The JI
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     
 

The Journal of Immunology, 2007, 179, 1022 -1029
Copyright © 2007 by The American Association of Immunologists, Inc.
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Schwartzkopff, S.
Right arrow Articles by Pircher, H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Schwartzkopff, S.
Right arrow Articles by Pircher, H.
Right arrowPubmed/NCBI databases
*Gene*GEO Profiles
*HomoloGene*Protein
*UniGene
*Genetics Home Reference

Tumor-Associated E-Cadherin Mutations Affect Binding to the Killer Cell Lectin-Like Receptor G1 in Humans1

Sabrina Schwartzkopff2,*, Carsten Gründemann2,*, Oliver Schweier*, Stephan Rosshart*, Klaus Erik Karjalainen{dagger}, Karl-Friedrich Becker{ddagger} and Hanspeter Pircher3,*

* Institute of Medical Microbiology and Hygiene, Department of Immunology, University of Freiburg, Germany; {dagger} Nanyang Technological University, Singapore; and {ddagger} Institute of Pathology, Technical University, Munich, Germany


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 Disclosures
 References
 
The killer cell lectin-like receptor G1 (KLRG1) is expressed by NK cells and memory T cells in man and mice. Cadherins were recently identified as ligands for mouse KLRG1 but ligands for human KLRG1 have not yet been defined. In this study, we first demonstrate that human E-cadherin is a ligand for human KLRG1. This finding is remarkable because human and mouse KLRG1 show only an intermediate degree of homology (57% aa identity). In addition, we show that E-cadherin, expressed on K562 target cells, inhibited polyclonal human NK cells. Inhibition of NK cell function was observed consistently in three independent functional assays but the extent of inhibition was modest and required high expression of E-cadherin on target cells. E-cadherin function is often inactivated during development of human carcinomas and splice-site mutations resulting in in-frame loss of exon 8 or 9 occur frequently in diffuse type gastric carcinomas. Our experiments further revealed that interaction of human KLRG1 to E-cadherin was susceptible to these tumor-associated mutations and that KLRG1+ NK cells were triggered more easily by K562 target cells carrying these mutations in comparison to target cells expressing wild-type E-cadherin. These results also indicate that the E-cadherin binding sites important for homophilic interaction are also involved in KLRG1 binding. Taken together, these data demonstrate that the main adhesion molecule of epithelial tissue, E-cadherin, is involved in regulation of NK cells in both humans and mice.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 Disclosures
 References
 
The killer cell lectin-like receptor G1 (KLRG1)4 is a type II transmembrane protein that belongs to the family of C-type lectins. It is found in humans (1, 2), rats (3), and mice (4, 5) and contains an ITIM in its cytoplasmic tail. The human KLRG1 gene maps to the NK gene complex (1), whereas the mouse homologue is located ~2 cM distant from this locus (6). In humans, 50–80% of CD56dim NK cells express KLRG1 (7), whereas in mice, kept under specific pathogen free conditions, KLRG1 is found on ~30% of NK cells (5). In both species, KLRG1 is also expressed by effector/memory-phenotype T cells that exhibit an impaired proliferation potential (7, 8). Infection of mice with viruses or parasites leads to a substantial increase in KLRG1 expression by NK cells and CD4 and CD8 T cells (8, 9). Moreover, repetitive Ag stimulation dramatically increases KLRG1 expression by virus-specific CD8 T cells in mice (8, 10) and corresponding data in humans revealed that virus-specific CD8 T cells are mostly KLRG1+ in chronic but not in resolved viral infections (11).

Members of the classical cadherin family were recently shown to serve as ligands for mouse KLRG1 (12, 13, 14). These studies further demonstrated that ligation of mouse KLRG1 by E-cadherin inhibited induction of CTLs and the lytic activity of an NK cell line in vitro. Down-regulation of E-cadherin represents a crucial step in epithelial tumor development and metastasis (15). Therefore, tumor cells lacking E-cadherin may be more susceptible to NK cell attack. Mouse KLRG1 also binds to human E-cadherin, but due to the high degree of homology between mouse and human E-cadherin (89% aa identity), this xenogeneic ligand reactivity is not entirely unexpected. More crucial is the question whether human E-cadherin also functions as a ligand for human KLRG1 because human and mouse KLRG1 exhibit only a moderate degree of homology (57% aa identity). In addition, it is important to know whether ligation of human KLRG1 by its physiological ligand interferes with the functional activity of KLRG1+ lymphocytes. Finally, it is well established that alterations in the E-cadherin gene occur frequently during tumor development (16). It was therefore of interest to determine whether naturally occurring tumor-associated E-cadherin mutations interfere with KLRG1 binding.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 Disclosures
 References
 
A5-KLRG1 reporter cells

To generate the CD3{zeta}.human KLRG1 fusion receptor (hKLRG1), the transmembrane and extracellular domains of human KLRG1 and the cytoplasmic domain of mouse CD3{zeta} were isolated by PCR using the following primers: hKLRG1-forward, 5'-ACCCTGGCCCCTCGCTGTTCTTGCCTTGTGGCAATAG-3'; hKLRG1-reverse, 5'-GCGGCCGCTCAAAGTCTGACCTTCTTACACAC-3'; CD3Z- forward, 5'-CTCGAGATGAGAGCAAAATTCAGCAGGAGTG-3'; and hCD3Z-reverse, 5'-CCACAAGGCAAGAACAGCGAGGGGCCAGGGTCTG-3'. The fusion receptor was generated by gene splicing by overlap extension using the hKLRG1- and the CD3Z-PCR products and CD3Z-forward and KLRG1-reverse primers. The PCR product was cloned into the XhoI site of pcDNA3.1/Zeo (Invitrogen Life Technologies). A5 T cell hybridomas that contained an NFAT-GFP expression cassette (17) were transfected by electroporation with a Bio-Rad Gene Pulser (250V, 960 microfarad (µF)). Cells were selected in medium containing 600 µg/ml Zeocin (Invitrogen Life Technologies) and enriched for high KLRG1 expressing cells using magnetic beads (Miltenyi Biotec). To generate the CD3{zeta}.mouse KLRG1 fusion receptor (mKLRG1) with mouse KLRG1 transmembrane and extracellular domains and CD3{zeta} cytoplasmic domain, the following primers were used: mKLRG1-forward, 5'-ACCCTGGCCCCTCGCCATCTTTCCCGCTTTGCAATGG-3'; mKLRG1-reverse, 5'-GCGGCCGCTCAGTATAGGACCTTCTTACAGATC-3'; mCD3Z-reverse, 5'-CAAAGCGGGAAAGATGGCGAGGGGCCAGGGTCTG-3' and CD3Z-forward. The fusion receptor was generated by gene splicing by overlap extension using the mouseKLRG1- and the mCD3Z-PCR products and CD3Z-forward and mKLRG1-reverse primers. The resulting PCR product was cloned into pcDNA3.1-TOPO (Invitrogen Life Technologies) and subcloned into the pMSCV2.2. proviral vector (a gift from Dr. A. Diefenbach, University of Freiburg, Germany). Retroviral supernatants were generated by double transfection of 293T cells with pMCSV2.2 and p10A1-gag/pol/env proviral vectors (also obtained from Dr. A. Diefenbach). After 48 h, supernatant was harvested and used to transduce A5 cells by "spinfection" (120 min, 700 x g). Fluorescence-activated cell sorting was used to isolate A5 cells expressing mouse KLRG1.

E-cadherin expressing target cells

K562 cells were transfected by electroporation (250V, 960 µF) with a pEF-BOS expression vector containing human E-cadherin cDNA (18). Cells were selected in medium containing 1.2 mg/ml G418 and fluorescence-activated cell sorting was used to isolate cells expressing different levels of E-cadherin. K562 cells expressing {Delta}8 or {Delta}9 human E-cadherin were generated by cotransfection using {Delta}8 and {Delta}9 human E-cadherin cDNA (18) cloned into the beta-actin based expression vector pBATEM (19) together with pcDNA3.1/neo (Invitrogen Life Technologies). Cells were selected in medium containing 1.2 mg/ml G418 and enriched for high-expressing cells by cell sorting. 721.221 cells were transduced with retroviral supernatants (48 h) generated by double transfection of 293T cells with pMSCV2.2 containing human E-cadherin or GFP and p10A1-gag/pol/env proviral vectors. Fluorescence-activated cell sorting was used to isolate cells expressing high levels of E-cadherin or GFP, respectively. Bone marrow-derived dendritic cells from B6 mice were generated by culturing bone marrow cells from femurs in RPMI 1640 medium (Sigma-Aldrich) supplemented with 10% FCS (Biochrom, Seromed) and mouse GM-CSF (20–50 ng/ml) and recombinant mouse IL-4 (10 ng/ml; PromoKine) for 8 days.

Reporter cell assay

A5-KLRG1 reporter cells (105) were cocultured for 8 h with 105 target cells in 24-well plates (CELLSTAR; Greiner Bio-One). Afterward, cells were harvested and GFP induction was analyzed by flow cytometry. For mAb blocking, purified anti-KLRG1 mAb (13A2 and 13F12F2) (7, 20) or isotype-matched control mAb (10 µg/ml) were added to the culture medium for the entire incubation period. For plate-bound stimulation, anti-KLRG1 mAb or isotype-matched control mAb were immobilized (10 µg/ml) on 96-well high-binding polystyrene flat-bottom EIA/RIA plates (Costar) by overnight incubation in PBS at 4°C. After an extensive wash with PBS, A5-KLRG1 reporter cells (105/well) were added and cultured for 8 h before analysis. Recombinant fusion proteins consisting of the extracellular domain of human or mouse E-cadherin fused to the Fc portion of human IgG1 were purchased from R&D Systems. For plate-bound stimulation, Fc-chimeras were immobilized (1 µg/ml) in 96-well high binding polystyrene flat-bottom EIA/RIA plates (Costar) by overnight incubation in PBS at 4°C.

NK cell assays and staining protocols

PBLs from adult healthy donors were isolated from buffy coats obtained from the Blood Transfusion Center (Freiburg, Germany) using Ficoll-Paque (Amersham Biosciences) gradient centrifugation. NK cells were purified from PBLs using the NK Cell Isolation kit II (Miltenyi Biotec) that allowed isolation of untouched NK cells. Purity of the isolated NK cells, as determined by anti-CD3 and -CD56 mAb staining, was >90%. Cytolytic activity of purified NK cells was determined in a standard 4 h 51Cr release assay in U-bottom 96-well plates. Purified anti-KLRG mAb (13A2 and 13F12F2) (7, 20) and isotype matched control mAb (10 µg/ml) were present during the entire incubation period. The CD107a degranulation assay was performed as described (21). In brief, 2 x 105 purified NK cells were cocultured with 2 x 105 K562 target cells in U-bottom 96-well plates. To each well containing 200 µl of cell suspension, 10 µl (~0.5 µg) PE-conjugated anti-CD107a mAb (BD Pharmingen) were added. After incubation at 37°C for 1 h, 2 µl of 1/10 diluted GolgiStop (BD Pharmingen) was added per well and the plates were incubated for another 2 h. Afterward, cells were surface-stained with Alexa647-conjugated anti-KLRG1 mAb (13F12F2). For intracellular IFN-{gamma} assays, 106 purified NK cells/well were incubated with 106 K562 cells in 24-well plates in the presence of 10 µg/ml brefeldin A (GolgiPlug; BD Pharmingen). After 5 h, cells were first surface-stained with Alexa647-conjugated anti-KLRG1 mAb (13F12F2), fixed and permeabilized using Cytofix/Cytoperm solution (BD Pharmingen), followed by staining with PE-conjugated anti-IFN-{gamma} mAb (BD Pharmingen). Human E-cadherin was detected by mAb SHE78–7 (Alexis Biochemicals) followed by PE-labeled goat-anti-mouse IgG (Caltag Laboratories). Samples were analyzed with a BD FACSCalibur flow cytometer (BD Biosciences) using CellQuest Pro software (BD Biosciences).


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 Disclosures
 References
 
Human E-cadherin is a ligand for human KLRG1

To determine the ligand specificity of human KLRG1, we generated the A5-hKLRG1 reporter cell line that expressed a fusion receptor consisting of the transmembrane and extracellular domain of human KLRG1 fused to the intracellular part of the ITAM-containing mouse CD3{zeta}-chain. Ligation of ITAM-containing molecules in this cell line leads to NFAT-dependent GFP expression (17). Cell surface staining with anti-KLRG1 mAb revealed that the chimeric KLRG1 protein was expressed by the transfected A5 cells at a density comparable to human NK cells (Fig. 1A, left and Fig. 4). Coculture of A5-hKLRG1 reporter cells with E-cadherin-transfected K562 cells resulted in robust NFAT-GFP activation, whereas coculture with mock-transfected K562 cells did not. Moreover, induction of GFP expression was inhibited by anti-KLRG1 mAb (Fig. 1B). Coculture of nontransfected A5 reporter cells with E-cadherin transfected K562 cells did not result in GFP induction (data not shown). GFP expression in A5-hKLRG1 reporter cells could also be induced by plate-bound recombinant fusion proteins consisting of the five extracellular domains of human E-cadherin fused to the Fc portion of human IgG1 (hE-cad/Fc-chimera). The extent of E-cadherin-Fc-chimera-induced GFP expression in A5-hKLRG1 reporter cells was similar to that obtained by stimulation with plate-bound anti-KLRG1 mAb (Fig. 1C).


Figure 1
View larger version (40K):
[in this window]
[in a new window]

  		FIGURE 1. Human E-cadherin is a ligand for human KLRG1. A, Staining of A5-hKLRG1 reporter cells with anti-KLRG1 mAb (left) and of E-cadherin-transfected K562 cells with anti-E-cadherin mAb (right). Open histograms, negative control staining. B, A5-hKLRG1 cells were cultured for 8 h with mock- or E-cadherin-transfected K562 cells in the presence of the indicated mAb. Afterward, cells were analyzed by flow cytometry. C, A5-hKLRG1 reporter cells were cultured for 8 h in wells that were precoated with isotype control mAb, recombinant human E-cadherin-Fc-chimeras or anti-KLRG1 mAb. Shown are representative results from three independent experiments.

 

Figure 4
View larger version (32K):
[in this window]
[in a new window]

  		FIGURE 4. Degranulation and IFN-{gamma} production of KLRG1+ NK cells in response to K562 target cells is inhibited by E-cadherin. A, Freshly isolated, purified human NK cells were cocultured with mock- or E-cadherin-transfected K562 cells in the presence of anti-CD107a mAb. After 3 h, cells were harvested and additionally surface-stained with anti-KLRG1 mAb. Representative dot plots (left) and ratios of the percentage of CD107a+ KLRG1+ (R2) to the percentage of CD107a+ KLRG1– (R1) NK cells after stimulation with mock- or E-cadherin-transfected K562 cells are shown (right). Values from six individual donors, indicated by the different symbols, tested in six independent experiments are displayed. B, NK cells were cocultured with mock- or E-cadherin-transfected K562 cells in the presence of brefeldin A. After 5 h, cells were surface-stained with anti-KLRG1 mAb, fixed, permeabilized, and stained for intracellular IFN-{gamma}. Representative dot plots (left) and the ratios of the percentage of IFN-{gamma}+ KLRG1+ (R2) to the percentage of IFN-{gamma}+ KLRG1– NK (R1) cells after stimulation with mock- or E-cadherin-transfected K562 cells are shown (right). Values from five individual donors, indicated by the different symbols, tested in four independent experiments are depicted.

 
Mouse KLRG1 exhibits a strong xenogeneic reactivity to human E-cadherin (12, 13). To determine whether this is also true for the human homologue, A5-hKLRG1 reporter cells were stimulated in parallel by plate-bound Fc-chimeras containing mouse or human E-cadherin. In addition, A5-mKLRG1 reporter cells bearing a fusion receptor with mouse KLRG1 ectodomains and CD3{zeta} cytoplasmic domains were included in the assay. A5-hKLRG1 reporter cells were triggered efficiently by the Fc-chimera containing human E-cadherin, whereas stimulation by the Fc-chimera with mouse E-cadherin induced significant lower amounts of GFP (Fig. 2). In contrast, A5-mKLRG1 reporter cells were strongly stimulated by both types of E-cadherin-Fc-chimeras. Moreover, coculture of A5-hKLRG1 reporter cells with murine bone marrow-derived dendritic cells expressing mouse E-cadherin led to minimal GFP induction, whereas the same stimulator cells were able to trigger strong GPF induction in A5-mKLRG1 reporter cells. Thus, these data indicate that human KLRG1 exhibits only a weak xenogeneic reactivity to mouse E-cadherin. Taken together, these results demonstrate that human E-cadherin is a ligand for human KLRG1.


Figure 2
View larger version (37K):
[in this window]
[in a new window]

  		FIGURE 2. Human and mouse KLRG1 differ in their xenogeneic reactivity to E-cadherin. A5-hKLRG1 and A5-mKLRG1 reporter cells expressing human and mouse KLRG1 ectodomains, respectively, were cultured for 8 h in wells that were precoated with Fc-chimeras containing human or mouse E-cadherin. As indicated, plate-bound isotype control mAb were used as negative control. In addition, A5 reporter cells were cocultured with murine bone marrow-derived dendritic cells (mBM-DC) expressing mouse E-cadherin. Results from a representative experiment from three independent experiments are shown.

 
Anti-KLRG1 mAb increases the cytolytic activity of polyclonal human NK cells

Due to the ability of anti-KLRG1 mAb to interfere with the binding of KLRG1 to E-cadherin, we tested whether the cytolytic activity of NK cells toward E-cadherin expressing K562 target cells could be enhanced by addition of anti-KLRG1 mAb. This proved to be the case (Fig. 3A). Although the increase in the percentage specific lysis was modest, it was observed in four independent experiments using four different donors. In addition, increase in lysis was only observed when E-cadherin expressing K562 target cells were used, whereas the addition of anti-KLRG1 mAb did not influence lysis of mock-transfected K562 target cells ruling out unspecific effects of the mAb addition (Fig. 3B). These data provide the first clues that KLRG1 can function as an inhibitory receptor in human NK cells when E-cadherin expressing cells are encountered.


Figure 3
View larger version (28K):
[in this window]
[in a new window]

  		FIGURE 3. Anti-KLRG1 mAb enhances the cytolytic activity of NK cells. Freshly isolated, purified human NK cells were tested against E-cadherin (A) or mock(B)-transfected K562 target cells in 4 h 51Cr release assays in the presence of 10 µg/ml anti-KLRG1 (•) or isotype matched control mAb ({circ}). Four different donors tested in four independent experiments are shown.

 
The relatively small increase in lysis of E-cadherin expressing target cells induced by anti-KLRG1 mAb might be due to the considerable number of KLRG1-negative NK cells (30–40%) in the total NK cell population. Our attempts to use cell sorter purified KLRG1+ NK+ cells failed because KLRG1 was not re-expressed in NK cells that had been stained for FACS-isolation with anti-KLRG1 mAb. In addition, KLRG1 was down-regulated within 2–3 days when freshly isolated NK cells were cultured in vitro and the NK cell lines tested (NKL, YTS, and NK92-C2) did not express KLRG1. Due to these constraints, we turned to assays that allowed measurement of effector functions of ex vivo isolated polyclonal NK cells at the single cell level.

E-cadherin inhibits degranulation of polyclonal human KLRG1+ NK cells

Degranulation is a crucial step in cell killing and cell surface externalization of the lysosomal-associated membrane protein-1 (CD107a) is a convenient assay to determine degranulation of NK cells following stimulation (21). Stimulation of human NK cells with mock-transfected K562 cells resulted in CD107a expression by both KLRG1+ and KLRG1– NK cell populations. When NK cells from the same donors were stimulated with E-cadherin-transfected K562 cells, the number of CD107a+ cells decreased significantly within the KLRG1+ but not within the KLRG1– NK cell subset (Fig. 4A, left). This resulted in a lower ratio of responding KLRG1+ (region R2) to KLRG1– NK cells (region R1) compared with stimulation with mock-transfected K562 cells. Decreased degranulation of KLRG1+ NK cells after stimulation with E-cadherin-transfected K562 cells was observed in six independent experiments using six different donors (Fig. 4A, right).

E-cadherin inhibits IFN-{gamma} secretion by polyclonal human KLRG1+ NK cells

Besides cell killing, secretion of IFN-{gamma} represents a second important effector cell function of NK cells. We therefore determined whether KLRG1 ligation by E-cadherin was also able to interfere with IFN-{gamma} production by human NK cells. Parallel stimulation of aliquots of the same batch of NK cells with E-cadherin- or mock-transfected K562 cells revealed a significant decrease in the number of IFN-{gamma}+ KLRG1+ NK cells after stimulation with E-cadherin- compared with mock-transfected K562 cells (Fig. 4B, left). Correspondingly, the ratio of IFN-{gamma} producing KLRG1+ (region R2) to KLRG1– NK cells (region R1) was lowered after stimulation with E-cadherin- compared with mock-transfected K562 cells. Decreased R2 to R1 ratios after stimulation with E-cadherin expressing K562 cells were observed in four independent experiments using five different donors (Fig. 4B, right).

High expression of E-cadherin is required for inhibition of NK cell function

To determine whether target cells other than K562 cells could also be used to demonstrate the inhibitory function of KLRG1, the EBV-transformed B cell line LCL 721.221 was transduced with E-cadherin or, as a control, with GFP. E-cadherin expressed on 721.221 cells also decreased induction of CD107a on KLRG1+ NK cells in degranulation assays, however, the extent of inhibition was lower compared with the effects obtained with K562-E-cadherin cells (Fig. 5A). In addition, the degree of inhibition as reflected by decreased R2 to R1 ratios after stimulation with E-cadherin expressing target cells varied among different donors (Fig. 5B). E-cadherin was expressed on 721.221 cells at lower levels compared with K562 transfectants (mean fluorescence (MF), 229 vs 575; Fig. 5C vs Fig. 1A). Therefore, we hypothesized that high expression of E-cadherin on target cells might be required for significant inhibition of NK cell function. Our attempts to generate 721.221 cells expressing E-cadherin at higher levels using different expression vectors failed (data not shown). However, we succeeded in generating K562 cells expressing E-cadherin at intermediate levels (MF 103; Fig. 6C). Although E-cadherin expression on these cells was still substantial (~25-fold over background), significant inhibition of NK cell function could only be achieved with K562 cells expressing E-cadherin at higher levels (Fig. 6, A and B).


Figure 5
View larger version (26K):
[in this window]
[in a new window]

  		FIGURE 5. Reactivity of KLRG1+ NK cells to 721.221-E-cadherin cells. NK cells were cocultured with 721.221 cells transduced with E-cadherin or GFP in the presence of anti-CD107a mAb. After 3 h, cells were harvested and additionally surface-stained with anti-KLRG1 mAb. Representative dot plots (A) and the ratios of of the percentage of CD107a+ KLRG1+ (R2) to the percentage of CD107a+ KLRG1– (R1) NK cells after stimulation with 721.221-E-cadherin or 721.221-GFP cells (B) are shown. Values from nine individual donors, indicated by the different symbols, tested in four independent experiments are displayed. C, E-cadherin expression by transduced 721.221 cells analyzed with anti-E-cadherin mAb. One of three experiments is shown.

 

Figure 6
View larger version (33K):
[in this window]
[in a new window]

  		FIGURE 6. Inhibition of NK cell function requires high E-cadherin expression on target cells. NK cells were cocultured with the indicated K562 cells in the presence of anti-CD107a mAb. After 3 h, cells were harvested and additionally surface-stained with anti-KLRG1 mAb. Representative dot plots (A) and the ratios of the percentage of CD107a+ KLRG1+ (R2) to of the percentage of CD107a+ KLRG1– (R1) NK cells after stimulation with the indicated K562 cells (B) are shown. Values from three individual donors, indicated by the different symbols, tested in three independent experiments are displayed. C, E-cadherin expression by the indicated K562 cells analyzed with anti-E-cadherin mAb. One of three experiments is shown. D, A5-hKLRG1 reporter cells were cultured for 8 h in the presence of the indicated K562 cells and GFP induction was determined by flow cytometry. One of three experiments is shown.

 
Induction of GFP in A5-hKLRG1 reporter cells also correlated with E-cadherin expression levels on stimulator cells (Fig. 6D). However, the A5-GFP reporter cell assay appeared to be more sensitive in detecting KLRG1 ligands than the NK cell inhibition assay because K562 cells expressing intermediate levels of E-cadherin were able to induce significant GFP expression despite their borderline activity in inhibiting NK cell function.

Tumor-associated E-cadherin mutations interfere with KLRG1 interaction

Somatic mutations in the E-cadherin gene are found in ~50% of diffuse-type gastric carcinomas in humans. Most of the observed gene alterations are splice-site mutations resulting in in-frame deletion of exon 8 or 9 (22). It was therefore of interest to test whether these alterations interfered with human KLRG1 binding. Interestingly, this proved to be the case. Coculture of A5-hKLRG1 reporter cells with K562 cells transfected with wild-type E-cadherin resulted in strong GFP induction, whereas coculture with K562 cells transfected with E-cadherin mutants lacking exon 8 or 9 ({Delta}8 or {Delta}9) did not (Fig. 7A). Anti-E-cadherin Ab staining further revealed that mutant {Delta}8 or {Delta}9 E-cadherins were expressed by transfected K562 cells at lower levels (MF, 170–185) compared with K562 transfectants expressing wild-type E-cadherin (MF, 522) (Fig. 7B). However, {Delta}8 or {Delta}9 E-cadherins were expressed at higher levels compared with K562-E-cadherinint cells (MF, 103) that were still able to trigger GFP induction in A5-hKLRG1 reporter cells (Fig. 6D). This excludes the possibility that the lack of GFP induction in A5-hKLRG1 reporter cells after coculture with {Delta}8 or {Delta}9 E-cadherins was simply due to lower E-cadherin expression levels.


Figure 7
View larger version (23K):
[in this window]
[in a new window]

  		FIGURE 7. Tumor-associated E-cadherin mutations interfere with KLRG1 binding. A, Reactivity of human KLRG1 to mutant E-cadherins. A5-hKLRG1 cells were cultured for 8 h in the presence of mock-transfected K562 cells or of K562 cells transfected with wild-type, {Delta}8, or {Delta}9 E-cadherin. Afterward, cells were analyzed for induction of GFP expression by flow cytometry. One of three experiments is shown. B, E-cadherin expression by transfected K562 cells analyzed by anti-E-cadherin mAb. One of three experiments is shown. C, Reactivity of mouse KLRG1 to mutant E-cadherins. A5-mKLRG1 reporter cells were cultured for 8 h in the presence of the indicated stimulator cells. One of three experiments is shown.

 
To determine whether the {Delta}8 and {Delta}9 deletions in the E-cadherin molecule also affected recognition by mouse KLRG1, reporter cell assays were performed with A5-mKLRG1 cells expressing mouse KLRG1 ectodomains (Fig. 7C). As expected, a robust GFP induction in A5-mKLRG1 reporter cells was found after coculture with E-cadherin- but not with mock-transfected K562 cells. Similar to the data with the human KLRG1, interaction of mouse KLRG1 with E-cadherin was abolished completely by the {Delta}8 deletion in the E-cadherin molecule. However, in contrast to A5-hKLRG1 reporter cells, A5-mKLRG1 reporter cells were triggered by mutant {Delta}9 E-cadherin, albeit at a somewhat reduced level.

Finally, we determined whether the {Delta}8 or {Delta}9 E-cadherin mutations also interfered with KLRG1 interaction in functional NK cell assays. The experiments revealed that expression of {Delta}8 or {Delta}9 E-cadherin on K562 cells did not inhibit induction of CD107a expression on KLRG1+ NK cells resulting in similar ratios of responding KLRG1+ (R2) to KLRG1– (R1) NK cells as obtained with mock-transfected K562 cells. In contrast, wild-type E-cadherin expressed on K562 cells selectively decreased induction of CD107a on KLRG1+ NK cells resulting in significant lower R2 to R1 ratios (Fig. 8, A and B). The same conclusion was obtained when induction of IFN-{gamma} production was used to measure NK cell activation (Fig. 8, C and D). Taken together, these data demonstrate that human KLRG1 failed to bind to {Delta}8 and {Delta}9 E-cadherin mutants. Thus, NK cells could be triggered more easily by target cells lacking E-cadherin expression or carrying these mutations compared with target cells expressing wild-type E-cadherin.


Figure 8
View larger version (47K):
[in this window]
[in a new window]

  		FIGURE 8. Tumor-associated E-cadherin mutations prevent KLRG1-mediated NK cell inhibition. A, NK cells were cocultured with mock-transfected K562 cells or with K562 cells transfected with wild-type, {Delta}8, or {Delta}9 E-cadherin in the presence of anti-CD107a mAb. After 3 h, cells were harvested and additionally surface- stained with anti-KLRG1 mAb. Representative dot plots from one experiment are shown. B, Ratio of the percentage of KLRG1+CD107a+ (R2) to the percentage of KLRG1– CD107a+ NK cells (R1) after stimulation with the different K562 cells as indicated. Values from five individual donors, indicated by the different symbols, tested in four independent experiments are shown. C, NK cells were cocultured with the different K562 cells as indicated in the presence of brefeldin A. After 5 h, cells were surface-stained with anti-KLRG1 mAb, fixed, permeabilized, and stained for intracellular IFN-{gamma}. Representative dot plots from one experiment are shown. D, Ratio of the percentage of KLRG1+IFN-{gamma}+ (R2) to the percentage of KLRG1–IFN-{gamma}+ NK cells (R1) after stimulation with the different K562 cells as indicated. Values from four individual donors, indicated by the different symbols, tested in three independent experiments are shown.

 

   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
Disclosures
 References
 
E-cadherin, a classic member of the cadherin superfamily, is expressed by epithelial cells, keratinocytes, and Langerhans cells and mediates Ca2+-dependent homophilic cell adhesion. Classical cadherins have a modular structure consisting of five extracellular domains of ~110 aa with internal sequence homology. Intracellularly, E-cadherin is linked to the catenins that connect E-cadherin to the actin cytoskeleton. It is well established that E-cadherin function is frequently inactivated during development of human carcinomas. Abrogation of E-cadherin function may occur by several mechanisms including deletions and mutations in the E-cadherin gene. In ~50% of diffuse type gastric carcinomas, E-cadherin alterations have been observed and most of these alterations were splice-site mutations resulting in in-frame loss of exon 8 ({Delta}8) or 9 ({Delta}9) of the E-cadherin gene (22). This results in altered E-cadherin molecules that carry amino acid deletions in the second ({Delta}8; aa 337–aa 379) or third ({Delta}9; aa 380–aa 440) domain of the E-cadherin molecule. Expression of mutant {Delta}8 and {Delta}9 E-cadherin in cell lines has been shown to alter cellular morphology, to decrease cellular adhesion, and to increase cellular motility (18). The present study demonstrates that these E-cadherin mutations also abolished interaction with KLRG1. Thus, the amino acids encoded by exon 8 and 9 are not only important for homophilic adhesive function but also for heterophilic interaction with KLRG1. E-cadherin also represents a ligand for the integrin {alpha}Ebeta7 (CD103) expressed predominantly by intraepithelial T cells (23). Mutational analysis localized the CD103 recognition site to the first N-terminal domain of E-cadherin (24). Moreover, this analysis revealed that E-cadherin residues critical for heterophilic adhesion to CD103 were distinct from those required for homophilic adhesion. Taken together, these data indicate that CD103 and KLRG1 bind to different sites on E-cadherin.

Mouse and human E-cadherin show a high degree of homology (89% aa identity). Nonetheless, our results indicate that human KLRG1 exhibits only a weak xenogeneic reactivity to mouse E-cadherin whereas mouse KLRG1 recognizes mouse and human E-cadherin with comparable efficiency. In addition, our data suggest that the E-cadherin binding sites for mouse and human KLRG1 are not identical because the {Delta}9 mutation completely abolished interaction with human but not with mouse KLRG1. In comparison to E-cadherin, human and mouse KLRG1 show a considerable lower degree of homology (57% aa identity). Our finding that the interaction of KLRG1 to E-cadherin is conserved in humans and mice is therefore quite remarkable.

Our experiments further demonstrated that high expression levels of E-cadherin on target cells were required for inhibition of NK cell functions. This raises the question whether E-cadherin is expressed in vivo at sufficient levels to mediate inhibitory effects. The intensity of E-cadherin Ab staining of freshly isolated keratinocytes and Langerhans cells (25) is comparable to data obtained here with E-cadherin-transfected cell lines. In epithelial tissue, E-cadherin clusters at sites of cell-cell contact in adherent junctions. At these sites, E-cadherin molecules are abundant but they may not be easily accessible to immune cells because they are hidden in small gaps between epithelial cells. In addition, our data show that the sites on the E-cadherin molecule that are important for homophilic interaction are also involved in KLRG1 binding. It is therefore conceivable that KLRG1 binding has to compete with homophilic E-cadherin interactions. Infections often lead to tissue disruption and then, E-cadherin molecules could become better "visible" for lymphocytes. Under these conditions, KLRG1-mediated inhibition of effector cell function may help to prevent collateral damage to healthy uninfected cells.

The interaction of KLRG1 with E-cadherin should also be discussed in the context of previous findings that lympocytes also express receptors for nectins and nectin-like proteins. Nectins and nectin-like proteins are, like cadherins, cell-cell adhesion molecules involved in the formation of adherent junctions in epithelial tissues (26). CD226 (DNAM-1) present on NK cells and T cells recognizes CD112 (nectin-2) and CD155 (necl-5) (27), CD96 binds to CD155 (28), and CRTRAM is specific for necl-2 (29). In contrast to KLRG1, CD226, CD96, and CRTRAM are triggering receptors. This illustrates that immune cells are equipped with various signaling molecules that bind to proteins involved in epithelial tissue organization. The final outcome of the interaction between immune cells and epithelial tissue, either disrupted or intact, will therefore depend on the balance of activating and inhibitory signals.

KLRG1-mediated inhibition of NK cell function was consistently observed in three independent functional assays. However, it is adequate to state that the degree of inhibition measured in these in vitro assays was modest and required high expression levels of E-cadherin. To date, in the murine system, E-cadherin-induced inhibition of NK cell function has only been observed with an NK cell line over-expressing KLRG1 after retroviral transfection (13). In these experiments, BW5147 cells that do not provide strong activation signals for mouse NK cells were used as target cells. In addition, these target cells also expressed high levels (MF, ~1000) of E-cadherin. Our attempts to demonstrate E-cadherin-mediated inhibition of mouse NK cells using the classical NK cell target line B16 failed (12). It remains to be clarified whether the physiological role of KLRG1 is restricted to inhibition of effector cell function or whether this molecule also serves other purposes. In this context, it is noteworthy that only one ITIM is present in KLRG1 in both species and that ITIMs of human and mouse KLRG1 differ slightly (VxYxxL vs SxYxxL, respectively).

In conclusion, we here demonstrate that human E-cadherin represents a ligand for KLRG1 in humans and that ligation of KLRG1 by E-cadherin inhibited effector cell functions of polyclonal human NK cells. E-cadherin is the main adhesion molecule of epithelia and plays an important role in carcinogenesis (15). In addition to the adhesive function, it is also involved in modulating signal-transduction pathways (16). Our finding adds this central adhesion molecule to the growing list of proteins involved in MHC-independent regulation of NK cells in both humans and mice.


   Acknowledgments
 
We thank Dr. Wolfgang Schamel for the CD3{zeta} cDNA and Dr. Andreas Diefenbach for proviral vectors.


   Disclosures
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 Disclosures
References
 
The authors have no financial conflict of interest.


   Footnotes
 
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1 This work was supported by the Deutsche Forschungsgemeinschaft (SFB620, Teilprojekt B2). Back

2 S. S. and C. G. contributed equally to this work. Back

3 Address correspondence and reprint requests to Dr. Hanspeter Pircher, Institute of Medical Microbiology and Hygiene, Hermann-Herder-Strasse 11, Freiburg, Germany. E-mail address: hanspeter.pircher{at}uniklinik-freiburg.de Back

4 Abbreviations used in this paper: KLRG1, killer cell lectin-like receptor G1; MF, mean fluorescence; int, intermediate. Back

Received for publication March 16, 2007. Accepted for publication April 12, 2007.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 Disclosures
 References
 

  1. Butcher, S., K. L. Arney, G. P. Cook. 1998. MAFA-L, an ITIM-containing receptor encoded by the human NK cell gene complex and expressed by basophils and NK cells. Eur. J. Immunol. 28: 3755-3762. [Medline]
  2. Lamers, M. B., A. G. Lamont, D. H. Williams. 1998. Human MAFA has alternatively spliced variants. Biochim. Biophys. Acta 1399: 209-212. [Medline]
  3. Guthmann, M. D., M. Tal, I. Pecht. 1995. A new member of the C-type lectin family is a modulator of the mast cell secretory response. Int. Arch. Allergy Immunol. 107: 82-86. [Medline]
  4. Blaser, C., M. Kaufmann, H. Pircher. 1998. Virus-activated CD8 T cells and lymphokine-activated NK cells express the mast cell function-associated antigen, an inhibitory C-type lectin. J. Immunol. 161: 6451-6454. [Abstract/Free Full Text]
  5. Hanke, T., L. Corral, R. E. Vance, D. H. Raulet. 1998. 2F1 antigen, the mouse homolog of the rat "mast cell function- associated antigen," is a lectin-like type II transmembrane receptor expressed by natural killer cells. Eur. J. Immunol. 28: 4409-4417. [Medline]
  6. Voehringer, D., M. Kaufmann, H. Pircher. 2001. Genomic structure, alternative splicing, and physical mapping of the killer cell lectin-like receptor G1 gene (KLRG1), the mouse homologue of MAFA. Immunogenetics 52: 206-211. [Medline]
  7. Voehringer, D., M. Koschella, H. Pircher. 2002. Lack of proliferative capacity of human effector and memory T cells expressing killer cell lectinlike receptor G1 (KLRG1). Blood 100: 3698-3702. [Abstract/Free Full Text]
  8. Voehringer, D., C. Blaser, P. Brawand, D. H. Raulet, T. Hanke, H. Pircher. 2001. Viral infections induce abundant numbers of senescent CD8 T cells. J. Immunol. 167: 4838-4843. [Abstract/Free Full Text]
  9. Robbins, S. H., S. C. Terrizzi, B. C. Sydora, T. Mikayama, L. Brossay. 2003. Differential regulation of killer cell lectin-like receptor G1 expression on T cells. J. Immunol. 170: 5876-5885. [Abstract/Free Full Text]
  10. Masopust, D., S. J. Ha, V. Vezys, R. Ahmed. 2006. Stimulation history dictates memory CD8 T cell phenotype: implications for prime-boost vaccination. J. Immunol. 177: 831-839. [Abstract/Free Full Text]
  11. Thimme, R., V. Appay, M. Koschella, E. Panther, E. Roth, A. D. Hislop, A. B. Rickinson, S. L. Rowland-Jones, H. E. Blum, H. Pircher. 2005. Increased expression of the NK cell receptor KLRG1 by virus-specific CD8 T cells during persistent antigen stimulation. J. Virol. 79: 12112-12116. [Abstract/Free Full Text]
  12. Grundemann, C., M. Bauer, O. Schweier, N. von Oppen, U. Lassing, P. Saudan, K. F. Becker, K. Karp, T. Hanke, M. F. Bachmann, H. Pircher. 2006. Cutting edge: identification of E-cadherin as a ligand for the murine killer cell lectin-like receptor G1. J. Immunol. 176: 1311-1315. [Abstract/Free Full Text]
  13. Ito, M., T. Maruyama, N. Saito, S. Koganei, K. Yamamoto, N. Matsumoto. 2006. Killer cell lectin-like receptor G1 binds three members of the classical cadherin family to inhibit NK cell cytotoxicity. J. Exp. Med. 203: 289-295. [Abstract/Free Full Text]
  14. Tessmer, M. S., C. Fugere, F. Stevenaert, O. V. Naidenko, H. Jonathan Chong, G. Leclercq, L. Brossay. 2007. KLRG1 binds cadherins and preferentially associates with SHIP-1. Int. Immunol. 19: 391-400. [Abstract/Free Full Text]
  15. Takeichi, M.. 1993. Cadherins in cancer: implications for invasion and metastasis. Curr. Opin. Cell Biol. 5: 806-811. [Medline]
  16. Cavallaro, U., G. Christofori. 2004. Cell adhesion and signalling by cadherins and Ig-CAMs in cancer. Nat. Rev. Cancer 4: 118-132. [Medline]
  17. Andersen, P. S., C. Geisler, S. Buus, R. A. Mariuzza, K. Karjalainen. 2001. Role of the T cell receptor ligand affinity in T cell activation by bacterial superantigens. J. Biol. Chem. 276: 33452-33457. [Abstract/Free Full Text]
  18. Handschuh, G., S. Candidus, B. Luber, U. Reich, C. Schott, S. Oswald, H. Becke, P. Hutzler, W. Birchmeier, H. Hofler, K. F. Becker. 1999. Tumour-associated E-cadherin mutations alter cellular morphology, decrease cellular adhesion and increase cellular motility. Oncogene 18: 4301-4312. [Medline]
  19. Nose, A., A. Nagafuchi, M. Takeichi. 1988. Expressed recombinant cadherins mediate cell sorting in model systems. Cell 54: 993-1001. [Medline]
  20. Marcolino, I., G. K. Przybylski, M. Koschella, C. A. Schmidt, D. Voehringer, M. Schlesier, H. Pircher. 2004. Frequent expression of the natural killer cell receptor KLRG1 in human cord blood T cells: correlation with replicative history. Eur. J. Immunol. 34: 2672-2680. [Medline]
  21. Alter, G., J. M. Malenfant, M. Altfeld. 2004. CD107a as a functional marker for the identification of natural killer cell activity. J. Immunol. Methods 294: 15-22. [Medline]
  22. Becker, K. F., M. J. Atkinson, U. Reich, I. Becker, H. Nekarda, J. R. Siewert, H. Hofler. 1994. E-cadherin gene mutations provide clues to diffuse type gastric carcinomas. Cancer Res. 54: 3845-3852. [Abstract/Free Full Text]
  23. Cepek, K. L., S. K. Shaw, C. M. Parker, G. J. Russell, J. S. Morrow, D. L. Rimm, M. B. Brenner. 1994. Adhesion between epithelial cells and T lymphocytes mediated by E-cadherin and the {alpha}Ebeta7 integrin. Nature 372: 190-193. [Medline]
  24. Taraszka, K. S., J. M. Higgins, K. Tan, D. A. Mandelbrot, J. H. Wang, M. B. Brenner. 2000. Molecular basis for leukocyte integrin {alpha}Ebeta7 adhesion to epithelial (E)-cadherin. J. Exp. Med. 191: 1555-1567. [Abstract/Free Full Text]
  25. Tang, A., M. Amagai, L. G. Granger, J. R. Stanley, M. C. Udey. 1993. Adhesion of epidermal Langerhans cells to keratinocytes mediated by E-cadherin. Nature 361: 82-85. [Medline]
  26. Sakisaka, T., Y. Takai. 2004. Biology and pathology of nectins and nectin-like molecules. Curr. Opin. Cell Biol. 16: 513-521. [Medline]
  27. Bottino, C., R. Castriconi, D. Pende, P. Rivera, M. Nanni, B. Carnemolla, C. Cantoni, J. Grassi, S. Marcenaro, N. Reymond, et al 2003. Identification of PVR (CD155) and Nectin-2 (CD112) as cell surface ligands for the human DNAM-1 (CD226) activating molecule. J. Exp. Med. 198: 557-567. [Abstract/Free Full Text]
  28. Fuchs, A., M. Cella, E. Giurisato, A. S. Shaw, M. Colonna. 2004. Cutting edge: CD96 (tactile) promotes NK cell-target cell adhesion by interacting with the poliovirus receptor (CD155). J. Immunol. 172: 3994-3998. [Abstract/Free Full Text]
  29. Boles, K. S., W. Barchet, T. Diacovo, M. Cella, M. Colonna. 2005. The tumor suppressor TSLC1/NECL-2 triggers NK-cell and CD8+ T-cell responses through the cell-surface receptor CRTAM. Blood 106: 779-786. [Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 BloodHome page
J. Van den Bossche, P. Bogaert, J. van Hengel, C. J. Guerin, G. Berx, K. Movahedi, R. Van den Bergh, A. Pereira-Fernandes, J. M. C. Geuns, H. Pircher, et al.
Alternatively activated macrophages engage in homotypic and heterotypic interactions through IL-4 and polyamine-induced E-cadherin/catenin complexes
Blood, November 19, 2009; 114(21): 4664 - 4674.
[Abstract] [Full Text] [PDF]

Home page
 Cancer Res.Home page
S. Attig, J. Hennenlotter, G. Pawelec, G. Klein, S. D. Koch, H. Pircher, S. Feyerabend, D. Wernet, A. Stenzl, H.-G. Rammensee, et al.
Simultaneous Infiltration of Polyfunctional Effector and Suppressor T Cells into Renal Cell Carcinomas
Cancer Res., November 1, 2009; 69(21): 8412 - 8419.
[Abstract] [Full Text] [PDF]

Home page
 J. Biol. Chem.Home page
S. Nakamura, K. Kuroki, I. Ohki, K. Sasaki, M. Kajikawa, T. Maruyama, M. Ito, Y. Kameda, M. Ikura, K. Yamamoto, et al.
Molecular Basis for E-cadherin Recognition by Killer Cell Lectin-like Receptor G1 (KLRG1)
J. Biol. Chem., October 2, 2009; 284(40): 27327 - 27335.
[Abstract] [Full Text] [PDF]

Home page
 BloodHome page
S. M. Henson, O. Franzese, R. Macaulay, V. Libri, R. I. Azevedo, S. Kiani-Alikhan, F. J. Plunkett, J. E. Masters, S. Jackson, S. J. Griffiths, et al.
KLRG1 signaling induces defective Akt (ser473) phosphorylation and proliferative dysfunction of highly differentiated CD8+ T cells
Blood, June 25, 2009; 113(26): 6619 - 6628.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Schwartzkopff, S.
Right arrow Articles by Pircher, H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Schwartzkopff, S.
Right arrow Articles by Pircher, H.
Right arrowPubmed/NCBI databases
*Gene*GEO Profiles
*HomoloGene*Protein
*UniGene
*Genetics Home Reference

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS