Journal of Leukocyte Biology eBioscience full spectrum cell analysis
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Originally published online as doi:10.1189/jlb.1005559 on July 20, 2006

Published online before print July 20, 2006
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
jlb.1005559v1
80/4/787    most recent
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
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 reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Avril, T.
Right arrow Articles by Crocker, P. R.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Avril, T.
Right arrow Articles by Crocker, P. R.
(Journal of Leukocyte Biology. 2006;80:787-796.)
© 2006 by Society for Leukocyte Biology

Probing the cis interactions of the inhibitory receptor Siglec-7 with {alpha}2,8-disialylated ligands on natural killer cells and other leukocytes using glycan-specific antibodies and by analysis of {alpha}2,8-sialyltransferase gene expression

Tony Avril*,1, Simon J. North{dagger}, Stuart M. Haslam{dagger}, Hugh J. Willison{ddagger} and Paul R. Crocker*,2

* Division of Cell Biology and Immunology, Wellcome Trust Biocentre, University of Dundee, Dundee, United Kingdom;
{dagger} Division of Molecular Biosciences, Imperial College London, South Kensington Campus, United Kingdom; and
{ddagger} University Department of Neurology, Southern General Hospital, Glasgow, United Kingdom

2 Correspondence: Wellcome Trust Biocentre, School of Life Sciences, University of Dundee, Dow Street, Dundee DD1 5EH, UK. E-mail: p.r.crocker{at}dundee.ac.uk

ABSTRACT

Siglec-7 is a CD33-related sialic acid-binding Ig-like lectin expressed strongly on NK cells, where it can function as an inhibitory receptor. Its sialic acid-binding activity on NK cells is masked by cis interactions with sialylated glycans, which are likely to be important for regulating the inhibitory function of Siglec-7, which exhibits an unusual preference for {alpha}2,8-linked disialic acids, a motif found in "b-series" gangliosides and some glycoproteins. To investigate the presence of {alpha}2,8-linked disialic acids on NK cells, T cells, monocytes, and B cells, we first analyzed their expression of all known {alpha}2,8-sialyltransferase genes by quantitative PCR. Unlike T cells, B cells, and monocytes, NK cells consistently expressed mRNA encoding ST8Sia VI, which creates {alpha}2,8-linked disialic acids on O-linked glycans of glycoproteins. All blood leukocytes expressed ST8Sia IV, implicated in polysialic acid synthesis, and NK cells variably expressed high levels of ST8Sia V mRNA required for GT3 expression. Two human IgM antibodies, Ha1 and Pi1, with specificity for the {alpha}2,8-disialyl motif reacted strongly with NK cells in a sialic acid-dependent manner and less strongly with T cells and monocytes. Antibody-induced clustering of Siglec-7 on NK cells resulted in partial colocalization with anti-Ha1. Finally, MALDI-TOF mass spectrometric analysis of isolated NK cell O-glycans revealed the presence of a peak at mass-to-charge ratio of 1619.4 mass units, corresponding to a putative {alpha}2,8-disialylated glycan. Together, these results suggest that NK cells are decorated with {alpha}2,8-disialic acid structures implicated in regulation of cellular activation via interactions with Siglec-7.

Key Words: NK cells • sialic acid • MALDI-TOF

INTRODUCTION

Siglecs are transmembrane sialic acid-binding receptors of the Ig superfamily expressed in the hemopoietic, immune, and nervous systems, where they play roles in cellular adhesion and signaling [1 , 2 ]. The CD33-related siglecs are a distinct subset, which are expressed predominantly by cells of the innate immune system and appear to be undergoing rapid and continuous evolution [3 ] (Table 1 ). In humans, there are eight CD33-related siglecs, all of which contain two conserved, tyrosine-based signaling motifs, comprising a membrane proximal ITIM and an ITIM-like motif. Tyrosine phosphorylation of the ITIM can lead to recruitment of Src homology-2-containing tyrosine phosphatase 1 (SHP-1) and SHP-2 tyrosine phosphatases and initiation of inhibitory signaling [12 ]. Recent studies have raised the possibility that low-level, constitutive recruitment of tyrosine phosphatases by CD33-related siglecs could be important for maintaining a tonic suppression of leukocyte activation [13 ].


View this table:
[in this window]
[in a new window]
  		Table 1. The Human Siglec Family

 
When expressed naturally by leukocytes, the ligand-binding activity of all CD33-related siglecs examined so far is masked at the cell surface as a result of cis interactions with uncharacterized, sialylated ligands expressed on the same cells [14 ]. As demonstrated recently for the B cell inhibitory receptor CD22/Siglec-2 [15 ], interactions of siglecs with sialylated glycans can play a key role in regulating leukocyte activation. It has been proposed that CD33-related siglecs have become adapted to the sialic acid "landscape" of the cell type on which they are expressed to promote cis interactions and hence, regulate leukocyte activation events [2 , 16 ]. An attractive feature of this hypothesis is that cell surface glycosylation patterns can be regulated temporally according to the dynamic pattern of glycosyltransferase gene expression. This could influence the nature of cis interactions of siglecs with relevant counter-receptors and hence, regulate inhibitory signaling functions in a controlled manner. To understand the link between CD33-related siglec function and glycan recognition, it is clearly important to identify potential ligands on leukocyte populations and characterize the expression patterns of the relevant glycosyltransferases.

