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


     
 

The Journal of Immunology, 2007, 179, 3804 -3811
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 Malcherek, G.
Right arrow Articles by Trowsdale, J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Malcherek, G.
Right arrow Articles by Trowsdale, J.

The B7 Homolog Butyrophilin BTN2A1 Is a Novel Ligand for DC-SIGN1

Georg Malcherek2, Luzia Mayr3, Pedro Roda-Navarro4, David Rhodes, Nigel Miller and John Trowsdale

Department of Pathology, Division of Immunology, University of Cambridge, Cambridge, United Kingdom


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 Disclosures
 References
 
The MHC-encoded butyrophilin, BTN2A1, is a cell surface glycoprotein related to the extended family of B7 costimulatory molecules. BTN2A1 mRNA was expressed in most human tissues, but protein expression was significantly lower in leukocytes. An Ig-fusion protein of BTN2A1 bound to immature monocyte-derived dendritic cells. Binding diminished upon MoDC maturation and no binding was detected to Langerhans cells. Induction of the counterreceptor was IL-4 dependent and occurred early during dendritic cell differentiation. The interaction required the presence of Ca2+ and was mediated by high-mannose oligosaccharides. These properties matched DC-SIGN, a DC-specific HIV-1 entry receptor. This was confirmed by binding of soluble BTN2A1 to DC-SIGN-transfectants and its inhibition by a specific Ab. DC-SIGN bound to native BTN2A1 expressed on a range of tissues. However, BTN2A1 was not recognized on some normal cells such as HUVECs despite a similar expression level. The BTN2A1 of tumor cells such as HEK293T have more high-mannose moieties in comparison to HUVECs, and those high-mannose moieties are instrumental for binding to DC-SIGN. The data are consistent with tumor- or tissue-specific glycosylation of BTN2A1 governing recognition by DC-SIGN on immature monocyte-derived dendritic cells.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 Disclosures
 References
 
The butyrophilins (BTN)5 belong to the Ig superfamily, and are encoded in a cluster of seven genes in the extended MHC class I region on chromosome 6 (1, 2). BTN1, BTN2, and BTN3 subfamilies share ~50% amino acid identity. The extracellular domains of the three members within each subfamily are more closely related and share 88% (BTN2A1, -A2 and -A3) and 95% (BTN3A1, -A2 and -A3) identity, respectively. BTN1, the prototype of the family, is a major component of the milk fat globule membrane and is regulated by lactogenic hormones. BTN1 has a crucial function in the secretion of lipids into milk (3). Collectively, BTN2 and BTN3 are cell surface transmembrane glycoproteins, transcripts of which are ubiquitously expressed at a low to intermediate level (2). The extracellular IgV and IgC folds of BTN are related to myelin oligodendrocyte glycoprotein, a molecule confined to the CNS and a potential autoantigen in multiple sclerosis, to B-G molecules of the chicken MHC and to the B7 (CD80/86) costimulatory molecules (1, 4). Indeed, a T cell regulatory function is likely as BTN3A1 bound to a ligand on T cells. Similarly, a related mouse gene, butyrophilin-like 2 (BTNL2) functioned as an inhibitor of T cell activation (5, 6, 7). A heptad repeat of a 7-aa sequence encoded by a single exon and a B30.2/SPRY domain at the C terminus are not found on other B7-like molecules (2). The B30.2/SPRY domain is also part of a large set of tripartite motif (TRIM) proteins, that include TRIM5{alpha}, which has a function in defense against retroviral infections (8).

Dendritic cells (DCs) are professional APCs that have a pivotal role in controlling immune responses, directing them toward immune activation or tolerance (9). An important family of Ag receptor involved in recognition and uptake of glycan structures are the C-type lectin receptors (10). DC-specific ICAM-3 grabbing nonintegrin (DC-SIGN) functions as an internalization receptor for HIV-1, HCV, Mycobacterium tuberculosis, and other pathogens and also mediates cellular interactions with T cells and endothelial cells (11). Recently, evidence has emerged that DC-SIGN recognizes carbohydrate structures on CEACAM-1/Mac-1 and carcinoembryonic Ag (CEA), specifically expressed on neutrophils and tumor tissues, respectively (12, 13).

In this study, we characterized the biochemical features of butyrophilin BTN2A1 and its counterreceptor on immature monocyte-derived DCs (MoDCs), which we identified as DC-SIGN. Binding of DC-SIGN was dependent on the tumor- and/or tissue-specific glycosylation of BTN2A1.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 Disclosures
 References
 
RNA analysis, cloning, and transfection

BTN2A1 RNA expression was analyzed by RT-PCR of a human multiple tissue cDNA panel (BD Clontech). Human BTN2A1 was amplified with forward primer BTN2A1f, 5'-CACCTCGTAGTGGCAGGACTA-3' and reverse primer BTN2r, 5'-TGGGCATAAAGGATTCTGGA-3', designed to bridge two adjacent exons of the IgC domain. The GAPDH primers used were GAPDHf, 5'-ACAACAGCCTCAAGATCATCAG-3', and GAPDHr, 5'-GGTCCACCACTGACACGTTG-3'. For RT-PCR analysis and cDNA cloning, total cellular RNA was extracted using the RNeasy kit (Qiagen). RNA was reverse transcribed using the Prostar kit (Stratagene) and random hexamers. Full-length DC-SIGN cDNA was amplified by RT-PCR with BIO-X-ACT DNA polymerase (Bioline) from total human MoDC cDNA, cloned into pcDNA5/Frt/TOPO (Invitrogen Life Technologies), and confirmed by DNA sequencing. Human embryonic kidney HEK293T cells were transfected with pcDNA5/Frt/TOPO-DC-SIGN using Effectene (Qiagen) and analyzed after 48 h.

To create a DC-SIGN-FLAG fusion protein, a second PCR insert of DC-SIGN was generated using sense oligonucleotide 5'-CCCAGCTCCATAAGTCAGGAA-3' and antisense primer 5'-AAGTTCTGCTACGCAGGAGG-3' corresponding to the C terminus, and ligated in frame into BamHI pFLAG –CMV-3 (Sigma-Aldrich).

Full-length human BTN2A1 cDNAs were cloned into mammalian expression vectors as described (2) and transfected into a panel of primary cells and cell lines using Fugene (Roche Applied Science). BTN2A1 fusion proteins (referred to as BTN2A1-Ig) were prepared by cloning the extracellular domain in-frame with the hinge-CH2-CH3 domain of human IgG1 (14). To produce fusion proteins, HEK293T cells were transfected using Effectene. BTN2A1-Ig, hFc and DC-SIGN-FLAG fusion proteins were purified from culture supernatant using protein A-Sepharose or M{alpha}FLAG-agarose (Sigma-Aldrich).

Immunohistochemistry, immunoprecipitation, and Western blot

Formalin-fixed sections of normal colon (Imgenex) were stained for BTN2A1 expression using the avidin-biotin-peroxidase system (Vectastain, Vector Laboratories). Sections were blocked with 2% normal rabbit serum and the tissue culture supernatant of a monoclonal rat anti-BTN2A1 Ab was applied followed by biotinylated rabbit anti-rat IgG and the avidine-biotin-HRP complex. Secondary Abs were preabsorbed with 10% human Ab serum before application. HRP activity was developed with diaminobenzidine and sections were counterstained with Carrazzi hematoxylin.

HEK293T cells were washed with PBS and suspended in lysis buffer containing 50 mM Tris-HCl (pH 7.2), 150 mM NaCl, 1 mM Ca2+, 1 mM Mg2+, 1% Triton X-100, and a mixture of protease inhibitors (Roche Applied Science) at 4°C for 1 h. sDC-SIGN-FLAG tissue culture supernatant was coupled to M{alpha}FLAG-agarose at 4°C for 12 h, washed with lysis buffer, and subsequently used for precipitation of HEK293T lysates at 4°C overnight. The eluate in SDS sample buffer was separated by 10% SDS-PAGE under reducing conditions, transferred onto an Immobilon P membrane (Millipore) and probed with tissue culture supernatant of a rat anti BTN2A1-B30.2 Ab followed by HRP-rabbit anti rat Ig (DakoCytomation). Western blots were developed using ECL system (Amersham Biosciences).

Cells, Abs, and reagents

PBMCs were obtained from buffy coats of healthy donors by Ficoll gradient centrifugation. Monocytes were prepared from plastic adherent PBMC and incubated for 72 h in the presence of IL-4 (500 U/ml) or for 5 days in IL-4 and GM-CSF (500 U/ml and 100 ng/ml, R&D Systems). At day 5, the phenotype of cultured MoDCs was confirmed by flow cytometry and typically was CD14low, HLA-DRhigh, and CD1ahigh with moderate levels of CD86. MoDCs were activated in the presence of 1 µg/ml LPS (L2654; Sigma-Aldrich) for 48 h. Langerhans cells (LCs) were generated as described (15).

HUVECs were grown in Endothelial Growth Medium (Promo Cell) and IMR-90 (human lung embryonic fibroblasts) were grown in DMEM with glutamax, glucose 1000 mg/l, and sodium pyruvate. MOLT-4 (human acute lymphoblastic leukemia), MelJuSo (human melanoma cell line), Hela (cervical carcinoma cells), HEK293 (embryonal kidney carcinoma cells), and BAF/3 (mouse B cell lymphoma) were grown in RPMI 1640 supplemented with 2 mM glutamine, penicillin/streptomycin (100 U/ml, Invitrogen Life Technologies) and 10% FCS (Harlan Sprague Dawley). The medium for BAF/3 was also supplemented with mIL-3. T cells were separated by a Dako MoFlo cell sorter by negative selection from PBL stained with M{alpha}CD14- (Diatec), M{alpha}CD19-FITC and M{alpha}CD56-RPE (DakoCytomation), and treated with PHA/IL-2 (5 µg/ml/100 U/ml; R&D Systems), PMA/calciumionophore (10/100 ng/ml), or immobilized UCHT-1 anti CD3 Ab (DakoCytomation).

The following mouse anti-human mAbs were used: NA1/34-HLK (CD1a, Insight Biotechnology); IT2.2 (CD86-R-PE); L243 (HLA-DR); DCN46 (DC-SIGN, BD Biosciences); Mab161 (DC-SIGN, R&D Systems); and 1B10 (DC-SIGN, provided by F. Arenzana-Seisdedos, Institute Pasteur, Paris).

The following rat anti-human mAbs were used: Rat{alpha}DC-SIGN-RPE (eBioscience); monoclonal anti-BTN2A1 Ab was raised by immunization of rats with the recombinant B30.2 domain of the BTN2A1 protein and does not cross-react with BTN3 family members, but may also recognize other BTN2 molecules.

Binding of soluble fusion proteins by flow cytometry

To analyze BTN2A1-counterreceptor expression, cells were incubated on ice with soluble BTN2A1-Ig (10 µg/ml, unless indicated otherwise) in FACS binding buffer, 1% FCS, 0.02% sodium azide in D-PBS with Ca2+ and Mg2+. After 20 min cells were washed and stained with FITC-conjugated goat F(ab')2 anti-human IgG (Caltag Laboratories) or PE-conjugated goat anti-human IgG (Jackson ImmunoResearch Laboratories). HIgG (Sigma-Aldrich), hFc, and CTLA-4-Ig were used at 10 µg/ml. For inhibition assays, MoDCs or DC-SIGN-transfectants were preincubated on ice for 20 min with increasing concentrations of mannan, Ca2+-chelator or rHIV-1SF2 gp120 (from Dr. L. Williams, NIBSC Centralised Facility for AIDS Reagents supported by EU Programme EVA and the U.K. Medical Research Council) before fusion proteins were added. To study the carbohydrate dependency of binding, BTN2A1 was treated with Endoglycosidase H (Endo H, 1 mU/µg; Roche Applied Science) at RT for 15 h. For Ab blocking experiments, DCs and DC-SIGN-transfectants were preincubated at 4°C for 20 min with the anti-DC-SIGN 1B10 mAb (35 µg/ml), before directly labeled fusion proteins were added. BTN2A1-Ig was labeled with the Zenon AlexaFluor 488 human IgG labeling kit (Molecular Probes). Cells were analyzed on a FACScan (BD Biosciences) using CellQuest software. To analyze the binding of DC-SIGN-FLAG to a panel of cells transfected with BTN2A1-GFP, a two-step detection protocol was used. An anti-DC-SIGN Ab (DCN46; BD Biosciences) was used to bridge bound sDC-SIGN-FLAG (5 µg/ml) to a goat anti-mouse Alexa 647 (Molecular Probes). Using Alexa 647 had the advantage that cells could be analyzed using SUMMIT 4.2 software on a Cyan ADP (DakoCytomation) without using compensation. For lectin blocking experiments, HEK293T-DC-SIGN transfectants were preincubated at 4°C for 20 min with 100 µg/ml Galanthus nivalis agglutinin (GNA; Vector Laboratories), before the two-step staining was conducted. Binding of biotinylated GNA was detected using streptavidin-R-PE (Sigma-Aldrich).

To examine internalization, DCs were labeled with soluble BTN2A1-Ig, washed, and subsequently incubated at 37°C. Aliquots were removed at 5 min and 15 min and internalization stopped by metabolic fixation (D-PBS with Ca2+ and Mg2+ and sodium azide) at 4°C. Bound BTN2A1-Ig was analyzed by a FITC-conjugated goat F(ab')2 anti-human IgG.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 Disclosures
 References
 
Expression of BTN2A1

Using oligonucleotides specific for BTN2A1, RT-PCR analysis indicated that BTN2A1 mRNA was ubiquitously expressed (Fig. 1A). Immunohistochemistry (IHC) using a BTN2A1-specific mAb localized a high level of expression of BTN2A1 to epithelial cells whereas lower levels were also found in leukocytes (Fig. 1B). Western blot analysis confirmed BTN2A1 expression on a variety of primary cells and cell lines, such as HUVECs, IMR-90, HEK293T, and Jurkat, whereas expression on freshly isolated leukocytes, such as PBLs and monocytes was low (Fig. 1C, data not shown). This is in contrast to other B7-like molecules, as well as the related set of BTN3 butyrophilins. BTN3 molecules are preferentially expressed on T cells, T cell lines, and at lower levels on other PBMCs and some tumor cell lines (5).


Figure 1
View larger version (78K):
[in this window]
[in a new window]

  		FIGURE 1. Expression pattern of BTN2A1. A, PCR analysis of BTN2A1 mRNA using multiple tissue panels (top). B, Tissue sections of normal colon were stained for BTN2A1 expression by IHC using diaminobenzidine. A representative section shows strong staining of epithelial cells and also some staining of lymphocytes. C, Western blots of cell lysates (15 µg/lane) with anti-BTN2A1 Ab (top) and anti beta-actin Ab (bottom). The ~69 kDa BTN2A1 protein band was prominent in cells and cell lines of various tissues, but less pronounced in PBLs and Mo.

 
BTN2A1-Ig binds to MoDCs, not to LCs

To identify the counterreceptor for BTN2A1, we constructed a fusion protein comprising its ectodomain with the Fc portion of human IgG1. The purified recombinant fusion protein was used to detect the presence of a counterreceptor by flow cytometry. First, we studied lymphocyte subsets including T cells, as BTN3A1 bound these cells (5). PMA/calcium ionophore (CI)-activated and nonactivated MOLT-4 T cells did not bind to BTN2A1, nor did NK-cells or B cells (Fig. 2A). Other tissues tested, including Hela cells, were also negative (data not shown). In contrast, MoDCs, but not epidermal LCs bound the BTN2A1 fusion protein with high intensity (Fig. 2B). We were concerned to rule out binding to Fc receptors expressed on DCs. Titration experiments showed only minimal binding of an hFc protein, hIgG, or hBTN3A3-Ig, a close homologue of BTN2A1, whereas BTN2A1-Ig bound in a dose-dependent manner (Fig. 2C, data not shown).


Figure 2
View larger version (31K):
[in this window]
[in a new window]

  		FIGURE 2. Putative BTN2A1-counterreceptor expression on MoDCs. A panel of cells and cell lines was screened for binding to soluble BTN2A1-Ig (10 µg/ml) by flow cytometry. A, BTN2A1 does not interact with untreated or PMA/CI-activated MOLT-4 cells, or with NK-cells or B-LCL YT or Raji. B, BTN2A1 binds to MoDCs, but not to LCs. Binding of soluble BTN2A1-Ig (10 µg/ml, open histogram) to MoDCs and LCs is shown vs binding to hFc protein (black-filled histogram). The phenotype of MoDCs used was HLA-DRhigh, CD1high, and CD14low. C, Binding of BTN2A1-Ig is concentration-dependent. BTN2A1 concentrations were 5, 12, 25 µg/ml (dotted, broken, and solid lines, respectively). Background binding of hFc (14, 35, 70 µg/ml) and hIgG (7, 18, 35 µg/ml) used as control proteins was low.

 
BTN2A1 counterreceptor expression is up-regulated by IL-4 and down-regulated on mature DCs

Immature MoDCs (imMoDCs) express low levels of B7 costimulatory molecules such as B7.2 (Fig. 3A). When triggered by microbial stimuli, pattern recognition receptors mediate maturation into immunogenic DCs that express high levels of MHC class II molecules such as HLA-DR and costimulatory molecules such as B7.1 and B7.2 (Fig. 3A, data not shown). To determine how DC activation affects BTN2A1 counterreceptor expression, we compared imMoDCs to LPS-treated mature MoDCs (mMoDCs). After LPS-activation, BTN2A1-counterreceptor expression diminished significantly (Fig. 3A). Down-regulation of the putative BTN2A1 counterreceptor was also observed on DCs activated by TNF-{alpha} (data not shown). As expected, a CTLA-4-Ig fusion protein showed a reciprocal staining pattern, bright on mMoDCs and low on imMoDCs, in accordance with the regulation of its ligands B7.1 and B7.2 (Fig. 3A). Human monocytes differentiated into a homogenous population of CD1ahigh, CD14low imDCs and expressed the BTN2A1 counterreceptor (Fig. 3B). To determine the factors responsible for the induction of the counterreceptor on differentiating DCs, monocytes were isolated and treated with IL-4. CD14 expression was low, as was CD1a expression, in contrast to IL-4/GM-CSF-treated cells (Fig. 3B, data not shown). However, IL-4 stimulated monocytes acquired significant levels of the BTN2A1-counterreceptor. Counterreceptor expression levels were similar on IL-4/GM-CSF treated cells, suggesting that expression was primarily IL-4 mediated. Expression of the counterreceptor was high at 48 h after IL-4 stimulation, indicating that it is acquired early during MoDC differentiation (Fig. 3B).


Figure 3
View larger version (29K):
[in this window]
[in a new window]

  		FIGURE 3. Putative BTN2A1-counterreceptor expression is IL-4-dependent and is down-regulated on mDCs. A, MoDCs were activated with LPS for 48 h and stained for B7.2 expression using M{alpha}CD86-RPE. Mo-derived imDC (top) and mDCs (bottom) binding to BTN2A1-Ig or CTLA-4-Ig was analyzed by flow cytometry. Specific binding is shown as open histograms vs binding of hIgG (filled histograms). B, Monocytes, isolated by plastic adherence, were incubated for 72 h in the presence of IL-4 (top) or a combination of IL-4/GM-CSF (bottom). Cells were stained with soluble BTN2A1-Ig, M{alpha}CD1a, or Rat{alpha}DC-SIGN-RPE. Specific Ab or fusion protein binding is shown as open histogram vs isotype control or hIgG binding, respectively (filled histogram).

 
The counterreceptor for BTN2A1 is an endocytic C-type lectin

MoDCs express a variety of cell surface receptors that exhibit dual functions as Ag receptors and as cellular adhesion receptors. To help distinguish between them, we tested whether the BTN2A1 counterreceptor could function as an endocytic receptor. MoDCs were incubated with BTN2A1-Ig under saturating conditions and transferred from 4°C to 37°C. Receptor internalization was stopped at 4°C at various time points. Cell surface binding of BTN2A1-Ig was analyzed by flow cytometry using a FITC-conjugated goat F(ab')2 anti-human IgG. After 5 min, >80% of the ligand had been removed from the cell surface (Fig. 4A). External loss of bound BTN2A1-Ig correlated with appearance of internalized BTN2-Ig as observed in permeabilized DCs (data not shown). Thus, bound BTN2A1 was rapidly internalized from the cell surface, consistent with endocytosis. Together, the IgV- and the IgC-like domains of the BTN2A1 monomer contain four potential N-linked glycosylation sites. To evaluate the role of sugars in BTN2A1 binding to DCs, we conducted a series of inhibition experiments. Interaction of BTN2A1-proteins with DCs required Ca2+, as cation chelators, such as EDTA and EGTA were inhibitory in a dose-dependent manner (Fig. 4B). Whereas monosaccharides blocked BTN2A1-Ig binding only weakly (data not shown), mannan was a potent, dose-dependent inhibitor (Fig. 4C). The significance of carbohydrates for binding, in particular high-mannose-type oligosaccharides, was further corroborated by Endo H digestion. Endo H-treated BTN2A1-Ig failed to bind to DCs (Fig. 4D). To rule out binding to high-mannose carbohydrate on the Fc-domain, we also tested binding of hFc and hBTN3A3-Ig. HFc and hBTN3A3, expressed in HEK293T like BTN2A1, did not bind (Fig. 2, data not shown). Thus, the counterreceptor recognized mannose moieties on BTN2A1, most likely branched mannose-structures. Taking these data together, the profile of BTN2A1 binding to MoDCs indicates that its counterreceptor belongs to the C-type lectin family.


Figure 4
View larger version (18K):
[in this window]
[in a new window]

  		FIGURE 4. BTN2A1-counterreceptor is a C-type lectin and rapidly internalized. A, BTN2A1 protein is rapidly internalized after ligation. DCs were labeled with BTN2A1-Ig (10 µg/ml) on ice and incubated at 37°C for 0 min (solid line), 5 min (dotted line), and 15 min (scattered line) (see inset). Internalization was stopped at 4°C and bound BTN2A1-Ig detected by flow cytometry. Percent DC-BTN2A1-Ig binding was calculated as follows: [(MFI of sample – MFI of negative control)/MFI of positive control – MFI of negative control] x 100. B, Binding of BTN2A1 protein to DCs is Ca-dependent. DCs were preincubated on ice for 20 min in the presence of 1 mM Ca2+ with increasing concentrations of Ca2+-chelator, before soluble BTN2A1-Ig (10 µg/ml) fusion protein was added. Binding without inhibitor (solid line) was compared with binding in the presence of 5 mM or 20 mM EDTA and EGTA (dotted or scattered lines, respectively). C, Binding of BTN2A1-Ig is blocked by mannan and inhibition is dose-dependent. MoDCs were preincubated as described in (B) with increasing doses of mannan and analyzed for binding of BTN2A1-Ig by flow cytometry (see also inset). D, BTN2A1 binding is abrogated by Endo H digestion. BTN2A1-Ig was treated with Endo H (1 mU/µg) at RT overnight and binding to DCs was analyzed by flow cytometry.

 
BTN2A1 binds DC-SIGN transfectants and binding is blocked by specific Abs and HIVgp120

DC-SIGN was first identified as a C-type lectin that binds to HIV envelope glycoprotein gp120 (16). This molecule is predominantly expressed on DCs, including MoDCs, but not LCs (17). Its expression is IL-4 dependent and is negatively regulated by LPS and TNF-{alpha} (18). We confirmed that DC-SIGN is IL-4 inducible and its induction peaks at 48 h (Fig. 3). Thus, DC-SIGN has several hallmarks exhibited by the BTN2A1 counterreceptor. To test this, we cloned and expressed the full-length DC-SIGN cDNA in HEK293T cells. Staining with a specific mAb confirmed transient DC-SIGN expression on a majority of the HEK293T transfectants. BTN2A1 bound to DC-SIGN-transfectants, but not to nontransfected cells (Fig. 5A and data not shown). Interaction was specific for the BTN2A1 domain of the chimera because a control Fc protein did not bind. We also assayed the interaction in the presence of an inhibitory Ab. As shown in Fig. 5A, binding of BTN2A1 was inhibited by preincubation with the DC-SIGN-blocking Ab 1B10 but was not affected by an isotype control. We then addressed the question of whether DC-SIGN is the only receptor on DCs or whether there are other C-type lectins interacting with BTN2A1. The 1B10 Ab completely abrogated binding to DCs (Fig. 5B), suggesting that DC-SIGN is the exclusive receptor for BTN2A1 on DCs. Thus, the biochemistry, Ig-fusion protein binding, and Ab blocking studies were all consistent with DC-SIGN as a counterreceptor of BTN2A1.


Figure 5
View larger version (9K):
[in this window]
[in a new window]

  		FIGURE 5. BTN2A1 binds to DC-SIGN-transfectants with high affinity and binding is blocked by DC-SIGN-specific Ab. A, Binding of soluble BTN2A1-Ig (10 µg/ml) and hFc (10 µg/ml) to HEK293T cells transfected with DC-SIGN and blocking of BTN2A1 binding by a DC-SIGN-blocking Ab (1B10, 35 µl/ml) or an isotype control (IgG2a, 35 µg/ml, BD Biosciences). The number of cells that were positive for "Zenon Alexa Fluor 488"-labeled BTN2A1-Ig was analyzed by flow cytometry. The data are representative of three independent experiments. B, BTN2A1 binding and inhibition of binding to DCs by 1B10 as described in (A). C, Binding of BTN2A1 (10 µg/ml) and inhibition of binding to DC-SIGN-transfectants by HIVgp120.

 
DC-SIGN is an HIV receptor, and we explored whether BTN2A1 and HIVgp120 compete for binding. Both molecules bind to DC-SIGN via high-mannose oligosaccharides and we found that binding of BTN2A1-Ig to DC-SIGN-transfectants was blocked by recombinant gp120 in a dose-dependent manner (Fig. 5C). Thus, recombinant BTN2A1 is a candidate to inhibit transmission of HIV to DCs.

Tissue- and/or tumor-specific glycosylation of native BTN2A1 is recognized by DC-SIGN

We demonstrated that a soluble BTN2A1-fusion protein binds to DC-SIGN on MoDCs. To obtain direct evidence for the interaction of DC-SIGN to native BTN2A1, we used DC-SIGN-FLAG for immunoprecipitation. Lysates of HEK293T expressing BTN2A1 endogenously (Fig. 1C) were precipitated and separated by SDS-PAGE. By Western blotting with an Ab to the recombinant B30.2 domain of BTN2A1, full-length BTN2 was detected, whereas a control immunoprecipitation was negative (Fig. 6A).


Figure 6
View larger version (24K):
[in this window]
[in a new window]

  		FIGURE 6. DC-SIGN binds to BTN2A1 on human tumor cell lines but not on HUVECs and HFF. A, HEK293T lysate immunoprecipitated with DC-SIGN-FLAG or a FLAG-BAP-protein bound to M{alpha}FLAG-agarose was analyzed on a PVDF membrane using rat mAb to BTN2A1 B30.2 domain. A full-length BTN2A1/2/3 protein at ~69kD was detected in the cell lysate and precipitate, but not in the control. B, PMNs, HEK293 T cells, and PBLs were analyzed for sDC-SIGN binding. SDC-SIGN binding (open histogram) is shown vs binding of a BAP-FLAG protein (black filled histogram). C, A panel of human cells and cell lines was transiently transfected with BTN2A1- or BTN3A3-GFP. Relative binding of DC-SIGN-FLAG was plotted against the level of BTN2A1-GFP expression (FITC Log). An arbitrary baseline was set to 100% binding for the mean FL 6-fluorescence intensity of DC-SIGN binding to untransfected cells. Percentage of normalized binding was calculated as the mean FL 6 ratio of transfected to untransfected cells.

 
As we showed in Fig. 1, BTN2A1 is expressed in most tissues and we wondered, therefore, why DC-SIGN has not been reported to exhibit widespread binding. DC-SIGN has been shown to bind selective oligosaccharide residues on various proteins. Therefore, binding of BTN2A1 to DC-SIGN could vary if glycosylation of the butyrophilin molecule is cell-type specific. To address this, we studied binding of DC-SIGN-FLAG to cells by flow cytometry. As a positive control, we first confirmed strong binding of the DC-SIGN fusion protein to MAC-1 and CEACAM-1 on neutrophils (Fig. 6B, (13)). We then showed that HEK293T cells also bound sDC-SIGN, in accordance with our immunoprecipitation data. However, we could not detect PBL binding to DC-SIGN (Fig. 6B).

A potential problem with clearly demonstrating that cell surface BTN2A1 binds DC-SIGN is that the lectin has been assigned a number of different cellular ligands (13, 17, 19). To overcome this problem, GFP-labeled BTN2A1 was transfected into a range of cells and DC-SIGN binding was assayed in relation to the level of BTN2A1 expression. BTN3A3, a homologue to BTN2A1 that does not bind to DC-SIGN (Fig. 6, data not shown) was used as a negative control. The cells expressed some BTN2A1 endogenously so the increase in binding of DC-SIGN after transfection of BTN2A1 was measured. There was a marked increase of DC-SIGN binding to HEK293T, Hela, and MelJuSo, as BTN2A1 expression increased (Fig. 6C). Interestingly, binding to transfected HUVECs, human foreskin fibroblasts (HFF), in contrast, was unchanged with increasing BTN2A1-GFP expression levels (Fig. 6C). This suggests that BTN2A1 on these cells does not contain the appropriate carbohydrate moieties for binding to DC-SIGN. Together, the data confirm that BTN2A1 is recognized by DC-SIGN, but only on certain tissues.

GNA inhibits binding of DC-SIGN and binds to BTN2A1 on tumor cells such as HEK293T, not on HUVECs

We were interested to understand why DC-SIGN binds to some but not other cells. We used the BTN2A1-GFP transfectants (Fig. 6) to study binding of DC-SIGN in the presence of GNA. GNA, a strictly mannose-binding plant lectin, blocked binding of DC-SIGN to HEK293T. Binding was inhibited by ~80% (Fig. 7A). Thus, mannose carbohydrates are mandatory in binding although we cannot rule out other carbohydrates being involved. Next, we used GNA to assess mannose-moieties of BTN2A1. GNA bound strongly to BTN2A1 on HEK293T and not to the related BTN3A3. However, GNA binding to BTN2A1 on HUVECs was low (Fig. 7B). Taking these data together, we demonstrated that BTN2A1 is differentially decorated with high-mannose moieties that determine binding to DC-SIGN.


Figure 7
View larger version (10K):
[in this window]
[in a new window]

  		FIGURE 7. GNA inhibits binding of DC-SIGN and binds to BTN2A1 on HEK293T, not HUVECs. A, HEK293T was transiently transfected with BTN2A1-GFP (see Fig. 6C) and binding of DC-SIGN-FLAG monitored in the presence and absence of GNA (100 µg/ml). Relative binding was normalized as described in Fig. 6C. B, HEK293T and HUVECs were transiently transfected with BTN2A1- and BTN3A3-GFP as described in Fig. 6C. Relative binding of GNA (20 µg/ml) was plotted against the level of BTN2A1-GFP expression (FITC Log) and evaluated as described in Fig. 6C.

 

   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
Disclosures
 References
 
The butyrophilin family member BTN3A1 and mouse BTNL2 were recently shown to interact with T cells. The murine BTNL2 functioned as an inhibitor of T cell activation (5, 6). Using an Ig-fusion protein, produced in HEK293T cells, we demonstrated that, rather than interacting with T cells, BTN2A1 is recognized by immature MoDCs. Biochemical "fingerprinting" and Ab-blocking studies revealed that DC-SIGN was the receptor expressed on immature MoDCs interacting with BTN2A1. A mAb directed against the DC-SIGN carbohydrate-recognition domain blocked binding of BTN2A1 (20). Other C-type lectins were not involved in the interaction of BTN2A1 with DC-SIGN.

Crystal structures of the carbohydrate recognition domain, in combination with binding studies, revealed that DC-SIGN has a dual binding specificity and selectively recognizes endogenous high-mannose oligosaccharides in addition to fucose-containing glycans (21, 22). These studies predicted binding of DC-SIGN to other cell surface or soluble glycoproteins with appropriately displayed high-mannose oligosaccharides. Interaction of DC-SIGN with endogenous glycans and HIVgp120 results from high affinity binding to a characteristic internal feature of high-mannose oligosaccharides (21). Our studies are consistent with BTN2A1 on some cells, but not all, having appropriate high-mannose moieties for binding to DC-SIGN.

We did not detect binding of DC-SIGN to PBLs. This could be explained by their low level of BTN2A1 expression. However, PBLs have high levels of ICAM-3, an alternative DC-SIGN ligand that was shown to support primary immune responses (17). Bogoevska et al. (23) showed DC-SIGN binding to ICAM-3 on PMNs but not on T cells. ICAM-3 on T cells is not fucosylated and lacks the appropriate mannose structures for binding, whereas PMNs bind DC-SIGN via fucosylated ICAM-3 (23). Our data are consistent with these observations.

DC-SIGN bound to BTN2A1 expressed on HEK293T, Hela, and MelJuSo cells. In contrast, primary cells such as HUVECs and HFF failed to bind, despite similar BTN2A1 expression levels. However, we found that BTN2A1 of HEK293T have more high-mannose moieties in comparison to HUVECs and those high-mannose moieties are instrumental for binding to DC-SIGN. DC-SIGN did not bind to BTN2A1 expressed on any normal tissues we have studied so far. This is reminiscent of the CEA, a tumor-associated Ag, reported to bind to DC-SIGN via Lewisx and Lewisy carbohydrates. CEA-glycosylation is dysregulated in a number of malignant tissues and DCs recognize the tumor-specific glycosylation on colorectal cancer cells through DC-SIGN (12). Thus, there is a precedent for binding of DC-SIGN molecules being restricted to tumor tissues. It is also possible that glycosylation of specific tissue-types is responsible for binding to tumor cells.

However, it is too early to conclude that transformed tissues are the target of DC-SIGN. Binding of ligands such as Mac-1, CEACAM-1, and ICAM-3 appears to be due to tissue-specific, rather than tumor-specific, glycosylation patterns (13, 23, 24). The tumor cell-binding data may be reinterpreted as being due to tissue-specific glycosylation patterns that happen to be found on tumor cells. Is BTN2A1 binding dependent on tumor-specific or tissue-specific glycosylation? What we know so far is that DC-SIGN binds to BTN2A1 only on cells decorated with oligomannosylated moieties. Further investigations on an extended panel of tissues, ideally, on pairs of normal and tumor tissue, as well as primary tumors are needed to resolve this issue.

We, and others, cloned a molecule related to DC-SIGN called L-SIGN or DC-SIGNR, which is expressed in liver sinusoids, placental capillaries, and the endothelium of lymph node sinuses (25, 26). Both SIGN receptors bind to pathogens such as HIV, CMV, ebola, and mycobacteria, and selectively recognize endogenous high-mannose oligosaccharides. Because DC-SIGN and L-SIGN share the structural basis for selective recognition of high-mannose oligosaccharides it is reasonable to assume that BTN2A1 also interacts with L-SIGN, implying a functional interaction of BTN2A1 beyond DCs (21).


   Acknowledgments
 
We thank Lucy Gardner for help with IHC, Dr. F. Arenzano-Seisdedos (Institute Pasteur, Paris) for DC-SIGN mAb 1B10, Dr. Yoshinaga (Amgen) for CTLA-4-Ig, Dr. M. Douglas (University of Birmingham) for ICAM3-Ig plasmid, Dr. L. Williams (NIBSC Centralised Facility for AIDS Reagents) for rHIV-1SF2 gp120, Drs. P. McAry (CIMR, Cambridge) for LCs, Drs. Reija Valtola, Jose Garcia-Bernardo, C. Chang, and Tina Rich (Department of Pathology) for HUVECs, BAF/3, and IMR-90. We also thank Prof. K. Drickamer and Dr. M. Taylor (Imperial College London) for a very fruitful discussion.


   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 Wellcome Trust. Back

2 Address correspondence and reprint requests to Dr. Georg Malcherek, Department of Pathology, Division of Immunology, Tennis Court Road, Cambridge, U.K. E-mail address: gfjm2{at}cam.ac.uk Back

3 Current address: Medical Research Council, Laboratory of Molecular Biology, Hills Road, Cambridge, U.K. Back

4 Current address: Max Planck Institute for Molecular Physiology, Department of System Cell Biology, Otto-Hahn-Str. 11, Dortmund, Germany Back

5 Abbreviations used in this paper: BTN, butyrophilins; BTNL2, butyrophilin-like 2; DC, dendritic cell; DC-SIGN, DC-specific ICAM-3 grabbing nonintegrin; CEA, carcinoembryonic Ag; MoDC, monocyte-derived DCs; LC, Langerhans cells; GNA, Galanthus nivalis agglutinin; IHC, immunohistochemistry; imMoDC, immature MoDCs; Endo H, Endoglycosidase H; TRIM, tripartite motif; CI, calcium ionophore; HFF, human foreskin fibroblasts. Back

Received for publication April 24, 2007. Accepted for publication July 12, 2007.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 Disclosures
 References
 

  1. Henry, J., M. M. Miller, P. Pontarotti. 1999. Structure and evolution of the extended B7 family. Immunol. Today 20: 285-288. [Medline]
  2. Rhodes, D. A., M. Stammers, G. Malcherek, S. Beck, J. Trowsdale. 2001. The cluster of BTN genes in the extended major histocompatibility complex. Genomics 71: 351-362. [Medline]
  3. Ogg, S. L., A. K. Weldon, L. Dobbie, A. J. Smith, I. H. Mather. 2004. Expression of butyrophilin (Btn1a1) in lactating mammary gland is essential for the regulated secretion of milk-lipid droplets. Proc. Natl. Acad. Sci. USA 101: 10084-10089. [Abstract/Free Full Text]
  4. Linsley, P. S., R. Peach, P. Gladstone, J. Bajorath. 1994. Extending the B7 (CD80) gene family. Protein Sci. 3: 1341-1343. [Medline]
  5. Compte, E., P. Pontarotti, Y. Collette, M. Lopez, D. Olive. 2004. Frontline: characterization of BT3 molecules belonging to the B7 family expressed on immune cells. Eur. J. Immunol. 34: 2089-2099. [Medline]
  6. Nguyen, T., X. K. Liu, Y. Zhang, C. Dong. 2006. BTNL2, a butyrophilin-like molecule that functions to inhibit T cell activation. J. Immunol. 176: 7354-7360. [Abstract/Free Full Text]
  7. Arnett, H. A., S. S. Escobar, E. Gonzalez-Suarez, A. L. Budelsky, L. A. Steffen, N. Boiani, M. Zhang, G. Siu, A. W. Brewer, J. L. Viney. 2007. BTNL2, a butyrophilin/B7-like molecule, is a negative costimulatory molecule modulated in intestinal inflammation. J. Immunol. 178: 1523-1533. [Abstract/Free Full Text]
  8. Bieniasz, P. D.. 2004. Intrinsic immunity: a front-line defense against viral attack. Nat. Immunol. 5: 1109-1115. [Medline]
  9. Banchereau, J., R. M. Steinman. 1998. Dendritic cells and the control of immunity. Nature 392: 245-252. [Medline]
  10. McGreal, E. P., J. L. Miller, S. Gordon. 2005. Ligand recognition by antigen-presenting cell C-type lectin receptors. Curr. Opin. Immunol. 17: 18-24. [Medline]
  11. Geijtenbeek, T. B., Y. van Kooyk. 2003. Pathogens target DC-SIGN to influence their fate DC-SIGN functions as a pathogen receptor with broad specificity. APMIS 111: 698-714. [Medline]
  12. van Gisbergen, K. P., C. A. Aarnoudse, G. A. Meijer, T. B. Geijtenbeek, Y. van Kooyk. 2005. Dendritic cells recognize tumor-specific glycosylation of carcinoembryonic antigen on colorectal cancer cells through dendritic cell-specific intercellular adhesion molecule-3-grabbing nonintegrin. Cancer Res. 65: 5935-5944. [Abstract/Free Full Text]
  13. van Gisbergen, K. P., I. S. Ludwig, T. B. Geijtenbeek, Y. van Kooyk. 2005. Interactions of DC-SIGN with Mac-1 and CEACAM1 regulate contact between dendritic cells and neutrophils. FEBS Lett. 579: 6159-6168. [Medline]
  14. Fawcett, J., C. L. Holness, L. A. Needham, H. Turley, K. C. Gatter, D. Y. Mason, D. L. Simmons. 1992. Molecular cloning of ICAM-3, a third ligand for LFA-1, constitutively expressed on resting leukocytes. Nature 360: 481-484. [Medline]
  15. MacAry, P. A., M. Lindsay, M. A. Scott, J. I. Craig, J. P. Luzio, P. J. Lehner. 2001. Mobilization of MHC class I molecules from late endosomes to the cell surface following activation of CD34-derived human Langerhans cells. Proc. Natl. Acad. Sci. USA 98: 3982-3987. [Abstract/Free Full Text]
  16. Curtis, B. M., S. Scharnowske, A. J. Watson. 1992. Sequence and expression of a membrane-associated C-type lectin that exhibits CD4-independent binding of human immunodeficiency virus envelope glycoprotein gp120. Proc. Natl. Acad. Sci. USA 89: 8356-8360. [Abstract/Free Full Text]
  17. Geijtenbeek, T. B., R. Torensma, S. J. van Vliet, G. C. van Duijnhoven, G. J. Adema, Y. van Kooyk, C. G. Figdor. 2000. Identification of DC-SIGN, a novel dendritic cell-specific ICAM-3 receptor that supports primary immune responses. Cell 100: 575-585. [Medline]
  18. Relloso, M., A. Puig-Kroger, O. M. Pello, J. L. Rodriguez-Fernandez, G. de la Rosa, N. Longo, J. Navarro, M. A. Munoz-Fernandez, P. Sanchez-Mateos, A. L. Corbi. 2002. DC-SIGN (CD209) expression is IL-4 dependent and is negatively regulated by IFN, TGF-beta, and anti-inflammatory agents. J. Immunol. 168: 2634-2643. [Abstract/Free Full Text]
  19. Geijtenbeek, T. B., D. J. Krooshoop, D. A. Bleijs, S. J. van Vliet, G. C. van Duijnhoven, V. Grabovsky, R. Alon, C. G. Figdor, Y. van Kooyk. 2000. DC-SIGN-ICAM-2 interaction mediates dendritic cell trafficking. Nat. Immunol. 1: 353-357. [Medline]
  20. Halary, F., A. Amara, H. Lortat-Jacob, M. Messerle, T. Delaunay, C. Houles, F. Fieschi, F. Arenzana-Seisdedos, J. F. Moreau, J. Dechanet-Merville. 2002. Human cytomegalovirus binding to DC-SIGN is required for dendritic cell infection and target cell trans-infection. Immunity 17: 653-664. [Medline]
  21. Feinberg, H., D. A. Mitchell, K. Drickamer, W. I. Weis. 2001. Structural basis for selective recognition of oligosaccharides by DC-SIGN and DC-SIGNR. Science 294: 2163-2166. [Abstract/Free Full Text]
  22. Guo, Y., H. Feinberg, E. Conroy, D. A. Mitchell, R. Alvarez, O. Blixt, M. E. Taylor, W. I. Weis, K. Drickamer. 2004. Structural basis for distinct ligand-binding and targeting properties of the receptors DC-SIGN and DC-SIGNR. Nat. Struct. Mol. Biol. 11: 591-598. [Medline]
  23. Bogoevska, V., P. Nollau, L. Lucka, D. Grunow, B. Klampe, L. M. Uotila, A. Samsen, C. G. Gahmberg, C. Wagener. 2007. DC-SIGN binds ICAM-3 isolated from peripheral human leukocytes through Lewis x residues. Glycobiology 17: 324-333. [Abstract/Free Full Text]
  24. Bogoevska, V., A. Horst, B. Klampe, L. Lucka, C. Wagener, P. Nollau. 2006. CEACAM1, an adhesion molecule of human granulocytes, is fucosylated by fucosyltransferase IX and interacts with DC-SIGN of dendritic cells via Lewis x residues. Glycobiology 16: 197-209. [Abstract/Free Full Text]
  25. Soilleux, E. J., R. Barten, J. Trowsdale. 2000. DC-SIGN; a related gene, DC-SIGNR; and CD23 form a cluster on 19p13. J. Immunol. 165: 2937-2942. [Abstract/Free Full Text]
  26. Bashirova, A. A., T. B. Geijtenbeek, G. C. van Duijnhoven, S. J. van Vliet, J. B. Eilering, M. P. Martin, L. Wu, T. D. Martin, N. Viebig, P. A. Knolle, et al 2001. A dendritic cell-specific intercellular adhesion molecule 3-grabbing nonintegrin (DC-SIGN)-related protein is highly expressed on human liver sinusoidal endothelial cells and promotes HIV-1 infection. J. Exp. Med. 193: 671-678. [Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 J. Biol. Chem.Home page
J. Jeong, A. U. Rao, J. Xu, S. L. Ogg, Y. Hathout, C. Fenselau, and I. H. Mather
The PRY/SPRY/B30.2 Domain of Butyrophilin 1A1 (BTN1A1) Binds to Xanthine Oxidoreductase: IMPLICATIONS FOR THE FUNCTION OF BTN1A1 IN THE MAMMARY GLAND AND OTHER TISSUES
J. Biol. Chem., August 14, 2009; 284(33): 22444 - 22456.
[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 Malcherek, G.
Right arrow Articles by Trowsdale, J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Malcherek, G.
Right arrow Articles by Trowsdale, J.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS