratories). domains of antiparallel CEA molecules on the opposed cell (17). Incubation was performed for 1 h and cells were analyzed by flow cytom- However, until now, nothing was known about the nature of the etry with a FACScan (BD Immunocytometry Systems). CEA interactions with CEACAM1 or the functional consequences Generation of 721.221 cells expressing truncated CEA and of these interactions. CEACAM1 proteins In this work, we demonstrate that CEACAM1 binding to CEA is direct and functional. The heterophilic interactions between For the generation of the truncated CEACAM proteins, we amplified by CEA and CEACAM1 inhibit the killing of NK cells that express PCR the leader peptide of CEA using 5Ј primer GATCCACTATCAGAG Ј the CEACAM1 protein. By using various combinations of the dif- CAACCCCAAC (including HindIII restriction site) and the 3 primer GAATTCGGCAGTGGTGGGCGGG (including EcoRI restriction site) ferent CEA domains and by using functional assays, we further The PCR-generated fragment was cloned into pCDNA3 mammalian ex- demonstrate that the N domains of both CEACAM1 and CEA are pression vector. Next, we amplified by PCR the relevant CEA domains crucial but not sufficient for the heterophilic interactions. Further- using different 5Ј primers (including EcoRI restriction site) and the same 3Ј more, we show that each pair of A and B domains within CEA primer CCGCTCGAGCTATATCAGAGCAACCCCAAC (including XhoI restriction site). The 5Ј primer CEA D2-D7 (CEA ⌬N), CCGGAATT contributes to the intensity of the CEA interaction with CAAGCCCTCCATCTCCAGCA; 5Ј primer CEA D4-D7, CCGGAATTC CEACAM1. Finally, we show that the N domain of CEACAM1 AAACCCTTCATCACCAGCAAC; 5Ј primer CEA D5-D7, CCGGAAT plays a critical role in the functional CEACAM1 homophilic TCGACGACCCCACCATTTCCC. interactions. We also amplified by PCR the CEACAM1 D2-D3 (CEACAM1 ⌬N) domains using 5Ј primer CCGGAATTCAAGCCCTCCATCTCCAGCA (including EcoRI restriction site) and 3Ј primer CCGCTCGAGTTACT GCTTTTTTACTTCTGA. The various PCR fragments were then cloned Materials and Methods Downloaded from Cells into the pcDNA3 containing the leader peptide of CEA and stably transfected into the 221 cell line. The cell lines used in this work were: the MHC class I-negative 721.221 (221), 221 cells stably transfected with the CEACAM1 protein (221/ Cytotoxicity assays and ELISA CEACAM1), and 221 expressing the CEA protein (221/CEA) (12). The murine thymoma BW cell line which lacks expression of ␣- and ␤-chains The cytotoxic activity of NK and YTS cells against the various targets was of the TCR. The generation of the BW cell line expressing the extracellular assessed in 5-h 35S release assays as previously described (18). In all pre- portion of the human CEACAM1 protein fused to mouse ␨-chain (BW/ sented cytotoxicity assays, the spontaneous release was Ͻ 25% of maximal http://www.jimmunol.org/ CEACAM1-␨) was previously described (10). The generation of YTS NK release. In experiments in which Abs were included, the final Abs concen- tumor expressing the CEACAM1 protein (YTS/CEACAM1) and the tration was 10 ␮g/ml, or 40 ␮l/ml in those cases where rabbit polyclonal CEACAM1 in which the cytoplasmic tail of the molecule was truncated Abs were used. The measurement of IL-2 production resulting from the (and therefore does not include the ITIMs; YTS/CEACAM1Trunc) was BW/CEACAM1 interactions with different cells was performed as previ- previously described (8). Primary NK cells were isolated from peripheral ously described (10). For ELISA, ELISA plates were coated with 0.1 ␮g/ blood lymphocytes, using the human NK isolation kit and the autoMACS well of purified CEA (Biodesign International) followed by incubation instrument (Miltenyi Biotec). For the enrichment of CEACAM1-negative with 5 ␮g of the relevant fusion protein. Bound proteins were detected NK cells, isolated NK cells were further purified by depletion of CD16- using AP-conjugated second mAb. positive NK cells, using the anti-CD16 mAb B73.1.1 and the autoMACS instrument. NK cells were grown in culture as previously described (18). Results by guest on October 6, 2021 Antibodies The heterophilic interactions between CEACAM1 and CEA are functional The Abs used in this work were mAb Kat4c (DakoCytomation), directed against CEACAM1, 5, 6, 8, and rabbit polyclonal anti-CEACAM (Dako- We have previously demonstrated that the entire family of the Cytomation). Anti-CD99 mAb 12E7 was used as control. CEACAM proteins can be divided into two groups based on the presence or absence of the RQ residues in the N domain (12). In Generation of Ig fusion protein and FACS staining addition, we and others demonstrated that CEACAM1 interacts The generation of LIR1-Ig and CD99-Ig fusion proteins was previously with CEA (12, 13). However, the functional consequences of the described (10, 19). The sequence encoding the full extracellular portion CEACAM1-CEA interactions were never investigated. For mea- of CEA protein was amplified by PCR using the following primers: 5Ј primer GGATCCACTATCAGAGCAACCCCAAC (including HindIII suring direct binding of CEACAM1 and CEA to the purified CEA restriction site) and the 3Ј primer GGGATCCGATGCAGAGACTGT protein, the extracellular portion of the CEACAM1 and CEA were GATGCT (including BamHI restriction site). A silent mutation, adenine fused to the Fc portion of human IgG1, thus creating 887 guanidine (no change in glycine 295) was performed by site-di- CEACAM1-Ig and CEA-Ig, respectively. Direct binding of rected mutagenesis to cancel the BamHI site in the amplified sequence. CEACAM1-Ig and CEA-Ig to purified CEA was tested in ELISA. The sequences encoding the truncations of CEA protein were amplified by PCR using the same 5Ј primer and the various 3Ј primers (including As demonstrated in Fig. 1A, heterophilic binding of CEACAM1-Ig BamHI restriction site) and the full CEA-Ig construct containing the and a stronger homophilic binding of CEA-Ig to the purified CEA adenine 887 guanidine was used as template. The 3Ј primer of CEA were observed. The control LIR1-Ig fusion protein showed only D5-D1-Ig, GGGATCCGCAGAGACTGTGATTGTCTT; 3Ј primer of weak binding (Fig. 1A). CEA D3-D1-Ig, GGGATCCGGCTCTGCATAGACTGTGAT; 3Ј primer of CEA D2-D1-Ig, GGGATCCAGCTCCGGGTATACGCG. We next examined the heterophilic and homophilic interactions These PCR-generated fragments were cloned into the mammalian ex- between CEACAM1 and CEA in the context of mammalian cells pression vector containing the Fc portion of human IgG1. The fusion pro- by using cell transfectants. 221 cells were transfected with teins were periodically analyzed for degradation by SDS-PAGE protein CEACAM1 cDNA (221/CEACAM1) and with the CEA cDNA gels. The constructs containing only the N domains of CEACAM1 and (221/CEA) and the NK tumor cell line YTS was transfected with CEA fused to Fc portion of human IgG1 were generated in the laboratory of Dr. S. Gray-Owen. Sequencing of the constructs revealed that cDNA of CEACAM1 cDNA (YTS/CEACAM1). Expression levels were all Ig fusion proteins were in frame with the human Fc and were identical monitored with Kat4c mAb (Fig. 1, B–D). Homophilic and het- with the reported sequences. The production of the various fusion proteins erophilic binding of the CEACAM1-Ig fusion protein was ob- by COS-7 cells and purification on protein G column were previously served to 221/CEACAM1 and 221/CEA, respectively (Fig. 1, B described (20). To assay for the fusion protein binding, various cells were incubated with 50 ␮g/ml fusion protein for2honice. The cells were and C). Similarly, homophilic and heterophilic binding of the washed and incubated with Fc fragment-specific (minimal cross-reaction to CEA-Ig fusion protein was observed to 221/CEA and 221/ bovine, horse, and mouse serum proteins), PE-conjugated affinity-purified CEACAM1, respectively (Fig. 1, B and C). A weak homophilic 6694 CEA INHIBITS NK KILLING VIA INTERACTION WITH CEACAM1 Downloaded from http://www.jimmunol.org/

FIGURE 1. Heterophilic binding of CEACAM1-Ig to CEA. A, CEACAM1-Ig and CEA-Ig directly interacts with purified CEA protein. ELISA plates were coated with 0.1 ␮g/well purified CEA and than incubated with 5 ␮g of the relevant fusion protein. The figure represents OD absorbance (405 nm) p Ͻ 0.001. Shown is one representative experiment of three performed. B–D, Binding of CEACAM1 by guest on October 6, 2021 ,ءءء .after reducing nonspecific binding to the plate and CEA to 221 and YTS transfectants. The expression levels of 221/CEACAM1 (B, left panel), 221/CEA (C, left panel), and YTS/CEACAM1 (D, left panel) were monitored with Kat4c mAb (open histogram). Binding of CEACAM1 and CEA to these transfectants was assessed by FACS staining using CEACAM1-Ig and CEA-Ig fusion proteins and the control CD99-Ig (open histogram). The background (gray histogram) is the corresponding staining of parental 221 and YTS cells. Shown is 1 representative experiment of 10 performed. E, Binding of CEACAM1 and CEA to CEACAM1-positive NK clone. The expression level of CEACAM1 on the NK clone was monitored with Kat4c mAb (E, left panel). Binding of CEACAM1 and CEA was assessed by FACS staining using CEACAM1-Ig and CEACAM5-Ig fusion proteins and the control CD99-Ig (open histogram). The background is the staining of the same NK clone with the secondary PE-conjugated goat anti-human IgG Abs only (gray histogram). Shown is one representative clone of five.

and heterophilic binding of CEACAM1-Ig and CEA-Ig was ob- those cells elicited the secretion of murine IL-2 (mIL-2) (10). The served to YTS/CEACAM1, respectively, probably due to rela- BW parental cells and the BW/CEACAM1-␨ were coincubated tively low expression levels of the CEACAM1 protein (Fig. 1D). with no cells, or with irradiated 221, 221/CEACAM1, and 221/ The control CD99-Ig fusion protein did not stain any of the trans- CEA cells for 48 h either with anti-CEACAM polyclonal blocking fectants tested (Fig. 1, B–D). Abs or with anti-human GST (hGST) Abs as a control. Significant We next examined the heterophilic and homophilic interactions amounts of mIL-2 were detected in the supernatant of BW/ between CEACAM1 and CEA with NK cells that endogenously CEACAM1-␨ cells coincubated with no cells or 221 cells in the express CEACAM1. NK cells were isolated from healthy donors, presence of the control anti-hGST Abs. Adding the anti- activated NK clones were grown and stained with Kat4c mAb CEACAM1 blocking Abs significantly reduced the mIL-2 secre- (representative CEACAM1-positive clone is shown in Fig. 1E). tion, demonstrating that the observed mIL-2 secretion was due to Importantly, homophilic and heterophilic binding of the homophilic interaction between different BW/CEACAM1-␨ cells CEACAM1-Ig and CEA-Ig fusion proteins were observed to the (Fig. 2A). Significantly higher amounts of mIL-2 were detected in CEACAM1-positive NK clones (Fig. 1E). Thus both the CEA and the supernatant of BW/CEACAM1-␨ cells coincubated with 221/ the CEACAM1 proteins are able to interact with CEACAM1 on CEACAM1 and with 221/CEA cells compared with BW/ the surface of various cells. CEACAM1-␨ cells coincubated with no cells or with 221 cells in To test the functional relevance of the heterophilic interactions the presence of control Abs. Adding the anti-CEACAM1 blocking between CEACAM1 and CEA we first used the BW cell system. Abs partially reduced the mIL-2 secretion demonstrating that the BW cells expressing the extracellular potion of CEACAM1 fused observed elevated secretion of mIL-2 resulted from CEACAM inter- to mouse ␨-chain (BW/CEACAM1-␨) were generated as previ- actions between the BW/CEACAM1-␨ cells and 221 cells expressing ously described (10). Engagement of the CEACAM1 protein on either CEACAM1 or CEA. (Fig. 2A). No secretion of mIL-2 was The Journal of Immunology 6695 Downloaded from http://www.jimmunol.org/ by guest on October 6, 2021

FIGURE 2. Functional interaction between CEA and CEACAM1. A, Engagement of CEACAM1-␨ by CEA and CEACAM1 induces IL-2 secretion. The mIL-2 secretion from BW/CEACAM1-␨ cells was measured by ELISA. BW/CEACAM1-␨ cells were coincubated for 48 h with no cells, irradiated 221, 221/CEACAM1, and 221/CEA cells. Cells were incubated either with rabbit anti-CEACAM polyclonal in a final concentration of 20 ␮l/ml (open column), .p Ͻ 0.005. Shown is one representative experiment of four performed ,ءء .or with anti-hGST (gray column). The y-axis is OD at wave length of 650 nm B–D, The interaction between CEA and CEACAM1 inhibits NK cytotoxicity. B, A representative CEACAM1-positive NK clone stained by Kat4c mAb Ј (open histogram). The background (gray histogram) is the staining of the secondary reagent FITC-conjugated goat anti-mouse F(ab ) 2 Abs. C, The CEACAM1-positive NK clone was tested in killing assays against the indicated target cells in the indicated E:T ratio. The NK clone was incubated with rabbit anti-GST polyclonal Abs in a final concentration of 20 ␮l/ml to block FcR on the NK clone. Target cells were incubated either with rabbit anti-CEACAM polyclonal Abs in a final concentration of 20 ␮l/ml, or with anti-CD99 mAb in a final concentration of 5 ␮g/ml. Shown is one representative NK clone of three. D, Killing of 221 and 221/CEA by various CEACAM1-positive NK clones. Killing assays were performed as described in C. The E:T p Ͻ 0.001. Figure shows 4 representative clones of 15 clones tested. E, CEACAM1 homophilic interactions do not ,ءءء ;p Ͻ 0.005 ,ءء .ratio was 5:1 significantly inhibit the overall NK cytotoxicity. YTS/CEACAM1 cells and CEACAM1-positive NK clone were tested in killing assays against 221 cells in E:T ratio of 5:1. The effector cells were incubated either with rabbit anti-CEACAM polyclonal Abs (open column) or with anti-hGST polyclonal Abs (gray column) in a final concentration of 20 ␮l/ml. Shown is one clone of five clones tested. observed when the parental BW cells were used (data not shown). CEACAM1-positive NK clones (representative clone is shown in These results clearly indicate that CEACAM1 and CEA proteins het- Fig. 2B) were assayed for killing against 221, 221/CEACAM1, and erophilically interact with each other and that these interactions are 221/CEA cells in various E:T ratios either with anti-CEACAM functional. blocking polyclonal Abs or with anti-CD99 Ab as a control (Fig. We have previously reported that CEACAM1 expression on NK 2C). As was previously reported (8, 10), inhibition of NK killing cells plays a major role in inhibition of NK cytotoxicity via the was observed when 221/CEACAM1 cells were used and this in- homophilic CEACAM1 interactions (8–10, 12). The CEA protein hibition could be reversed by the anti-CEACAM Abs (Fig. 2C). is an important tumor marker expressed in colorectal and other Importantly, a significant inhibition of CEACAM1-positive NK carcinomas (15). Because CEA is expressed on tumor cells and as clones was also observed when 221/CEA cells were used (Fig. shown above, is able to interact heterophilically with CEACAM1 2C). This inhibition was the result of CEACAM1 and CEA het- (Fig. 1), it was of a particular importance to test whether the CEA erophilic interactions, as anti-CEACAM Abs abrogated this effect protein can also inhibit the killing by CEACAM1-positive NK (Fig. 2C). In healthy people CEACAM1 is expressed on CD16- cells. NK cells were isolated from healthy donors; activated NK negative NK cells (21). Indeed, the CEACAM1-positive NK clones were grown and stained for CEACAM1 expression. clones tested in this study did not express CD16 (data not shown). 6696 CEA INHIBITS NK KILLING VIA INTERACTION WITH CEACAM1

CD16 is the major FcR on NK cells involved in ADCC (22). Nev- CEACAM1 and with 221/CEA compared with the 221 parental ertheless, to prove beyond any doubt that the restoration of NK cells (Fig. 3A). The moderate inhibition correlated with the low killing mediated by using the polyclonal anti-CEACAM Abs was CEACAM1 expression on the YTS cells (Fig. 1D). Importantly no not due to ADCC phenomenon; the NK clones were incubated inhibition of lysis by YTS/CEACAM1 trunc cells was observed with polyclonal anti-hGST Abs before killing to block any avail- (Fig. 3B). These results indicate that the CEACAM1 inhibitory able FcR (Fig. 2, C and D). signal is probably transduced via the ITIM sequences in its The CEA-mediated inhibition was variable between the different cytoplasmic tail. NK clones tested (Fig. 2D) probably because the killing of target cells is determined by a balance between the inhibitory and activating signals. However, as can be seen in Fig. 2, C and D,in The N domain of both CEACAM1 and CEA is critical but not all clones expressing CEACAM1, the CEA-mediated inhibition sufﬁcient for the heterophilic and homophilic binding was signiﬁcant and was abrogated by the addition of anti- The CEACAM-related proteins share a common basic structure of CEACAM Abs. several sequential Ig like domains beginning with an Ig V-like N

We demonstrated above that CEACAM1 homophilic interac- domain and proceeding with different number of Ig C2-like do- tions between opposing BW/CEACAM1-␨ cells are functional, mains. It was postulated before that CEACAM1 interacts ho- leading to the secretion of mIL-2 without the presence of target mophilically in “head-to-head” N domains interactions (11). Ad- cells (Fig. 2A). Therefore, it can be suggested that NK cells ex- ditionally, it was reported that the CEA protein interacts pressing CEACAM1 will undergo homophilic interactions with homophilically through “head-to-tail” interactions between the Ig each other and that these interactions would result in a general V-like N domain and the Ig C2-like domain (17). However, the inhibition of NK cytotoxicity. To elucidate whether homophilic nature of the CEACAM1-CEA interactions was not investigated. Downloaded from interactions between CEACAM1-positive NK cells affect the over- To determine whether the Ig V-like N domains are crucial for the all cytotoxicity of these cells, we incubated YTS/CEACAM1 and binding of CEACAM1 to CEA, we generated a mutated construct CEACAM1-positive NK clones either with anti-CEACAM Abs or of CEACAM1 and CEA that does not contain the N domain with anti-hGST Abs as a control. The cells were then tested in (CEACAM1 ⌬N) and (CEA ⌬N), respectively. The 221 cells were killing assays against 221 parental cells. No significant increase in stably transfected with those constructs and tested for expression the lysis of 221 cells by YTS/CEACAM1 cells was observed when using the anti-CEACAM polyclonal Abs (Fig. 4, A and B). Next, http://www.jimmunol.org/ anti-CEACAM Abs were added (Fig. 2E). In addition, in most of we tested the binding of CEACAM1-Ig to 221/CEA and 221/CEA the CEACAM1-positive NK clones tested, the addition of the ⌬N. Remarkably, the deletion of the N domain from CEA abol- blocking CEACAM Abs had little or no significant effect on the ished the heterophilic binding, as CEACAM1-Ig did not bind to observed killing (one representative clone is shown in Fig. 2E). 221/CEA ⌬N (Fig. 4A). The homophilic binding of CEA-Ig was Therefore, we conclude that homophilic interactions between NK also abolished when 221/CEA ⌬N were used (data not shown). We cells expressing CEACAM1 do not effect the overall cytotoxicity then performed the reciprocal experiments and tested the CEA-Ig of these cells. and CEACAM1-Ig for binding to 221/CEACAM1 and 221/ The CEACAM1 protein contains two ITIM sequences located CEACAM1 ⌬N cells. Importantly, the deletion of the N domain of by guest on October 6, 2021 within its cytoplasmic tail. These motifs have been shown to me- CEACAM1 abolished the homophilic binding of CEACAM1-Ig diate inhibitory functions in epithelial cells and T cells (5–7, 23). and the heterophilic CEA-Ig binding (Fig. 4B). These results in- To demonstrate that the ITIM motifs are important for the ob- dicated that the N domains of both CEA and CEACAM1 proteins served inhibition of NK cell cytotoxicity, we used the YTS cells. are critically involved in their heterophilic and homophilic We have previously transfected this cell line either with interactions. CEACAM1 protein (YTS/CEACAM1) or with CEACAM1 pro- Our next question was whether the N domain by itself would be tein in which the ITIM motifs were truncated (YTS/CEACAM1 sufficient for the heterophilic and homophilic interactions. There- trunc) (8). The various YTS transfectants were tested in killing fore, we constructed two additional fusion proteins, containing ei- assays against the 221 transfectants. In agreement with the above ther the CEACAM1-N domain (CEACAM1-N-Ig) or the CEA-N results, moderate, but significant inhibition of YTS/CEACAM1 domain (CEA-N-Ig). Surprisingly the CEACAM1-N-Ig and CEA- killing was observed when cells were incubated with 221/ N-Ig did not bind 221/CEACAM1 or 221/CEA cells (Fig. 4C).

FIGURE 3. The ITIM motifs are important for the observed inhibition of NK cell cytotoxicity. A and B, YTS cells expressing the CEACAM1 protein (YTS/CEACAM1) or truncated CEACAM1 (YTS/CEACAM1 trunc) were tested in killing assays against 221 and 221 transfectants. The E:T cell ratio was .p Ͻ 0.005. Shown is one experiment of two performed ,ءء .5:1 The Journal of Immunology 6697 Downloaded from http://www.jimmunol.org/ by guest on October 6, 2021

FIGURE 4. The N domains are crucial but not sufficient for the CEA and CEACAM1 recognition. A, Deletion of the N domain from CEA abolished CEACAM1 binding. Shown is the staining of 221/CEA and 221/CEA ⌬N transfectants with anti-CEACAM polyclonal Abs or with CEACAM1-Ig fusion protein (open histogram). The corresponding staining of 221 parental cells was used as background (gray histogram). B, Deletion of the N domain from CEACAM1 abolished CEA and the CEACAM1 binding. The figure shows staining with anti-CEACAM polyclonal Abs, with CEA-Ig and with CEACAM1-Ig as in A. C, The N domain of both CEA and CEACAM1 is not sufficient for binding. Shown is the staining of 221/CEA (upper two histograms) and 221/CEACAM1 (lower two histograms) with the indicated CEACAM1-N-Ig and CEA-N-Ig fusion proteins (open histogram). The corresponding staining of 221 parental cells was used as background (gray histogram). D, The N domains of both CEA and CEACAM1 are unable to directly bind the purified CEA protein. ELISA plates were coated with 0.1 ␮g/well of purified CEA and then incubated with 5 ␮g of the indicated fusion p Ͻ 0.01. For all parts, one representative experiment ,ء .protein. Shown is the OD absorbance (405 nm) after reducing nonspecific binding to the plate is shown of three performed.

Furthermore, even when measuring direct binding of CEACAM1- tional interactions and whether the N domain is crucial for these N-Ig and CEA-N-Ig to purified CEA protein no binding was de- interactions. To test the involvement of different domains, we tected (Fig. 4D). It is unlikely that the binding was abolished due generated truncated constructs of CEACAM1 and CEA as il- to the possible altered conformation of the N domains because we lustrated in Fig. 5A. The 221 cells were stably transfected with used the entire domain, which by definition can fold indepen- these various constructs and tested for expression using Kat4c dently. Furthermore, when domain A and B were included in the mAb (Fig. 5B). Ig fusion protein, binding was restored (see Fig. 7B) and finally, The interactions between the various truncated proteins ex- the CEACAM3 contains only the N domain. These combined re- pressed on 221 cells were first examined using the BW transfected sults suggest that the N domains on both CEACAM1 and CEA are system. Significant amounts of mIL-2 were detected in the super- critical but not sufficient for heterophilic and homophilic binding. natant of BW/CEACAM1-␨ cells coincubated with 221/ CEACAM1 and with 221/CEA cells (Fig. 6A). Importantly, min- The functional CEACAM1-CEA heterophilic interactions and imal amounts of mIL-2 were detected in supernatants of BW/ CEACAM1 homophilic interactions are crucially dependent on CEACAM1-␨ coincubated with 221/CEACAM1 ⌬N, 221/CEA the N domain ⌬N, 221/CEA D4-D7, and with 221/CEA D5-D7. Therefore, de- As shown above (Fig. 4), the N domains of CEA and CEACAM1 letion of CEACAM1 N domain and CEA N domain was sufficient are crucial, but not sufficient, for the heterophilic binding between to abolish the functional homophilic and heterophilic interactions CEA and CEACAM1 and the homophilic binding of CEACAM1. with BW/CEACAM1-␨ cells. Murine IL-2 was not detected when Therefore, we examined which domains are involved in the func- BW parental cells were used (data not shown). 6698 CEA INHIBITS NK KILLING VIA INTERACTION WITH CEACAM1

FIGURE 5. Various truncations of CEA and CEACAM1 and their expression on 221 cells. A, Sche- matic illustration of the CEACAM1 and various CEA truncations. The CEA protein contains one Ig V-like domain (N) and six Ig C2-like domains. The six Ig C2-like domains are classiﬁed based on sequence similarity to the A and B domains (2). For convenience, we numbered the various domains of CEA starting from the N domain (D1) and ending with the B3 domain (D7). The CEACAM1 isoform used in our work contains one Ig V-like domain

(N) and two Ig C2-like, A1 and B1. The generation of the CEA and CEACAM1 truncations is described in Mate- rials and Methods, and the various truncations are pre- sented in the ﬁgure. B, Expression levels of the various CEACAM1 and CEA truncations on 221 cells. 221 transfectants were stained with Kat4c mAb (open histogram). The Kat4c mAb staining of 221 parental cells was used as background (gray histogram). Downloaded from

The BW results were further conﬁrmed using functional killing N domain, none of the other CEA domains were able to confer assays. CEACAM1-positive NK clones were tested in killing as- inhibition (Fig. 6B). says against the various 221 transfectants. Inhibition of NK killing was observed when 221/CEACAM1 cells 221/CEA cells were The heterophilic binding of CEACAM1 and CEA is dependent used (Fig. 6B). Importantly and in agreement with the above bind- on the presence of the A, B, and N domains in CEA http://www.jimmunol.org/ ing results (Fig. 4A), no inhibition was observed when 221/ Because we could not detect any binding between CEACAM1 and CEACAM1 ⌬N were used. Therefore, the deletion of the N do- CEA when the N domains were removed (Fig. 4A), we generated main is enough to abolish any functional interactions between two truncated CEA fusion proteins containing the CEA V Ig-like N

CEACAM1 proteins. In addition, no inhibition was observed when domain and different number of C2 Ig like domains as illustrated in 221/CEA ⌬N, 221/CEA D4-D7, and 221/CEA D5-D7 cells were Fig. 7A. used. Therefore, although we demonstrate that other domains be- These fusion proteins enabled us to examine which domains, in sides the N domain are involved in the CEACAM1 binding to addition to the N domain, are important for the heterophilic inter-

CEA, the CEA N domain is crucial, because in the absence of the action between CEACAM1 and CEA. We tested the 221/ by guest on October 6, 2021

FIGURE 6. The interaction between CEACAM1 and CEA is functional and exclusively dependent on the N domain. A, The N domain (D1) of CEA and CEACAM1 is important for engagement of BW/CEACAM1- ␨. Mouse IL-2 secretion by BW/ CEACAM1-␨ cells was measured by ELISA. BW/CEACAM1-␨ cells were coincubated for 48 h with various irradiated 221 transfectants. The y-axis is the optic density at wave length of p Ͻ 0.01. Shown is one ,ء .nm 650 representative experiment of three performed. B, The N domain of CEA and CEACAM1 is crucial for the inhibition of CEACAM1-positive NK clones. CEACAM1-positive NK clones were tested in killing assays against 221 and 221 transfectants. The E:T ratio was -p Ͻ 0.001. Shown is 1 repre ,ءءء .5:1 sentative clone of 30 clones tested. The Journal of Immunology 6699

FIGURE 7. The heterophilic binding of CEACAM1 and CEA is dependent on the presence of A, B, and N domains. A, Schematic illustration of the soluble proteins used in this work. The generation of the various fusion proteins is described in Materials and Methods. Shown are the different CEA extracellular domains fused to the Fc portion of human IgG1. The names of the fusion proteins indicate which domains were left in the soluble protein. B, Staining of 221/CEACAM1 with various fusion proteins as indicated in each histogram (open histogram). The corresponding staining of 221 parental cells was used as background (gray histogram). Shown is one representative staining of four performed. Downloaded from

CEACAM1 for binding by the various fusion proteins. The full onstrates the possible importance of CEA-CEACAM1 interactions length CEA-Ig strongly recognized the 221/CEACAM1 with mean as a defense mechanism enabling tumors to escape NK cell attack. fluorescence intensity of 127. Deletion of D6 and D7 domains (A3, In addition the CEACAM1 protein is expressed on a variety of B3) in CEA-D5-D1-Ig led to a small decrease in the binding ef- other immune cells, including monocytes, granulocytes, activated http://www.jimmunol.org/ ficiency to the 221/CEACAM1 cells (Fig. 7B). The CEA-D3- T cells, and B cells (1, 2, 8). Therefore, it is possible that the D1-Ig containing only the A1, B1 domains together with the N CEACAM1-CEA interactions might be important for tumor es- domain still binds the 221/CEACAM1 cells, but with lower mean cape not only from NK killing but also from other immune cells. fluorescence intensity demonstrating that these domains are suffi- It can be suggested that NK cells expressing CEACAM1 will cient for the heterophilic binding. Remarkably, when the D3 (B1) undergo homophilic interactions with each other and that these domain was deleted as in CEA-D2-D1-Ig, the binding to 221/ interactions would result in general inhibition of NK cytotoxicity. CEACAM1 cells was abolished (Fig. 7B). These results reveal that Similarly it can be argued that interactions between different killer together with the N domain, each pair of the A and B domains of cell Ig-like receptors expressed on a particular NK cell with MHC by guest on October 6, 2021 the CEA protein is probably sufficient for binding to CEACAM1 class I molecules expressed on opposed NK cell would result in and that the presence of all A and B domains in the native form of inhibition of NK killing. It was previously demonstrated that in the the protein increases the intensity of the binding of CEA to case of killer cell Ig-like receptors and MHC class I interactions CEACAM1. there is no inhibition of NK cytotoxicity in trans (26). In agreement with these findings, in most of the NK clones tested in this Discussion study, no significant increase in the killing of 221 cells was ob- Different functions were implicated for the CEA protein on tu- served when anti-CEACAM Abs were added (Fig. 2E). Therefore mors; mediating intercellular adhesion strengthens the metastatic we propose that in NK clones, homophilic interactions between potential of the tumor cells (24), disrupting tissue structure and different NK cells expressing CEACAM1 do not significantly af- inhibition of cell differentiations (16, 25). However, until now, no fect the overall cytotoxicity of these NK cells. However, it is clear immunological function was attributed to CEA. that homophilic interactions between CEACAM1 proteins on ef- We have recently demonstrated that the homophilic CEACAM1 fector cells are formed as secretion of mIL-2 was observed from interactions are strong enough to deliver inhibitory signals, thereby BW/CEACAM1-␨ cells even without the presence of target cells decreasing human NK-mediated cytotoxicity and decidual lym- (Fig. 2A). Thus the homophilic CEACAM1 interactions on NK phocyte functions (8, 10, 12). In this study, we demonstrate for the cells do exist and might be effective in long term assays. Indeed, first time that the CEA protein can potentially protect the tumors we often observed that CEACAM1-positive NK clones are quite from NK cell-mediated killing. We demonstrate by direct ELISA difficult to grow and that the proliferation of such clones is often (Fig. 1A) and by FACS staining (Fig. 1, B–E) that CEACAM1 slower than that of CEACAM1-negative clones. Alternatively, it is binds to the CEA protein. We further proved by blocking and lysis possible that the CEACAM1 homophilic interactions on NK cells experiments that these heterophilic interactions between the would affect the overall NK cytotoxicity in rare situations when CEACAM1 on NK cells and CEA on the target cells inhibit NK CEACAM1 is expressed in high levels on NK cells. Indeed, we killing (Fig. 2). These findings suggest a new key role for the CEA, could observe a significant increase in the secretion of mIL-2 when enabling tumors to evade attack by CEACAM1-positive NK cells. targets cells expression CEACAM1 or CEA in high levels (Fig. 1, In healthy people, CEACAM1 is expressed only on the surface B and C) were included in the assay (Fig. 2A). of activated, CD16-negative NK cells (21). However, as we have The N-domains have an important role in mediating interactions demonstrated previously, in some melanoma patients, a dramatic between the various CEACAM proteins (27, 28). It was previously increase in CEACAM1 expression can be observed on the surface demonstrated that CEACAM1 interacts homophilically through in- of NK cells, independently of activation and CD16 expression (8). teractions between the N-domains of the two molecules (11). Ad- The reason why CEACAM1 expression is up-regulated on the sur- ditionally, it was reported that the CEA protein interacts ho- face of NK cells in these patients is not understood, but it dem- mophilically through reciprocal interactions between the Ig V-like 6700 CEA INHIBITS NK KILLING VIA INTERACTION WITH CEACAM1

N domain and a Ig C2-like domain on opposite molecules (17). We erophilic binding between CEACAM1 and CEA is important for show in this study that the N domain of both CEACAM1 and CEA regulating the balance between cis-(interaction between molecules is crucial but not sufﬁcient for heterophilic interactions between on the same cell) and trans (interaction between molecules on CEACAM1 and CEA (Fig. 4). Furthermore, the N domain is re- opposed cells) interactions. It is possible that through binding of quired for functional interactions between CEA and CEACAM1 the N domain of one molecule to the Ig C2-like domain on the (Fig. 5). In addition, we show that deletions of each pair of A and opposing cell the counter receptors can achieve optimal contact in

B domains (Ig C2 like domains) reduce the intensity of the CEA trans while the cis-recognition of the receptors on the same cell is binding to CEACAM1 (Fig. 6). These results indicate that the N prevented. domains and the Ig C2-like domains participate in the heterophilic The regulation of various immune responses is achieved by interactions between CEACAM1 and CEA. Therefore, it is likely many different ways. One of which is the distribution of the li- that the heterophilic interactions between CEACAM1 and CEA gands on target cells and the receptors on the effector cells. This are similar to those reported between CEA and CEA, and are also regulation can be hampered when the proteins interact homophili- mediated by the reciprocal binding of the N domain on one mol- cally, because a particular molecule can interact on the same cell ecule with the Ig C2-like domain on the opposite molecule. with another protein of the same kind. Therefore a different type of Because we show that deletions of each pair of A and B domains regulation is needed to prevent such scenario. The results pre-

(Ig C2-like domains) reduce the intensity of the CEA binding to sented in this study demonstrate for the first time that CEA can CEACAM1 (Fig. 6), we assume that the CEACAM1 N domain inhibit the killing of NK cells by interacting with CEACAM1. can bind to each pair of A and B domains and, therefore, the Furthermore, we demonstrated a unique molecular basis for the avidity of CEACAM1 binding to intact CEA protein, containing CEACAM1 and CEA interactions that involves a critical binding three pairs of A and B domains, is the strongest. of the N domain, but additionally includes binding of the other Downloaded from When considering the binding of CEA to CEACAM1, we must domains of the proteins. bear in mind that the CEA protein is four domains longer than the CEACAM1 isoform used in this work (illustrated in Fig. 4A). For Disclosures this reason, we assume that the CEA protein is able to bend and, The authors have no financial conflict of interest. therefore, enable the binding of the N domain of CEACAM1 to any of the CEA Ig C -like domains reciprocally to the binding of http://www.jimmunol.org/ 2 References the bended CEA N domain to the Ig C2-like domains of 1. Obrink, B. 1997. CEA adhesion molecules: multifunctional proteins with signal- CEACAM1. Although we suggest reciprocal head-to-tail binding regulatory properties. Curr. Opin. Cell Biol. 9: 616–626. between CEACAM1 and CEA, we demonstrate that the deletion 2. Beauchemin, N., P. Draber, G. Dveksler, P. Gold, S. Gray-Owen, F. Grunert, S. Hammarstrom, K. V. Holmes, A. Karlsson, M. Kuroki, et al. 1999. Redefined the N domain of CEA was sufficient to abolish the inhibitory effect. nomenclature for members of the carcinoembryonic antigen family. Exp. Cell Therefore, to deliver inhibitory signals, both N domains of CEA Res. 252: 243–249. and CEACAM1 have to participate in the binding (Fig. 5B). 3. Bates, P. A., J. Luo, and M. J. Sternberg. 1992. A predicted three-dimensional structure for the carcinoembryonic antigen (CEA). FEBS Lett. 301: 207–214. We also present some new insights regarding the CEACAM1 4. Benchimol, S., A. Fuks, S. Jothy, N. Beauchemin, K. Shirota, and C. P. Stanners. homophilic interactions. It has been argued that the CEACAM1 1989. Carcinoembryonic antigen, a human tumor marker, functions as an inter- by guest on October 6, 2021 interactions are mediated through interactions of the two N do- cellular adhesion molecule. Cell 57: 327–334. 5. Boulton, I. C., and S. D. Gray-Owen. 2002. Neisserial binding to CEACAM1 mains (11). However we demonstrate that the N domain by itself arrests the activation and proliferation of CD4ϩ T lymphocytes. Nat. Immunol. 3: is not sufficient for homophilic interactions between CEACAM1 229–236. 6. Izzi, L., C. Turbide, C. Houde, T. Kunath, and N. Beauchemin. 1999. cis-deter- molecules (Fig. 3C). These results suggest that even in the case of minants in the cytoplasmic domain of CEACAM1 responsible for its tumor in- homophilic CEACAM1 interactions the Ig C2-like domains are hibitory function. Oncogene 18: 5563–5572. involved in the binding. Previous studies have shown similar re- 7. Chen, D., H. Iijima, T. Nagaishi, A. Nakajima, S. Russell, R. Raychowdhury, V. Morales, C. E. Rudd, N. Utku, and R. S. Blumberg. 2004. Carcinoembryonic sults as deletions of the N domain alone (11, 29) or of all the Ig antigen-related cellular adhesion molecule 1 isoforms alternatively inhibit and C2-like domains together (29) abrogated the adhesion of costimulate human T cell function. J. Immunol. 172: 3535–3543. CEACAM1. Importantly, deletion of the CEACAM1 N domain 8. Markel, G., N. Lieberman, G. Katz, T. I. Arnon, M. Lotem, O. Drize, R. S. Blumberg, E. Bar-Haim, R. Mader, L. Eisenbach, and O. Mandelboim. completely abolished the inhibition of NK cytotoxicity mediated 2002. CD66a interactions between human melanoma and NK cells: a novel class via CEACAM1 homophilic interactions. Therefore, similar to the I MHC-independent inhibitory mechanism of cytotoxicity. J. Immunol. 168: CEA and the CEACAM1 heterophilic interactions the CEACAM1 2803–2810. 9. Markel, G., H. Mussaffi, K. L. Ling, M. Salio, S. Gadola, G. Steuer, H. Blau, homophilic interactions are fully dependent on the N domains. H. Achdout, M. de Miguel, T. Gonen-Gross, et al. 2004. The mechanisms con- Nevertheless, this does not preclude that in addition to the binding trolling NK cell autoreactivity in TAP2-deficient patients. Blood 103: 1770–1778. between the N domains of CEACAM1 there is a lower affinity 10. Markel, G., D. Wolf, J. Hanna, R. Gazit, D. Goldman-Wohl, Y. Lavy, S. Yagel, reciprocal binding of the N domains to the Ig C2-like domains on and O. Mandelboim. 2002. Pivotal role of CEACAM1 protein in the inhibition of the opposite molecule (similar to the CEA homophilic binding) activated decidual lymphocyte functions. J. Clin. Invest. 110: 943–953. 11. Watt, S. M., A. M. Teixeira, G. Q. Zhou, R. Doyonnas, Y. Zhang, F. Grunert, which increases the avidity of the binding. Another possible mode R. S. Blumberg, M. Kuroki, K. M. Skubitz, and P. A. Bates. 2001. Homophilic of interaction that can explain the involvement of the Ig C2-like adhesion of human CEACAM1 involves N-terminal domain interactions: struc- domains in the homophilic CEACAM1 binding may be similar to tural analysis of the binding site. Blood 98: 1469–1479. 12. Markel, G., R. Gruda, H. Achdout, G. Katz, M. Nechama, R. S. Blumberg, the suggested interactions between ICAM-1 and LFA-1. The crys- R. Kammerer, W. Zimmermann, and O. Mandelboim. 2004. The critical role of tal structure of the two N-terminal domains of ICAM-1 has re- residues 43R and 44Q of carcinoembryonic antigen cell adhesion molecules-1 in vealed that both domains function in LFA-1 binding, the first in- the protection from killing by human NK cells. J. Immunol. 173: 3732–3739. 13. Oikawa, S., M. Kuroki, Y. Matsuoka, G. Kosaki, and H. Nakazato. 1992. Ho- teracting with residues on LFA-1 and the second is involved in motypic and heterotypic Ca2ϩ-independent cell adhesion activities of biliary gly- orienting the recognition surface of the first domain (30). It is coprotein, a member of carcinoembryonic antigen family, expressed on CHO cell surface. Biochem. Biophys. Res. Commun. 186: 881–887. possible that in the CEACAM1 interactions residues in the N do- 14. Thompson, J. A., F. Grunert, and W. Zimmermann. 1991. Carcinoembryonic mains interact with each other whereas the Ig C2-like domains are antigen gene family: molecular biology and clinical perspectives. J. Clin. Lab. responsible for the correct orientation. Anal. 5: 344–366. 15. Hammarstrom, S. 1999. The carcinoembryonic antigen (CEA) family: structures, It might be speculated that the involvement of the Ig C2-like suggested functions and expression in normal and malignant tissues. Semin. Can- domains in the homophilic binding of CEACAM1 and in the het- cer Biol. 9: 67–81. The Journal of Immunology 6701

16. Screaton, R. A., L. Z. Penn, and C. P. Stanners. 1997. Carcinoembryonic antigen, 24. Jothy, S., S. Y. Yuan, and K. Shirota. 1993. Transcription of carcinoembryonic a human tumor marker, cooperates with Myc and Bcl-2 in cellular transformation. antigen in normal colon and colon carcinoma: in situ hybridization study and J. Cell Biol. 137: 939–952. implication for a new in vivo functional model. Am. J. Pathol. 143: 250–257. 17. Zhou, H., A. Fuks, G. Alcaraz, T. J. Bolling, and C. P. Stanners. 1993. Ho- 25. Eidelman, F. J., A. Fuks, L. DeMarte, M. Taheri, and C. P. Stanners. 1993. mophilic adhesion between Ig superfamily carcinoembryonic antigen molecules Human carcinoembryonic antigen, an intercellular adhesion molecule, blocks fu- involves double reciprocal bonds. J. Cell Biol. 122: 951–960. sion and differentiation of rat myoblasts. J. Cell Biol. 123: 467–475. 18. Mandelboim, O., H. T. Reyburn, M. Vales-Gomez, L. Pazmany, M. Colonna, G. Borsellino, and J. L. Strominger. 1996. Protection from lysis by natural killer 26. Colonna, M., G. Borsellino, M. Falco, G. B. Ferrara, and J. L. Strominger. 1993. cells of group 1 and 2 specificity is mediated by residue 80 in human histocom- HLA-C is the inhibitory ligand that determines dominant resistance to lysis by patibility leukocyte antigen C alleles and also occurs with empty major histo- NK1- and NK2-specific natural killer cells. Proc. Natl. Acad. Sci. USA 90: compatibility complex molecules. J. Exp. Med. 184: 913–922. 12000–12004. 19. Gonen-Gross, T., H. Achdout, R. Gazit, J. Hanna, S. Mizrahi, G. Markel, 27. Oikawa, S., C. Inuzuka, M. Kuroki, F. Arakawa, Y. Matsuoka, G. Kosaki, and D. Goldman-Wohl, S. Yagel, V. Horejsi, O. Levy, et al. 2003. Complexes of H. Nakazato. 1991. A specific heterotypic cell adhesion activity between mem- HLA-G protein on the cell surface are important for leukocyte Ig-like receptor-1 bers of carcinoembryonic antigen family, W272 and NCA, is mediated by N- function. J. Immunol. 171: 1343–1351. domains. J. Biol. Chem. 266: 7995–8001. 20. Katz, G., G. Markel, S. Mizrahi, T. I. Arnon, and O. Mandelboim. 2001. Rec- ognition of HLA-Cw4 but not HLA-Cw6 by the NK cell receptor killer cell 28. Kuroki, M., H. Abe, T. Imakiirei, S. Liao, H. Uchida, Y. Yamauchi, and Ig-like receptor two-domain short tail number 4. J. Immunol. 166: 7260–7267. S. Oikawa. 2001. Identification and comparison of residues critical for cell-ad- 21. Moller, M. J., R. Kammerer, F. Grunert, and S. von Kleist. 1996. Biliary glyco- hesion activities of two neutrophil CD66 antigens, CEACAM6 and CEACAM8. protein (BGP) expression on T cells and on a natural-killer-cell sub-population. J. Leukocyte Biol. 70: 543–550. Int. J. Cancer 65: 740–745. 29. Cheung, P. H., W. Luo, Y. Qiu, X. Zhang, K. Earley, P. Millirons, and S. H. Lin. ␥RIII (CD16) 22. Lanier, L. L., G. Yu, and J. H. Phillips. 1991. Analysis of Fc 1993. Structure and function of C-CAM1: the first immunoglobulin domain is membrane expression and association with CD3␨ and Fc⑀RI-␥ by site-directed required for intercellular adhesion. J. Biol. Chem. 268: 24303–24310. mutation. J. Immunol. 146: 1571–1576. 23. Beauchemin, N., T. Kunath, J. Robitaille, B. Chow, C. Turbide, E. Daniels, and 30. Wang, J., and T. A. Springer. 1998. Structural specializations of immunoglobulin A. Veillette. 1997. Association of biliary glycoprotein with protein tyrosine phos- superfamily members for adhesion to integrins and viruses. Immunol. Rev. 163: phatase SHP-1 in malignant colon epithelial cells. Oncogene 14: 783–790. 197–215. Downloaded from http://www.jimmunol.org/ by guest on October 6, 2021