Siglec-7 is the major CD33-related siglec expressed by human NK cells [17 , 18 ]. It is also present at lower levels on monocytes and a subset of CD8+ T cells [18 ] (Table 1) . Siglec-7 is able to inhibit NK cell cytotoxicity in redirected killing experiments with mAb [17 ] and in direct cytotoxicity against GD3-expressing target cells [11 ]. Recently, we showed that Siglec-7 is able to function as an inhibitory receptor by recruiting SHP-1 and -2 via its two intracellular, tyrosine-based motifs [12 ]. It is interesting that, and unlike other CD33-related siglecs, when Siglec-7 was expressed on transfected COS cells, it was able to mediate high-level binding to glycoprobes and human red blood cells and was therefore unmasked [18 ]. However, on NK cells, Siglec-7 was shown to be masked, suggesting that cis interactions with endogenous sialylated glycans prevent trans interactions with exogenous probes [11 ]. Several studies have demonstrated an unusual preference of Siglec-7 for {alpha}2,8-disialylated structures over terminal {alpha}2,3- and {alpha}2,6-linked sialic acids [11 , 19 20 21 ]. Disialic acids are commonly found on "b-series" gangliosides in the nervous system such as GD3, GT1b, and GQ1b, but data from Kitajima’s laboratory [22 , 23 ] have shown that these structures can also be presented on subsets of glycoproteins. Therefore, one potential explanation for the strong preference of Siglec-7 for the {alpha}2,8-disialyl motif is that NK cells naturally express this structure on glycolipids and/or glycoproteins, and this could lead to Siglec-7 masking on NK cells via cis interactions and be involved in inhibitory signaling functions of Siglec-7.

The formation of {alpha}2,8-linked sialic acids is catalyzed by a family of six {alpha}2,8-sialyltransferases, designated ST8Sia I–VI. ST8Sia I and V are the GD3/GT3 and GT1a/GQ1b/GT3 synthases, respectively, transferring sialic acid to glycolipid substrates, whereas ST8Sia II, III, and IV transfer sialic acid residues to glycoproteins [24 ]. In humans, the ST8Sia VI was recently shown to be relatively specific for O-linked glycans and required the trisaccharide Neu5Ac{alpha}2,3Galß1,3GalNAc, adding a single {alpha}2,8-linked sialic acid [24 ]. ST8Sia VI is therefore a prime candidate for the formation of Siglec-7 ligands on glycoproteins. ST8Sia II and IV catalyze the formation of polysialic acid on neural cell adhesion molecule (NCAM) [25 ]. Although NK cells are known to express polysialylated NCAM (CD56) [26 ], this is unlikely to be important for cis interactions with Siglec-7, as increasing the number of sialic acid residues in {alpha}2,8-linked oligosialic acid beyond 3 leads to reduced binding of Siglec-7 (unpublished observations).

In this study, we show that human NK cells express the {alpha}2,8-sialyltransferases ST8Sia IV, V, and VI, involved in the transfer of {alpha}2,8-linked disialic acid residues on glycolipids and glycoproteins. Using two human antibodies, Ha1 and Pi1, which preferentially bind {alpha}2,8-linked disialic acid residues, we demonstrate strong and relatively selective binding to NK cells and T cells, and in the case of NK cells, Siglec-7 is partially associated with antigens recognized by the Ha1 antibody. Finally, MALDI-TOF mass spectrometry (MS) of O-glycans isolated from NK cells revealed the presence of a peak at mass-to-charge ratio (m/z) of 1619.4 mass units, corresponding to a putative {alpha}2,8-disialylated glycan.

MATERIALS AND METHODS

Reagents and antibodies
Unless specified otherwise, all reagents and chemicals were purchased from Sigma-Aldrich (Poole, UK), and all antibodies were obtained from Caltag Laboratories (Buckingham, UK). Vibrio cholerae sialidase was obtained from Calbiochem (Nottingham, UK). The Siglec-Fc fusion proteins were described previously [27 ]. The chimeric proteins contain the entire extracellular domain of Siglec-7 and -9. The biotinylated polyacrylamide probes (PAA probes), conjugated to lactose or {alpha}2,3- or {alpha}2,8-linked disialic acid, were obtained from Syntesome (Munich, Germany). The anti-GD3 mAb, R24, was a kind gift from Dr. Paul Chapman (Memorial Sloan-Kettering Cancer Center, New York, NY) and was used as a purified IgG. The human Ha1 and Pi1 IgM mAb were described and purified as reported previously [28 , 29 ]. Biotinylation of these antibodies was performed using the EZ-link biotinylation kit (Pierce, Cramlington, UK).

Cells
The parental P815 mouse mastocytoma cells and P815 cells stably expressing GD3 synthase (GD3-P815) [11 ] were grown in RPMI 1640 (Invitrogen, Paisley, GB) supplemented with 10% FCS (Invitrogen). Human PBMC were isolated from the blood of healthy donors by Ficoll-Paque (Amersham Biosciences, Little Chalfont, UK) density gradient centrifugation. Purified B cells, T cells, and NK cells were isolated from PBMC by positive selection using a cell isolation kit and an autoMACS (Miltenyi Biotec, Bisley, GB). Monocytes were obtained from the negative fraction.

Total RNA extraction and real-time RT-PCR
RNA was extracted using a quiazol lysis reagent (Qiagen, Crawley, UK) followed by column-based purification using the RNeasy lipid tissue kit from Qiagen, according to the manufacturer’s recommendations. RNA concentrations were determined by OD260. cDNA was prepared from 1 µg purified RNA (RevertAid first-strand cDNA synthesis kit, Helena Biosciences, Sunderland, UK) and was tenfold-diluted. Real-time PCR was performed using a spectrofluorometric thermal cycler (ABI Prism 7000, Applied Biosystems, Warrington, UK) following the manufacturer’s recommendations except for the final PCR volume, which was decreased to 25 µl. The ST8Sia I–V primers and probes were purchased from Applied Biosystems (TaqMan assays-on-demand). The ST8Sia VI primers and probe were designed by Applied Biosystems (TaqMan assays-by-design). To normalize the data, 18S RNA was chosen as an endogenous control and tested in separate wells. The amplification efficiencies of the gene target and endogenous control were consistently found to be similar. Therefore, results are expressed as relative mRNA expression using the comparative threshold cycle method and are compared arbitrarily with B cell expression taken as 1. When no amplification was observed with B cells, results are compared with T cell expression taken as 1.

PAA probe-binding assay
Siglec-Fc proteins (10 µg/ml) were coated in 96-well plates overnight at 4°C. Biotinylated polyacrylamide (PAA) attached to lactose, {alpha}2,3-sialyllactose, or {alpha}2,8-linked disialic acid was complexed with alkaline phosphatase-conjugated streptavidin (1:1000) for 1 h at room temperature with rotation. Wells coated with Siglec-Fc proteins were incubated for 1 h at room temperature with precomplexed PAA probes. Unbound PAA probes were then washed off, and 100 µl per well of 10 µM fluorescein diphosphate (Molecular Probes, Eugene, OR) was added. After incubation at 37°C for 15–30 min, fluorescence readings were measured using a fluorescence plate reader (Cytofluor, PerSeptive Biosystems, Framingham, MA). PAA probe binding was expressed as arbitrary fluorescence unit (AFU).

Glycan array
The sugar-binding specificities of Ha1 antibody were analyzed by the Consortium for Functional Glycomics (CFG; http://www.functionalglycomics.org/static/consortium) using Glycan Array v2.8 as described previously [30 , 31 ]. Briefly, streptavidin-coated microtiter plates (Reacti-Bind NeutrAvidin-coated high-binding capacity black 384-well plates, Pierce, Rockford, IL) were coated with various biotinylated glycosides (30 pmol/well) in PBS, pH 7.4, overnight at 4°C. After washing, wells were incubated with 17.5 µg/ml Ha1 antibody at 4°C for 1 h followed by Alexa 488-conjugated goat anti-human IgM (Molecular Probes), and binding was measured on a fluorescent plate reader using excitation at 485 nm and emission at 535 nm. The glycosides used in this study are listed on the CFG website.

Flow cytometry
All incubations were carried out on ice. Cells were incubated with mAb (10 µg/ml) for 30 min, washed, and then analyzed using a FACSCalibur flow cytometer with CellQuest software (Becton Dickinson, Oxford, UK). For PAA probe-binding experiments, control or sialidase-treated NK cells were incubated with biotinylated PAA probes at 20 µg/ml for 1 h on ice. After washing in PBS + 0.1% BSA, cells were incubated with streptavidin-FITC (Jackson ImmunoResearch Laboratories, West Grove, PA) for 20 min, washed, and then analyzed using a FACSCalibur. For staining with biotinylated human Ha1 and Pi1 antibodies, a second incubation with FITC-conjugated streptavidin was performed.

Fluorescence microscopy
Cells were incubated with sheep anti-Siglec-7 polyclonal antibodies (pAb; 10 µg/ml) for 30 min at 4°C, washed, and then incubated with FITC-conjugated anti-sheep rabbit pAb for 30 min at 37°C to induce capping. After washes, cells were fixed with 2% formaldehyde for 10 min on ice and then stained on ice with biotinylated Ha1 antibody. After washing, cells were incubated with TRITC-conjugated streptavidin for 30 min on ice, washed, and analyzed using a Zeiss immunofluorescence microscope (Zeiss, Jena, Germany). Images were acquired using the AxioVision imaging system (Imaging Associates, Bicester, UK).

O-glycan profiling of NK cells
NK cells were isolated from PBMC by positive selection using a cell isolation kit and an autoMACS. Full details of the glycan isolation protocols are found at the following CFG website: http://www.functionalglycomics.org/coreCStatic/disclaimer.shtml. Briefly, glycoproteins extracted from washed cell pellets were reduced, carboxymethylated, and then digested with trypsin. The tryptic peptides were then treated with peptide N-glycosidase F to release the N-glycans. O-glycans were released by reductive elimination in 400 µl sodium borohydride (38 mg/ml in 0.05 M sodium hydroxide) at 45°C for 16 h. Reactions were terminated by drop-wise addition of glacial acetic acid, followed by Dowex 50W-X8 (H), 50–100 mesh chromatography, and borate removal prior to permethylation. MALDI data were acquired using a PerSeptive Biosystems Voyager-DETM STR MS in the reflectron mode with delayed extraction. Derivatized glycans were dissolved in 10 µl methanol, and 1 µl dissolved sample was premixed with 1 µl matrix (2,5-dihydrobenzoic acid) before loading onto a target plate.

Statistics
Values represent the mean ± SD of n different experiments. Student’s t-test was applied using a two-tailed distribution of two samples of equal or unequal variances.

RESULTS

Masking state of Siglec-7 on the surface of human NK cells
Among the different members of the siglec family (Table 1) , Siglec-7 has the unusual characteristic of binding preferentially to {alpha}2,8-linked disialic acid residues present on molecules such as synthetic polyacrylamide (PAA) probes (Fig. 1A ), the ganglioside GD3 [11 ], and other disialylated gangliosides [19 , 21 ] or the lipooligosaccharide of the HS:19 (GM1+GT1a+) Campylobacter jejuni strain [32 ]. In contrast, the highly related Siglec-9 bound {alpha}2,3-sialyllactose-PAA but not the {alpha}2,8-linked disialic acid structure (Fig. 1A) . As shownin Figure 1C , {alpha}2,8-linked disialic acid-PAA probes bound only to NK cells pretreated with sialidase, whereas no binding was observed with untreated NK cells. These results suggest that large amounts of sialic acid residues, possibly {alpha}2,8-linked disialic acid residues, are present on the cell surface of human NK cells and inhibit trans interactions of sialic acid structures with Siglec-7 (Fig. 1B) .


Figure 1
View larger version (43K):
[in this window]
[in a new window]
  		Figure 1. Human Siglec-7 is masked by sialic acid structures on the surface of NK cells. (A) Siglec-7- and -9-Fc chimeric proteins were used in a binding assay with lactose-PAA, {alpha}2,3-sialyllactose-PAA, or {alpha}2,8-linked disialic acid-PAA probes. (B) Purified human NK cells were untreated or treated with sialidase and incubated with lactose-PAA or {alpha}2,8-linked disialic acid-PAA probes. Binding of PAA probes was assessed by flow cytometry. (C) Schematic representation of masked and unmasked states of Siglec-7 on the cell surface.

 
ST8Sia mRNAs expressed in human PBMC
To study the expression of ST8Sia enzymes in human PBMC (and especially NK cells), B cells, T cells, NK cells, and monocytes from four different donors were purified and analyzed by quantitative RT-PCR for mRNA expression of the six different ST8Sia enzymes (Fig. 2 ). Cell purity was confirmed by flow cytometry using specific markers to each leukocyte population (Fig. 2A) . ST8sia II and III have been shown previously to be expressed predominantly in the nervous system [33 , 34 ], and as expected, no amplification of either was observed in any lymphocyte population (data not shown). However, as shown in Figure 2B , a significant amount of ST8Sia I (GD3 synthase) mRNA was found in T cells. This is in accordance with previous studies demonstrating CDw60 expression, a 9-O-acetylated form of GD3 present on glycolipids and glycoproteins, on the surface of T cells [35 ]. NK cells and to a lesser extent, monocytes expressed significant amounts of ST8Sia V (in two of four donors) and VI mRNAs. ST8Sia VI has been characterized in mice [36 ] and humans [37 ]. In humans, ST8Sia VI has been shown to transfer a single {alpha}2,8-linked sialic acid to {alpha}2,3-sialylated O-glycans as acceptors, and overexpression in mammalian cells led to a selective decrease in cell-surface {alpha}2,3-linked sialic acids, detected with the Mackia Amurensis lectin [37 ]. Therefore, the consistent expression of ST8Sia VI in NK cells lends strong support to the notion that at least a subset of their O-glycans is modified with {alpha}2,8-linked disialic acids. It is surprising that expression of ST8Sia IV mRNA was found in all leukocyte populations at variable levels. ST8sia II and IV are known to be important for polysialylation of the molecule NCAM, which is thought to be the major acceptor for polysialic acid in the body [38 ]. As polysialylated NCAM is expressed in NK cells as the CD56 marker [26 ], it would be expected that ST8Sia IV is expressed in NK cells, but this would not be the case for the other leukocyte populations, which do not express CD56/NCAM. It is possible that ST8Sia IV is not only involved in the polysialylation of the NCAM molecule but could also act on other glycoproteins as substrates. For example, it has been proposed that ST8Sia IV transfers polysialic acid to the "preferred" substrate NCAM but may transfer only one or a small number of sialic acid residues on less-preferred substrates [39 ]. Taken together, these results are consistent with the possibility that NK cells and other blood leukocytes express significant amounts of {alpha}2,8-linked di- and/or oligosialic acid residues at the cell surface.


Figure 2
View larger version (49K):
[in this window]
[in a new window]
  		Figure 2. Human {alpha}2,8-sialyltransferase mRNA expression in lymphocytes. (A) B cells, T cells, NK cells, and monocytes were purified from PBMC and tested for purity by flow cytometry using the indicated antibodies. (B) RNA was isolated from purified B cells, T cells, NK cells, and monocytes obtained from four different donors and analyzed by quantitative RT-PCR for expression of {alpha}2,8-sialyltransferases ST8Sia I–VI. No amplification was seen with ST8Sia II and III for any cell population (not shown). The results shown are displayed relative to B cells taken as 1 (or to T cells when no amplification was observed with B cells), after normalizing all cell populations to signals obtained with 18S RNA. *, No amplification was observed.

 
Characterization of human Ha1 antibody, which recognizes disialic acid structures
To investigate the potential expression of {alpha}2,8-linked disialic acid moieties on the surface of NK cells, we used human IgM mAb Ha1 and Pi1, both of which are derived from patients with peripheral neuropathy [28 , 29 ]. In previous studies, these antibodies have been shown to react similarly with b-series gangliosides GD1b, GT1b, GQ1b, and GD3 in a manner consistent with their minimal epitope being NeuAc{alpha}2,8-NeuAc{alpha}2,3-Gal, a structure present in the above gangliosides, as well as being potentially present as capping structures in O-glycans and N-glycans. To characterize their fine specificity further, Ha1 was submitted to the CFG for glycan array analysis against a panel of 172 synthetic glycans including 50 sialosides [30 , 31 ]. As shown in Figure 3A , Ha1 antibody recognized sialic acid-containing glycosides exclusively and showed a strong preference for glycans containing NeuAc{alpha}2,8-NeuAc{alpha}2,3-Gal. In comparison, the well-known R24 antibody, which has also been tested in the same glycan array, bound at high levels to the GD3 oligosaccharide but similar to the Ha1 antibody, also interacted weakly with {alpha}2,8-trisialic structures (Table 2 ). It is interesting that the glycan recognition profile observed with Ha1 antibody is similar to the profile obtained with Siglec-7-Fc chimera (Table 2) , supporting the possibility that Ha1 antibody can detect glycostructures recognized by Siglec-7.


Figure 3
View larger version (41K):
[in this window]
[in a new window]
  		Figure 3. Binding of glycan-specific antibodies to {alpha}2,8-linked disialic acid structures. (A) Ha1 antibody was tested in a glycan array containing 172 glycans, of which 50 were sialosides. The glycosides used in the array are listed on the CFG website. Data are the average ± SD of triplicate determinants. The glycans, which showed clear binding above background, are indicated in cartoon form following the glycan nomenclature established by the CFG. (B) Untreated or sialidase-treated, parental P815 and GD3 synthase-transfected P815 (GD3-P815) were labeled with anti-GD3 R24 mAb or human Ha1 and Pi1 antibodies and analyzed by flow cytometry.

 

View this table:
[in this window]
[in a new window]
  		Table 2. Comparison of Glycoside Recognition by Siglec-7-Fc, Ha1, and R24 Antibodiesa

 
To test if Ha1 and Pi1 antibodies could be used in flow cytometry experiments to probe for the presence of putative {alpha}2,8-linked disialosyl Siglec-7 ligands, we initially used parental and GD3-expressing P815 cells, which were labeled with Ha1 and Pi1 antibodies and analyzed by flow cytometry. The R24 antibody was used as a positive control. As expected, R24 bound GD3-P815 cells, whereas no staining was observed with parental P815 cells (Fig. 3B) . Strong binding of Ha1 and Pi1 antibodies was also observed with GD3-P815 cells but not with parental P815. To demonstrate that binding was sialic acid-dependent, GD3-P815 cells were pretreated with sialidase and then stained with R24, Ha1, and Pi1 antibodies. As expected, binding of these antibodies was abrogated by the sialidase treatment (Fig. 3B) .

Expression of putative disialic acid structures on human NK cells
We next used human PBL isolated from four different donors and stained them with R24, Ha1, or Pi1 antibodies in conjunction with specific markers and then analyzed labeled cells by flow cytometry. As expected, a fraction of CD3+ T cells was recognized by R24 (Fig. 4A ), probably as a result of interactions with CD60/GD3, known to be expressed by these cells [35 ]. Similar staining was observed with CD4+ and CD8+ T cells (data not shown). In addition, half or all of the T cells were stained with Ha1 and Pi1 antibodies, respectively (Fig. 4A ; Table 3 ). In all cases, antibody staining was abrogated by sialidase treatment of T cells. In contrast to T cells, only a small fraction of CD19+ B cells was recognized by all antibodies tested (Fig. 4A ; Table 3 ). With CD56+ NK cells, no staining was observed with R24, whereas Ha1 and Pi1 antibodies reacted strongly (Fig. 4A ; Table 3 ) in a sialidase-sensitive manner. These results were confirmed using purified leukocyte populations (Fig. 4B) , which in addition, showed that Pi1 bound moderately to purified monocytes (Fig. 4B) .


Figure 4
View larger version (72K):
[in this window]
[in a new window]
  		Figure 4. Expression of putative {alpha}2,8-linked disialic acid structures on human peripheral blood leukocytes. (A) Cells were labeled with anti-GD3 R24 mAb or human Ha1 and Pi1 antibodies and analyzed by flow cytometry. (B) Purified B cells, T cells, NK cells, and monocytes were stained with human Pi1 antibody and analyzed by flow cytometry. (C) Purified NK cells were stained with anti-Siglec-7 sheep pAb and incubated or not at 37°C for 30 min to induce capping. After fixation with paraformaldehyde, cells were labeled with human Ha1 antibody and analyzed by fluorescence microscopy.

 

View this table:
[in this window]
[in a new window]
  		Table 3. Ha1 and Pi1 Staining on Human Lymphocytesa

 
To test the possibility that Siglec-7 on NK cells interacts with the sialic acid structures recognized by the Ha1 antibody, we conducted cocapping experiments with human NK cells using sheep anti-Siglec-7 pAb. As shown in Figure 4C , expression of Siglec-7 on formaldehyde-fixed NK cells was uniform, as was the Ha1 antibody staining. After capping with anti-Siglec-7 pAb, colocalization of Ha1 was observed in some areas (Fig. 4C) . Taken together, these results show that sialic acid residues recognized by Ha1 antibody, potentially {alpha}2,8-linked, are expressed strongly on human NK cells and are, at least in part, associated with Siglec-7.

Finally, purified NK cells, T cells, B cells, and monocytes were submitted to the Glycan Analysis Core of the CFG. MALDI-TOF MS of isolated, permethylated N- and O-glycans revealed that for NK cells, the O-glycan fraction contained a peak at m/z 1619.4 mass units, corresponding to an {alpha}2,8-disialylated motif (Fig. 5 ). Although further experiments, especially MS/MS fragmentation analyses, are required to confirm this structural assignment, from biosynthetic and mass considerations, this peak is likely to reflect the presence of disialylated O-linked glycans on NK cells. In the case of human T cells, the N-linked glycan pool contained a peak at m/z 3965.7, which also corresponds to an {alpha}2,8-disialylated motif (data not shown). Equivalent peaks were not seen with B cells or monocytes (data not shown). Thus, the strong reactivity of NK cells and T cells with the Ha1 and Pi1 antibodies is mirrored by the presence of putative {alpha}2,8-linked disalic acids detected by MALDI-TOF MS.


Figure 5
View larger version (11K):
[in this window]
[in a new window]
  		Figure 5. MALDI-TOF MS profile of O-linked glycans from purified NK cells. Major peaks are annotated with the corresponding glycan structure in symbol form, following the glycan nomenclature established by the CFG. Complete annotation of the spectra can be found at the CFG website.

 
DISCUSSION

Siglec-7 is masked at the surface of human NK cells as a result of its engagement in cis with sialic acid residues. Given the strong preference of Siglec-7 for {alpha}2,8-linked disialic acids, these are likely candidates for masking of Siglec-7. Consistent with expression of {alpha}2,8-linked disialic acids on NK cells, we show that they consistently express mRNA for sialyltransferase ST8Sia VI, which transfers one sialic acid onto sialylated O-glycan acceptors. Variable expression of ST8Sia IV and V was also seen, enzymes that transfer {alpha}2,8-linked sialic acids to glycoprotein and glycolipid acceptors, respectively. Thus, at the level of mRNA and likely, protein expression, NK cells are well-equipped to generate {alpha}2,8-linked disialic acids at the cell surface. Consistent with this, we showed that the disialo-specific antibodies Ha1 and Pi1 reacted strongly with NK cells and less strongly with T cells and monocytes, both of which express lower levels of Siglec-7. It is interesting that B cells, which do not express Siglec-7, were not recognized by these antibodies. Furthermore, Siglec-7 and Ha1 were found to partially co-cap, suggesting that Siglec-7 was associated with {alpha}2,8-linked disialic acid-containing ligands at the cell surface. Identification of counter-receptors, which are decorated with {alpha}2,8-linked disialic acids and interact in cis with Siglec-7, will be important future goals. Using co-capping experiments, we have so far been unable to see colocalization of Siglec-7 with candidate counter-receptors such as NCAM (CD56), CD16, or CD2 (data not shown). Besides glycoprotein counter-receptors, it is possible that {alpha}2,8-disialogangliosides are important for cis interactions with Siglec-7 on leukocytes. Although we did not detect expression of GD3 on NK cells using the R24 mAb, other candidates such as GD2, GT1b, GQ1b, GT1a, and GT3 may be expressed. Currently, little is known about the ganglioside composition of human NK cells, although it is of interest that in the current study, we showed that NK cells variably expressed the ST8Sia V gene, whose product can lead to synthesis of GT1a, GQ1b, and GT3. In addition, Ha1, which bound strongly to NK cells, recognized GT3 oligosaccharide in the glycan array. Further studies aimed at structural characterization of glycoproteins and glycolipids modified with {alpha}2,8-linked disialic acids are likely to shed light on the nature of putative Siglec-7 counter-receptors expressed on NK cells and other leukocyte populations.

ACKNOWLEDGEMENTS

This work was supported by a Wellcome Trust Senior Fellowship (GR047677MA) awarded to P. R. C. The resources and collaborative efforts provided by the CFG were funded by NIGMS-GM62116. We thank Simon Clark and Joanna Warren for help with the real-time PCR experiments, Eric Wagner and Jean Veitch for antibody purification, and Rick Alvarez from the CFG for performing glycan array analyses. We are grateful to Dr. Reinhard Schwartz-Albiez for allowing us to include CFG data on the R24 antibody. We thank Drs. Chihiro Sato and Ken Kitajima for valuable discussions.

FOOTNOTES

1 Current address: UPRES EA3889, Université de Rennes 1, Centre Eugène Marquis, Rennes, France. Back

Received October 4, 2005; revised June 19, 2006; accepted June 20, 2006.

REFERENCES

  1. Varki, A., Angata, T. (2006) Siglecs—the major subfamily of I-type lectins Glycobiology 16,1R-27R[Abstract/Free Full Text]
  2. Crocker, P. R. (2005) Siglecs in innate immunity Curr. Opin. Pharmacol. 5,431-437[CrossRef][Medline]
  3. Angata, T., Margulies, E. H., Green, E. D., Varki, A. (2004) Large-scale sequencing of the CD33-related Siglec gene cluster in five mammalian species reveals rapid evolution by multiple mechanisms Proc. Natl. Acad. Sci. USA 101,13251-13256[Abstract/Free Full Text]
  4. Kumamoto, Y., Higashi, N., Denda-Nagai, K., Tsuiji, M., Sato, K., Crocker, P. R., Irimura, T. (2004) Identification of sialoadhesin as a dominant lymph node counter-receptor for mouse macrophage galactose-type C-type lectin 1 J. Biol. Chem. 279,49274-49280[Abstract/Free Full Text]
  5. Nath, D., Hartnell, A., Happerfield, L., Miles, D. W., Burchell, J., Taylor-Papadimitriou, J., Crocker, P. R. (1999) Macrophage-tumor cell interactions: identification of MUC1 on breast cancer cells as a potential counter-receptor for the macrophage-restricted receptor, sialoadhesin Immunology 98,213-219[CrossRef][Medline]
  6. Van den Berg, T. K., Nath, D., Ziltener, H. J., Vestweber, D., Fukuda, M., van Die, I., Crocker, P. R. (2001) Cutting edge: CD43 functions as a T cell counterreceptor for the macrophage adhesion receptor sialoadhesin (Siglec-1) J. Immunol. 166,3637-3640[Abstract/Free Full Text]
  7. Zhang, M., Varki, A. (2004) Cell surface sialic acids do not affect primary CD22 interactions with CD45 and surface IgM nor the rate of constitutive CD22 endocytosis Glycobiology 14,939-949[Abstract/Free Full Text]
  8. Vinson, M., Strijbos, P. J., Rowles, A., Facci, L., Moore, S. E., Simmons, D. L., Walsh, F. S. (2001) Myelin-associated glycoprotein interacts with ganglioside GT1b. A mechanism for neurite outgrowth inhibition J. Biol. Chem. 276,20280-20285[Abstract/Free Full Text]
  9. Domeniconi, M., Cao, Z., Spencer, T., Sivasankaran, R., Wang, K., Nikulina, E., Kimura, N., Cai, H., Deng, K., Gao, Y., He, Z., Filbin, M. (2002) Myelin-associated glycoprotein interacts with the Nogo66 receptor to inhibit neurite outgrowth Neuron 35,283-290[CrossRef][Medline]
  10. Strenge, K., Brossmer, R., Ihrig, P., Schauer, R., Kelm, S. (2001) Fibronectin is a binding partner for the myelin-associated glycoprotein (siglec-4a) FEBS Lett. 499,262-267[CrossRef][Medline]
  11. Nicoll, G., Avril, T., Lock, K., Furukawa, K., Bovin, N., Crocker, P. R. (2003) Ganglioside GD3 expression on target cells can modulate NK cell cytotoxicity via siglec-7-dependent and -independent mechanisms Eur. J. Immunol. 33,1642-1648[CrossRef][Medline]
  12. Avril, T., Floyd, H., Lopez, F., Vivier, E., Crocker, P. R. (2004) The membrane-proximal immunoreceptor tyrosine-based inhibitory motif is critical for the inhibitory signaling mediated by Siglecs-7 and -9, CD33-related Siglecs expressed on human monocytes and NK cells J. Immunol. 173,6841-6849[Abstract/Free Full Text]
  13. Avril, T., Freeman, S. D., Attrill, H., Clarke, R. G., Crocker, P. R. (2005) Siglec-5 (CD170) can mediate inhibitory signaling in the absence of immunoreceptor tyrosine-based inhibitory motif phosphorylation J. Biol. Chem. 280,19843-19851[Abstract/Free Full Text]
  14. Razi, N., Varki, A. (1999) Cryptic sialic acid binding lectins on human blood leukocytes can be unmasked by sialidase treatment or cellular activation Glycobiology 9,1225-1234[Abstract/Free Full Text]
  15. Collins, B. E., Smith, B. A., Bengtson, P., Paulson, J. C. (2006) Ablation of CD22 in ligand-deficient mice restores B cell receptor signaling Nat. Immunol. 7,199-206[CrossRef][Medline]
  16. Sonnenburg, J. L., Altheide, T. K., Varki, A. (2004) A uniquely human consequence of domain-specific functional adaptation in a sialic acid-binding receptor Glycobiology 14,339-346[Abstract/Free Full Text]
  17. Falco, M., Biassoni, R., Bottino, C., Vitale, M., Sivori, S., Augugliaro, R., Moretta, L., Moretta, A. (1999) Identification and molecular cloning of p75/AIRM1, a novel member of the sialoadhesin family that functions as an inhibitory receptor in human natural killer cells J. Exp. Med. 190,793-802[Abstract/Free Full Text]
  18. Nicoll, G., Ni, J., Liu, D., Klenerman, P., Munday, J., Dubock, S., Mattei, M. G., Crocker, P. R. (1999) Identification and characterization of a novel siglec, siglec-7, expressed by human natural killer cells and monocytes J. Biol. Chem. 274,34089-34095[Abstract/Free Full Text]
  19. Yamaji, T., Teranishi, T., Alphey, M. S., Crocker, P. R., Hashimoto, Y. (2002) A small region of the natural killer cell receptor, Siglec-7, is responsible for its preferred binding to {alpha} 2,8-disialyl and branched {alpha} 2,6-sialyl residues. A comparison with Siglec-9 J. Biol. Chem. 277,6324-6332[Abstract/Free Full Text]
  20. Blixt, O., Collins, B. E., van den Nieuwenhof, I. M., Crocker, P. R., Paulson, J. C. (2003) Sialoside specificity of the siglec family assessed using novel multivalent probes: identification of potent inhibitors of myelin-associated glycoprotein J. Biol. Chem. 278,31007-31019[Abstract/Free Full Text]
  21. Rapoport, E., Mikhalyov, I., Zhang, J., Crocker, P., Bovin, N. (2003) Ganglioside binding pattern of CD33-related siglecs Bioorg. Med. Chem. Lett. 13,675-678[CrossRef][Medline]
  22. Yasukawa, Z., Sato, C., Kitajima, K. (2005) Inflammation-dependent changes in {alpha}2,3-, {alpha}2,6-, and {alpha}2,8-sialic acid glycotopes on serum glycoproteins in mice Glycobiology 15,827-837[Abstract/Free Full Text]
  23. Sato, C., Fukuoka, H., Ohta, K., Matsuda, T., Koshino, R., Kobayashi, K., Troy, F. A., II, Kitajima, K. (2000) Frequent occurrence of pre-existing {alpha} 2->8-linked disialic and oligosialic acids with chain lengths up to 7 Sia residues in mammalian brain glycoproteins. Prevalence revealed by highly sensitive chemical methods and anti-di-, oligo-, and poly-Sia antibodies specific for defined chain lengths J. Biol. Chem. 275,15422-15431[Abstract/Free Full Text]
  24. Harduin-Lepers, A., Mollicone, R., Delannoy, P., Oriol, R. (2005) The animal sialyltransferases and sialyltransferase-related genes: a phylogenetic approach Glycobiology 15,805-817[Abstract/Free Full Text]
  25. Weinhold, B., Seidenfaden, R., Rockle, I., Muhlenhoff, M., Schertzinger, F., Conzelmann, S., Marth, J. D., Gerardy-Schahn, R., Hildebrandt, H. (2005) Genetic ablation of polysialic acid causes severe neurodevelopmental defects rescued by deletion of the neural cell adhesion molecule J. Biol. Chem. 280,42971-42977[Abstract/Free Full Text]
  26. Lanier, L. L., Testi, R., Bindl, J., Phillips, J. H. (1989) Identity of Leu-19 (CD56) leukocyte differentiation antigen and neural cell adhesion molecule J. Exp. Med. 169,2233-2238[Abstract/Free Full Text]
  27. Jones, C., Virji, M., Crocker, P. R. (2003) Recognition of sialylated meningococcal lipopolysaccharide by siglecs expressed on myeloid cells leads to enhanced bacterial uptake Mol. Microbiol. 49,1213-1225[CrossRef][Medline]
  28. Herron, B., Willison, H. J., Veitch, J., Roelcke, D., Illis, L. S., Boulton, F. E. (1994) Monoclonal IgM cold agglutinins with anti-Pr1d specificity in a patient with peripheral neuropathy Vox Sang. 67,58-63[Medline]
  29. Willison, H. J., O’Hanlon, G. M., Paterson, G., Veitch, J., Wilson, G., Roberts, M., Tang, T., Vincent, A. (1996) A somatically mutated human antiganglioside IgM antibody that induces experimental neuropathy in mice is encoded by the variable region heavy chain gene, V1–18 J. Clin. Invest. 97,1155-1164[Medline]
  30. Bochner, B. S., Alvarez, R. A., Mehta, P., Bovin, N. V., Blixt, O., White, J. R., Schnaar, R. L. (2005) Glycan array screening reveals a candidate ligand for siglec-8 J. Biol. Chem. 280,4307-4312[Abstract/Free Full Text]
  31. Tateno, H., Crocker, P. R., Paulson, J. C. (2005) Mouse Siglec-F and human Siglec-8 are functionally convergent paralogs that are selectively expressed on eosinophils and recognize 6'-sulfo-sialyl Lewis X as a preferred glycan ligand Glycobiology 15,1125-1135[Abstract/Free Full Text]
  32. Avril, T., Wagner, E. R., Willison, H. J., Crocker, P. R. (2006) Sialic acid-binding immunoglobulin-like lectin 7 mediates selective recognition of sialylated glycans expressed on Campylobacter jejuni lipooligosaccharides Infect. Immun. 74,4133-4141[Abstract/Free Full Text]
  33. Kojima, N., Yoshida, Y., Tsuji, S. (1995) A developmentally regulated member of the sialyltransferase family (ST8Sia II, STX) is a polysialic acid synthase FEBS Lett. 373,119-122[CrossRef][Medline]
  34. Yoshida, Y., Kojima, N., Kurosawa, N., Hamamoto, T., Tsuji, S. (1995) Molecular cloning of Sia {alpha} 2,3Gal ß 1,4GlcNAc {alpha} 2,8-sialyltransferase from mouse brain J. Biol. Chem. 270,14628-14633[Abstract/Free Full Text]
  35. Fox, D. A., He, X., Abe, A., Hollander, T., Li, L. L., Kan, L., Friedman, A. W., Shimizu, Y., Shayman, J. A., Kozarsky, K. (2001) The T lymphocyte structure CD60 contains a sialylated carbohydrate epitope that is expressed on both gangliosides and glycoproteins Immunol. Invest. 30,67-85[CrossRef][Medline]
  36. Takashima, S., Ishida, H. K., Inazu, T., Ando, T., Ishida, H., Kiso, M., Tsuji, S., Tsujimoto, M. (2002) Molecular cloning and expression of a sixth type of {alpha} 2,8-sialyltransferase (ST8Sia VI) that sialylates O-glycans J. Biol. Chem. 277,24030-24038[Abstract/Free Full Text]
  37. Teintenier-Lelievre, M., Julien, S., Juliant, S., Guerardel, Y., Duonor-Cerutti, M., Delannoy, P., Harduin-Lepers, A. (2005) Molecular cloning and expression of a human hST8Sia VI ({alpha}2,8-sialyltransferase) responsible for the synthesis of the diSia motif on O-glycosylproteins Biochem. J. 392,665-674[CrossRef][Medline]
  38. Angata, K., Fukuda, M. (2003) Polysialyltransferases: major players in polysialic acid synthesis on the neural cell adhesion molecule Biochimie 85,195-206[Medline]
  39. Kitazume-Kawaguchi, S., Kabata, S., Arita, M. (2001) Differential biosynthesis of polysialic or disialic acid structure by ST8Sia II and ST8Sia IV J. Biol. Chem. 276,15696-15703[Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 J. Biol. Chem.Home page
P.-C. Pang, B. Tissot, E. Z. Drobnis, P. Sutovsky, H. R. Morris, G. F. Clark, and A. Dell
Expression of Bisecting Type and Lewisx/Lewisy Terminated N-Glycans on Human Sperm
J. Biol. Chem., December 14, 2007; 282(50): 36593 - 36602.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
jlb.1005559v1
80/4/787    most recent
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
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 reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Avril, T.
Right arrow Articles by Crocker, P. R.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Avril, T.
Right arrow Articles by Crocker, P. R.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS