Involvement of the Gal(beta)1-3GalNAc(beta) structure in the recognition of apoptotic bodies by THP-1 cells

Apoptosis – galactose-specific lectin – macrophages neoglycoconjugates – phagocytosis

A specific apoptotic glycosylation pattern may play an assistant or even a causative role in phagocytosis of apoptotic bodies. To elucidate the role of macrophages in lectin-mediated phagocytosis, an experimental system was used, where monocyte-derived THP-1 cells engulf the apoptotic bodies from the melanoma cell line MELJUSO. A flow cytometry assay was performed to reveal lectin expression and quantify the phagocytosis of apoptotic bodies. Taking into account that siglecs, a mannose receptor and galectins expressed on macrophages could be involved in engulfment of apoptotic bodies we studied their potential expression on THP-1 cells by means of polyacrylamide glycoconjugates. A strong binding of the cells to siglec ligands (3′SiaLac, 6′SiaLac, [Neu5Ac^agr;2-8]^sub 2^) and galectin ligands (LacNAc, GalNAc[beta]1-4GlcNAc, Gal[beta]1SGalNAc[beta] and asialoGM1) was observed. To reveal the corresponding targets on apoptotic bodies, the carbohydrate pattern of MELJUSO cells was analyzed. The apoptotic membrane was characterized by a high level of glycans terminated by galactose or sialic acid. To study lectin-mediated phagocytosis of apoptotic bodies by THP-1 cells, an inhibitory phagocytosis assay was performed. Binding of Gal[beta]1-3GalNAc- or LacNAc-specific reagents (lectins and antibodies) to apoptotic bodies abolished their engulfment by the THP-1 cells whereas blocking of Neu5Ac[alpha]2 – 6 or Neu5Ac[alpha]2 – 3 sites by the corresponding lectins was not effective. Furthermore, Gal[beta]1 3GalNAc[beta]- PAA or asialoGM1-PAA binding to the THP-1 cells decreased phagocytosis, whereas two other potent THP-1binding probes, LacNAc- PAA and GalNAc[beta]1 – 4GlcNAcPAA did not inhibit phagocytosis. Thus, Gal[beta]1 -3GalNAc[beta] terminated chains represented on the apoptotic bodies but not the other tested galectin ligands appear to be a target for THP-1 cells.

Abbreviations. FITC Fluorescein isothiocyanate. – flu Fluorescein residue. MAbs Monoclonal antibodies. – (Man)^sub 2^Man Man[alpha]1 – 3(Man[alpha]1 -6) Man[alpha]. – PBA PBS containing 0.2% bovine serum albumin. – PBS Phosphate-buffered saline. – PE Phycoerythrin. – PMA Phorbol 12myristate 13-acetate. – Sug-PAA Polyacrylamide glycoconjugates. – T^sub p[beta][beta]^ Gal[beta]1 -3GalNAc[beta]. – TF Gal[beta]1 -3GalNAc[alpha]. – Versene solution PBS containing 0.02% EDTA.

Introduction

Apoptosis is an active mode of cell death developing throughout life and participating in cell turnover of many tissues. It pathologically occurs in cancers, autoimmune, infectious or neurodegenerative diseases. Apoptosis is characterized by cell shrinkage, nuclear condensation, protease activation and, finally, by DNA fragmentation (Granville et al., 1998). Cells undergoing apoptosis are recognized and rapidly eliminated by neighboring phagocytes. Removal of apoptotic cells is crucial for many biological processes including normal tissue turnover, the development of the immune system and resolution of inflammation. The engulfment of apoptotic bodies by phagocytes requires specific recognition systems. Participation in this process of integrin molecules including the vitronectin receptor and the thrombospondin receptor has been demonstrated (Rigotti et al,1995; Fadok et al., 1992). Phosphatidylserine and scavenger receptors, including CD36 (Fraser et al., 1993; Bratton et al., 2000) of macrophages were shown to bind to phosphatidylserine of apoptotic bodies that appears on the cell surface at early apoptosis. Alterations of glycosylation represent another class of changes that occur on apoptotic cell surfaces. An assistant or even causative participation can be traced for carbohydrate chains in phagocytosis of apoptotic bodies by macrophages. The mannose/fucose-specific lectin was shown to participate in the phagocytosis of apoptotic neutrophilsby fibroblasts (Hall et al., 1994). Phagocytosis of rat apoptotic hepatocytes by native hepatocytes is inhibited by antibodies against the asialoglycoprotein receptor (Dini et al., 1992). Not long ago we investigated glycosylation alterations of colon carcinoma cells in late-phase apoptosis and observed that induction of apoptosis led to actual loss of most fucose- and partial loss of sialic acid- containing structures accompanied by an increased exposure of [beta]- galactose residues (Rapoport and LePendu, 1999). Such results suggested that [beta]-galactoside-specific lectins (possibly galectins) could be involved in recognition events of phagocytosis. Galectins are revealed in many tumors and their expression correlates with the cell metastatic potential (Leffler, 2001). Galectins secreted by tumor cells might favour tumor cell aggregation and adhesion to the cell matrix and promote tumor cell metastases (Perillo et al, 1998; Leffler, 2001). Galectins are implicated in apoptosis and cell proliferation and may also be involved in the phagocytosis of apoptotic bodies.

This investigation is aimed at studying lectin involvement in the recognition of tumor apoptotic bodies by macrophages. Macrophages express galectins, mannose receptor (Ezekowitz et al., 1990) and siglecs (Munday et al., 1999). One cannot exclude that selecting as adhesive molecules may also affect macrophage activation (Langermans et al, 1994). Using a panel of fluorescein-labeled glycoconjugates, the possible expression of selectins, the mannose receptor, siglecs, and galectins on the monocyte-derived THP-1 cell line was studied. In parallel we analyzed the glycosylation pattern of MELJUSO melanoma cells after induction of apoptosis thus determining potential carbohydrate ligands of the mentioned lectins. Our results showed that receptors for sulfated oligosaccharides, for sialylated oligosaccharides and for galactosides were present on phagocytic cells. Yet, only galactoside receptors participated in engulfment. Moreover, only Gal[beta]1 -SGalNAc[beta]- but not lactosamine- related chains of the apoptotic tumor cells seemed to be a trigger in the phagocytosis process.

Materials and methods

Cell culture

Human monocytic leukaemia cell line THP-1 (ECACCN 88081201) was kindly provided by Dr. V. A. Nesmyanov (Moscow, Russia), melanoma cell line MELJUSO is a gift of Dr. C. Watt (Dundee, UK). THP-1 was cultured in RPMI1640 (Flow laboratories, UK) supplemented with 10% FCS, 2 mM glutamine (Flow Laboratories, UK) and 0.5 mM 2- mercaptoethanol (Serva, Germany). MELJUSO was cultured in Dulbecco medium (Flow Laboratories) under the same conditions. For induction of apoptosis, MELJUSO cells at about 80% confluency were UV irradiated for 5 min and cultivated in complete medium for 24 h. After changing the medium the cells were cultivated again for 24 h, detached cells were then collected. Under these conditions, >80% detached cells still excluded trypan blue, indicating that they had not yet reached the stage of post-apoptotic necrosis. Induction of apoptosis was confirmed by analysis of DNA fragmentation as described in (Rapoport and LePendu, 1999).

Tab. I. Epitope specificity of MAbs related to T^sub [beta][beta]^ and TF-containing structures.

Lectins, glycoconjugates and antibodies

Digoxigenin-labeled lectins from Maakia amurensis (MAA), Sambucus nigra (SNA) and Arachis hypogaea (PNA) were from Roche (Mannheim, Germany), biotin-labeled lectin Erythrina cristagali (ECA) was from Vecta (USA). MAbs A78-G/A7, A68-B/A11, A68-E/A2, and A84-A/F10 were generated as described before (Karsten et al., 1995; Cao et al., 1995). Their specificity, as determined by Sug-PAA-based ELISA (Vlasova et al., 1994), is shown in Table I. FITC-labeled anti- digoxigenin, streptavidin and goat anti-mouse IgG were purchased from Roche and Sigma, respectively. PE-labeled antibodies T29/33 to human CD45 were purchased from Dako (Denmark). Sug-PAA and Sug-PAA- flu were from Syntesome (Munich, Germany). AsialoGM1 was synthesized as described in (Baidina et al., 2000).

Assay of Sug-PAA-flu binding to THP-1 cells

To obtain macrophage-like cells, THP-1 cells were cultured in medium containing PMA for 24 h. After medium removal the cells were cultured in the same medium without PMA for 24 h. Adherent cells were harvested with cold Versene solution. Cells were then washed three times in cold PBA and incubated with Sug-PAA-flu (100 [mu]g/ ml) for 40 min at 4 [degrees]C. After washing with cold PBA, fluorescence analysis was performed using a flow cytometer (Dako Galaxy, Denmark, or EPICS ELITE Coulter, USA).

Binding of lectins to apoptotic bodies

The cells in suspension obtained after induction of apoptosis were washed three times in PBA and incubated with lectins at the appropriate concentrations in the same buffer for 30 min at 4 [degrees]C. The cells were then incubated with FITC-labeled anti- digoxigenin or streptavidin. After final washing with PBA, fluorescence analysis was performed as described above. Cells without UV irradiation were used as control after being suspended using Versene solution.

FITC labeling of apoptotic bodies

Apoptotic bodies obtained from the MELJUSO cell line as described above were washed 3 times with cold Dulbecco medium by centrifugation at 1200 rpm/min for 3 min. The cells (10(6) cells/ ml) were suspended in 1 ml medium containing 300 ngFITC (Sigma, USA) and incubated at 37 [degrees]C for 40 min followed by washing with medium. FITC-labeled cells were kept frozen in medium containing 10% DMSO until use.

Phagocytosis assay

To avoid adherence of THP-1 cells to wells, 24-well plates (Nunc, Denmark) were rinsed with 2% poly(2-hydroxyethyl methacrylate) in ethanol. Activated THP-1 cells were harvested with Versene solution, washed three times with medium, and 10(6) cells/ml were cultured overnight with apoptotic bodies. To confirm that phagocytosis occurred, cell suspension was smeared on coverslips and stained with Giemsa. A minimum of 300 cells was examined for phagocytosis.

For the flow cytometry assay, THP-1 cells were cultured with FITC- labeled apoptotic bodies in pretreated 24-well plates. The reaction mixture was collected, washed three times in cold PBA and incubated with anti-human CD45-PE for 60 min al 4 [degrees]C. After the final washing FACS analysis was performed.

Fig. 1. Binding of Sug-PAA-flu probes to THP-1 cells, cytofluorometric analysis. THP-1 cells were cultured in medium containing PMA for 24 h. Then the suspension of non-differentiated cells was removed and adherent cells were cultured without PMA as described in “Materials and methods”. Only adherent differentiated cells were subjected to analysis. Fluorescence increase was calculated as [(F^sub i^/F^sub 0^) x 100]-100, where F^sub i^ – fluorescence intensity of cells after incubation with the probe, F^sub 0^ – fluorescence intensity of untreated cells. (A) Interaction of THP-1 cells with probes related to selectins and mannose receptor ligands. Increase of fluorescence after incubation with 6HSO^sub 3^SiaLe^sup x^, SiaLe^sup a^, Man[alpha] probes was less than 10% and therefore not shown. (B) Interaction of THP-1 cells with probes related to siglec ligands; increase of fluorescence after incubation with Neu5Ac[beta]Bn and Neu5Ac[alpha]O(CH^sub 2^)^sub 3^, probes was less than 10%. (C) Interaction of THP-1 cells with probes related to galectin ligands. Increase of fluorescence after incubation with GlcNAc[beta]1- 4GlcNAc, Ga1[alpha]1-3GalNAc[alpha]1, Gal[alpha]13GalNAc[beta], Gal[beta]1 -3GalNAc[alpha] and GalNAc[alpha]1-3GalNAc[beta] probes was less than 10%. Results shown correspond to mean values + S.D. from triplicates of one representative experiment out of four.

In inhibition assays THP-1 cells were pretreated with Sug-PAA (100 [mu]g/ml) or free sugars (100 mM or 200 mM). Alternatively, apoptotic bodies were incubated with lectins (1 [mu]g/ml) or MAbs at appropriate concentrations for 40 min at 37 [degrees]C. Cells were washed with RPMI-1640 and THP-1 cells were then cultured with apoptotic bodies before analysis by flow cytometry as described above.

Results

Binding of Sug-PAA-flu to THP-1 cells after PMA activation

The presence of galectins, mannose/GalNAc receptor, siglecs or selectins on THP-1 cells was revealed by flow cytometry using Sug- PAA-flu probes that correspond to their respective ligands (Galanina et al., 1998). Figure 1A represents binding to THP-1 cells of selectin ligands and related saccharidcs as Sug-PAA-flu probes. THP- 1 cells were weakly labeled by 6′HSO^sub 3^. SiaLe^sup x^ (6- sulfated Gal) whereas SiaLe^sup a^ and 6HSO^sub 3^SiaLe^sup x^ (6- sulfated GlcNAc) were inactive (not shown). We also investigated the interaction of THP-1 cells with three mannose-containing glycoconjugates Man-PAA, Man2Man-PAA and (GlcNAcMan)^sub 2^ManGlcNAc^sub 2^. Only the last probe (Fig. 1A) displayed significant (increase of fluorescence above 10%) binding to THP-1 cells. Figure IB displays staining of THP-1 cells by typical siglec ligands. THP-1 cells were labeled by 3′SiaLac, 6′SiaLac and [Neu5Ac[alpha]2-8]^sub 2^ probes but not by Neu5Ac[alpha]-PAA (18- 25% increase of fluorescence). Figure 1C represents probing of THP- 1 cells with galectin ligands: Gal[beta]1 – 4GlcNAc (LacNAc), GalNAc[beta]1-4GlcNAc (Lac-di-NAc) and Gal[beta]1 -SGalNAc[beta] (T^sub [beta][beta]^) probes demonstrated binding to the cells. At the same time THP-1 did not recognize related disaccharides Gal[beta]1-SGalNAc[alpha] (TF) and GalNAc[alpha]1-3GalNAc[beta] (Fs). These results indicated that the THP-1 cells display carbohydrate-binding sites related to those of selectins, siglecs, the mannose receptor, and galectins (see Discussion).

Induction of apoptosis

MELJUSO cells were irradiated with UV. In order to confirm that this treatment induced apoptosis, DNA analysis was performed and DNA fragmentation was monitored. As shown in Figure 2A, a DNA ladder typical of apoptosis was observed in extract from UV-treated cells (lane 2) but not from control cells (lane 1).

Probing the carbohydrate chains of apoptotic bodies

Terminal residues of carbohydrate chains on MELJUSO melanoma cells were studied with four lectins. Binding of MAA (recognizing Neu5Ac[alpha]2-3Gal) to the UV-irradiated cells did not differ much from the binding to untreated cells (Fig. 2B). A weak increase in Gal- and sialic acid-terminated chain expression was observed as probed with ECA and SNA (directed to Gal[beta]1 -4GalNAc and Neu5Ac[alpha]2-6Gal, respectively). Binding of PNA (specific to Gal[beta]1-3GalNAc) to apoptotic bodies slightly decreased.

Inhibition of phagocytosis

The phagocytosis of apoptotic bodies was visually evaluated by Giemsa staining. A THP-1 cell ready to engulf an apoptotic body (Fig. 3, upper panel) and an apoptotic body localized inside the cytoplasm of a THP-1 cell (Fig. 3, lower panel) are shown. Further, to measure the population of apoptotic bodies that was ingested by THP-1 cells, a flow cytometry assay was performed as described in “Materials and methods”. A MAb to CD45 (leucocyte-associated antigen) was used to detect monocyte-derived THP-1 cells (Fig. 4). Figure 4A-C shows the population of CD45-positive THP-1 cells that engulf apoptotic bodies. As shown in Figure 4B, the engulfment of apoptotic bodies by activated THP-1 cells increased almost 3 times compared to that of non-activated cells. Addition of EDTA decreased the engulfment ability of THP-1 cells (Fig. 4C).

Fig. 2. A. DNA fragmentation in MELJUSO cells after UV irradiation. DNA extraction from treated cells in suspension or adherent untreated cells followed by electrophoresis on a 1.8% agarose gel was performed according to (Rapoport and Le Pendu, 1999). DNA fragments were visualized under UV light by ethidium bromide staining. Lane 1: DNA from adherent untreated cells; lane 2: DNA from cells in suspension collected 48 h after UV irradiation. B. Lectin cytofluorometric analysis of MELJUSO cell glycosylation before (white bar) and after (grey bar) induction of apoptosis by UV irradiation. Only cells in suspension were analyzed after apoptosis induction, untreated MELJUSO cells were detached with Versene solution. Intensity of fluorescence is given in arbitrary units. One hundred units correspond to autofluorescence, without lectin incubation. Results shown correspond to mean values + S.D. from triplicates of one representative experiment out of four.

Fig. 3. Phagocytosis of apoptotic bodies by THP-1 cells. Coverslip smears from the cultured cell suspension were stained with Giemsa (magnification x 1000). Upper panel: apoptotic body present near a THP-1 cell; lower panel: apoptotic body inside THP-1 cytoplasm. The morphological characteristics of THP-1 cells have been described earlier (Tsuchiya et al., 1980).

We next investigated the phagocytosis of apoptotic bodies from MELJUSO by activated THP-1 cells in the presence of two types of inhibitors: plant lectins and glycoconjugates. As shown in Figure 5A, PNA and ECA that recognize terminal [beta]-galactosides, led to about a 50% decrease of phagocytosis, whereas SNA or MAA gave less than 5% inhibition. Synthetic glycoconjugate probes were chosen for inhibition experiments based on their binding ability to THP-1 cells described above. Sialic acid-containing probes 3′SiaLac-PAA, [Neu5Ac[alpha]2-8]^sub 2^PAA or 6′SiaLac-PAA did not affect the engulfment (Fig. 5B), SiaLe(x)-PAA and 3′HSO^sub 3^Le(x)-PAA were also inactive. At the same time pretreatment of THP-1 cells with Gal[beta]1-3GalNAc[beta]-PAA or asialoGM1-PAA resulted in a strong (more than 50%) inhibition of phagocytosis, whereas two other, LacNAcPAA and GalNAc[beta]1 -4GlcNAc[beta]-PAA (Fig. 5B), as well as free LacNAc, lactose and galactose at 0.2 M were inactive (data not shown).

Fig. 4. Phagocytosis of melanoma MELJUSO apoptotic bodies by THP- 1 cells. The two-parametric histogram displays light scatter of logarithmic FITC (X axis) and PE (Y axis) fluorescence intensities. Quadrants represent THP-1 cells stained only with anti-CD45-PE (left top), FITC-labeled apoptotic bodies (right bottom) and anti-CD45-PE- labeled THP-1 cells that engulf apoptotic bodies (right top). Values in the quadrants represent the proportion of engulfment-positive THP- 1 cells. Dot plots were generated after gating anti-CD45-PE- positive THP-1 cells. (A) Phagocytosis of apoptotic bodies by non- activated THP-1 cells; (B) phagocytosis of apoptotic bodies by activated THP-1 cells; (C) phagocytosis by activated THP-1 cells in the presence of 1 mM EDTA; (D) apoptotic bodies only.

These results strongly suggest that Gal[beta]1 -3GalNAc[beta] is crucial for recognition of apoptotic bodies. We therefore studied the effect of MAbs related to this epitope on the engulfment ability of THP-1 cells. In the phagocytosis inhibition assay, MAbs A68-B/ A11 and A68-E/A2, specific for the [beta]-form of TF-antigen (Gal[beta]1-3GalNAc[beta]), especially the first one, appeared to be potent inhibitors, whereas the anti-TF (Gal[beta]1 – 3GalNAc[alpha]) MAbs A84-A/F10 and A78-G/A7 were not inhibitory (Fig. 6).

Discussion

Apoptosis is one of the mechanisms responsible for the elimination of tumor cells at early stages of tumor transformation. The surviving tumor cells undergo several changes, resulting in increased resistance to the action ofmacrophages and killer cells (Fidler and Kleinerman, 1993; Deichman, 2000). Investigation of the phagocytosis mechanisms, in particular, the receptors of the primary recognition is key to the development of ways of phagocytosis stimulation. Lectins can participate in phagocytosis by binding to complementary carbohydrates of target cells (lectin-dependent phagocytosis). In our earlier studies we observed a high expression of terminal residues such as galactose and sialic acid on apoptotic tumor cells as well as a near complete loss of fucose-containing chains (Rapoport and LePendu, 1999). Data available indicate a possible involvement of galactose- and sialospecific lectins in engulfment of apoptotic bodies. Siglecs (Munday et al., 1999; Nicoll et al., 1999; Floyd et al., 2000), an asialoglycoprotein receptor (Sakamaki et al., 1995) and galectins (Reichert et al., 1994; Perillo et al., 1998) are known to be expressed on macrophages. One may speculate that these lectins and complementary carbohydrate chains of glycoproteins and glycolipids of apoptotic bodies participate in phagocytosis. Besides, the involvement of macrophage mannose receptor in tumor cells engulfment has been shown previously (Stahl and Ezekowitz, 1998). In the present study engulfment of apoptotic bodies derived from UV-irradiated MELJUSO melanoma cells by THP-1 cells was used as a phagocytosis model. The activated THP- 1 cells display macrophage characters (Tsuchiya etal., 1980; Scubitz et al., 1983). Cytofluorometric analysis is a convenient and reliable method to monitor engulfed cell populations (Langermans et al., 1994). We used it in order to quantify phagocytosis. To confirm that uptake of apoptotic bodies occurs, a phagocytosis assay in the presence of EDTA was performed. A thrombospondin receptor (Fadok et al., 1992) activated by Ca^sup 2+^ ions participates in the phagocytosis of apoptotic bodies by THP-1 cells. Chelation of Ca^sup 2+^ ions by EDTA proved to reduce the phagocyte activity of the THP- 1 cells (Fig. 4C).

Fig. 5. Phagocytosis of apoptotic bodies by THP-1 cells in the presence of lectins (A) and PAA-glycoconjugates (B). Inhibition of phagocytosis was calculated as 100-[(Ph^sub i^/Ph^sub 0^) x 100], where Ph^sub i^ – phagocytosis degree in the presence of the inhibitor, Ph^sub 0^ – phagocytosis degree without inhibitor. Phagocytosis degree is given as percentage of anti-CD45-PE-positive cells. Results shown correspond to mean values + S.D. from triplicates of one representative experiment out of four.

Since THP-1 is a macrophage cell line, siglecs, selectins, the mannose receptor, and galectins may be expressed on THP-1 cells and participate in the phagocytosis of apoptotic bodies. Polyacrylamide carbohydrate probes were used to reveal the [beta]-galacto-, sialo- , manno-, and some other carbohydrate-binding activities.

Fig. 6. Phagocytosis of apoptotic bodies by THP-1 cells in the presence of MAbs against the Gal[beta]1 – 3GaINAc epitope. Inhibition of phagocytosis was calculated as 100-[(Ph^sub i^Ph^sub 0^) x 100], where Ph^sub i^ – phagocytosis degree in the presence of the inhibitor, Ph^sub 0^ – phagocytosis degree without inhibitor. Phagocytosis degree is given as percentage of anti-CD45-PE-positive cells. Apoptotic bodies were obtained after UV irradiation. Only floating cells were collected and analyzed by flow cytometry. Results shown correspond to mean values + S.D. from triplicates of one representative experiment out of four.

The binding of 3′HSO^sub 3^Le^sup x^ and 3′HSO^sub 3^Le^sup a^ to THP-1 cells (Fig. 1A) indicates possible expression of L-selectin on this cell line of monocytic origin. However, this is unlikely since the profile of L-selectin ligands binding to THP-1 cells did not fit with their profile of affinity for L-selectin. Indeed, 6HSO^sub 3^SiaLe^sup x^ and SiaLe^sup a^ known as ligands of high affinity for L-selectin (Rosen, 1993; Hemmerich et al., 1995) bound THP-1 cells more weakly than SiaLe^sup x^ and 6′HSO^sub 3^Le^sup x^. In addition we could not observe any significant binding of the following L-selectin ligands: 6HSO^sub 3^SiaLe^sup x^ and SiaLe^sup a^. Finally, THP-1 cells interaction with 3′HSO^sub 3^Le^sup x^ and 3′HSO^sub 3^Le^sup a^ was not inhibited by antibodies against E-, P- and L-selectins, but was inhibited by free 3′HSO^sub 3^Le^sup x^ and 3′HSO^sub 3^Le^sup a^ as well as their polyacrylamide conjugates (data not shown). Thus, available results indicate possible expression on THP-1 cells of a receptor different from selectins but capable of recognising the sulfated Gal motif in a complex ligand, e.g. the mannose receptor. The macrophage mannose receptor, besides the Man-containing domain is known to have a second, cysteine-reach one interacting with sulfated Ga1NAc residue (Roseman and Baenziger, 2001). There is evidence that 3′HSO^sub 3^Le^sup x^, 3′HSO^sub 3^Le^sup a^, and some other oligosaccharides with a sulfate at the 3- OH of the galactose residue can bind to the macrophage mannose receptor (Liu et al., 2001). One cannot exclude that binding of some sulfated probes to THP-1 cells as we observed also involves the second site of the mannose receptor. Regardless, even if we actually manage to reveal a mannose receptor on the THP-1 cells by sulfated probes, it is unlikely to participate in the apoptotic bodies phagocytosis – as deduced from the absence of phagocytosis inhibition by the conjugate 3′HSO^sub 3^Le^sup x^ – PAA (Fig. 5B).

We also looked for the presence of a siglec-type activity on the THP-1 cell line. A strong labeling of these cells by 3′SiaLac and 6′SiaLac, typical ligands of siglecs (Kelm et al., 2001), was observed (Fig. 1B). Furthermore, the THP-1 cells displayed strong staining by the disaccharide [Neu5A[alpha]2-8]^sub 2^ probe, which was found to be a high-affinity ligand for siglec 7 (Yamaji et al., 2002). To understand if THP-1 siglecs are potential binding partners for apoptotic bodies, we tried to reveal sialylated chains on these bodies using the plant lectins SNA and MAA that recognise Neu5Ac[alpha]2-6Gal and Neu5Ac[alpha]2-3Gal, respectively. It turned out that sialic acid-containing terminal residues were revealed both on the parent MELJUSO cells and on the apoptotic bodies derived therefrom (Fig. 2B). Thus prerequisites for THP-1 cell siglec involvement in the phagocytosis of apoptotic bodies are fulfilled: THP-1 cells display the sialoside-binding activity, at the same time sialic acid-containing residues do not disappear from MELJUSO cells during apoptosis. However, neither SNA, MAA nor the polyvalent conjugates 3′SiaLac-PAA, 6′SiaLac-PAA and [Neu5Ac[alpha]2-S]^sub 2^- PAA inhibited phagocytosis of apoptotic bodies by THP-1 cells. Thus, there is no reason to consider siglecs, selectins or the mannose receptor as involved in elimination of apoptotic bodies by THP-1 cells.

It is known that galectin-1, galectin-3 and galectin-8 mRNAs are expressed in THP-1 cells (Lahm et al., 2001). The oligosaccharide GM1 has been shown to be the ligand of highest affinity for galectin- 1 (Gabius et al., 1990; Kopitz et al., 1998). The sialic acid residue in Gal[beta]1 – 3GalNAc[beta]1 – 4(Neu5Ac7agr;2- 3)Gal[beta]1 – 4Glc-Cer does not evidently affect the binding to galectin since asialoGM1 and GM1 inhibited erythrocyte agglutination by galectin-1 (BBL-14) equally well (Kannan and Appukuttan, 1993, 1997). To reveal galectin-1 on THP-1 cells, we employed probes containing the asialoGM1 tetrasaccharide, Gal[beta]1 – 3GalNAc[beta]1 – 4Gal[beta]1 – 4GlcNAc as well as the terminal disaccharide Gal[beta]1 – 3GalNAc[beta]. Galectins are known to recognize also terminal GalNAc residues (Leffler, 2001). It was thus of interest to study GalNAc[beta]1 – 4GlcNAc (Lac-diNAc) as a ligand. We observed that after activation, the THP-1 cells bound to LacNAc, Gal[beta]1 – 3GalNAc[beta], GalNAc[beta]1 – 4GlcNAc, and asialoGM1 (Fig. 1C). Since terminal [beta]-galactoside residues detected by the PNA and ECA lectins were present on apoptotic bodies the proposed involvement of galectins on the one hand, and of galactose chains on the other, was tested in phagocytosis inhibition experiments. Blocking of Gal-terminated carbohydrate chains by lectins PNA and ECA significantly decreased phagocytosis. Phagocytosis was also specifically inhibited by Gal[beta]1 – 3GalNAc[beta]-PAA and asialoGM1-PAA indicating that the Gal[beta]1 – 3GalNAc[beta] motif is a crucial structure in the THP-1-apoptotic cell interaction. Finally, MAbs specific for the Gal[beta]1 – 3GalNAc[beta] motif could also inhibit phagocytosis of apoptotic bodies by THP-1 cells.

In conclusion, Gal[beta]1 – 3GalNAc[beta] proved to be a major glycotope for the phagocytosis of apoptotic bodies by THP-1 cells. Noteworthy, the ss-configuration of the GalNAc moiety is important, neither TF-PAA nor TF-specific MAbs affected the phagocytosis. However, we do not have enough data to link the Gal[beta]1 – 3GalNAc[beta] motif to any specific glycoconjugate. Perhaps it is a glycolipid, namely asialoGM1 as suggested by the phagocytosis inhibition by anti-asialoGM1 MAbs, which do not recognize the sialylated ganglioside GM1. Nevertheless, we cannot exclude that glycoproteins of apoptotic bodies are involved in the phagocytosis. The question as to which lectin mediates THP-1 phagocytosis, is still to be tackled. Data on the carbohydrate specificity of phagocytosis suggest the involvement of galectin-1. Our further studies are to be focused on the identification of the molecules acting as phagocytosis mediators.

Acknowledgements. This work was supported by a grant from INTAS grant N97 – 32036 and DRI/637 MAE PECO 1999 grant from INSERM, France. Presently, the work is supported by a grant from the Russian Foundation for Basic Research (# 01-04-49253). We are very thankful to Dr. A. Buryakov (Moscow, Russia) for synthesis of poly(2- hydroxyethyl methacrylate).

European Journal of Cell Biology 82, 295-302(200\3, June) [middot] (C) Urban & Fischer Verlag [middot] Jena http:// www.urbanfischer.de/journals/ejcb

0171-9335/03/82/06-285 $15.00/0

References

Baidina, O. V., Korchagina, E. J., Bovin, N. V. (2000): Chemical synthesis of ganglioside GA^sub 1^ tctrasaccharide, Abstract. 20th Int. Carbohydrate Symp., LCI Publisher GmbH, Hamburg, Germany, p. 212.

Bratton, D. L., Rose, D. M., Oearson, A., Ezekowitz, R. A., Henson, P. M. (2000): A receptor for phosphatidylserine-specific clearance of apoptotic cells. Nature 405, 85-90.

Cao, Y., Karsten, U, Liebrich, W., Haensch, R., Springer, G. F., Schlag, P. (1995): Expression of Thomsen-Friedenreich-related antigens in primary and metastatic colorectal carcinomas: a reevaluation. Cancer 76, 1700-1708.

Deichman, G. (2000): Natural selection and early phenotypic changes of tumor cells in vivo: acquisition of new defence mechanisms. Biochemistry (Moscow) 65, 92-111.

Dini, L., Autuori, F., Lentini, A., Oliverio, S., Piacentini, M. (1992): The clearance of apoptotic cells in the liver is mediated by the asialoglycoprotein receptor. FEBS Lett. 296, 174-178.

Ezekowitz, R. A. B., Sastry, K., Bailly, P., Warner, A. (1990): Molecular characterization of the human macrophage receptor: demonstration of multiple carbohydrate recognition-like domains and phagocytosis of yeast in Cos-1 cells. J. Exp. Med. 172, 1785-1794.

Fadok, V. A., Savill, J. S., Haslett, C, Bratton, D. L., Doherty, D., Campbell, P. A., Henson, P. M. (1992): Different populations of macrophages use either the vitronectin receptor or the phosphatidylserine receptor to recognize and remove apoptotic cells. J. Immunol. 149, 4029-4035.

Fidler, I. J., Kleinerman, E. S. (1993): Therapy of cancer metastasis by systemic activation of macrophages: from the bench to the clinic. Res. Immunol. 144, 284-287.

Floyd, H., Ni, J., Cornish, A. L., Zeng, Z., Liu, D., Carter, K. C., Steel, J., Crocker, P. (2000): Siglec-8. A novel eosinophil- specific member of the immunoglobulin superfamily. J. Biol. Chem. 275, 861-866.

Fraser, I., Hughes, D., Gordon, S. (1993): Divalent cation- independent macrophage adhesion inhibited by monoclonal antibodies to murine scavenger receptor. Nature 364, 343-346.

Gabius, S., Kayser, K., Hellmann, K., Ciesiolka, T, Trittin, A., Gabius, H.-J. (1990): Carrier-immobilized derivatized lysogangliosidc GM, is a ligand for specific sites in various human tumor cell types and peripheral blood lymphocytes and monocytes. Biochcm. Biophys. Res. Commun. 169, 239-244.

Galanina, O., Tuzikov, A., Rapoport, E., LePendu, J., Bovin, N. (1998): Carbohydrate-based probes for detection of cellular lectins. Anal. Biochem. 265, 282-289.

Granville, D. J., Carthy, C. M., Hunt, D. W. C, McManus, B. M. (1998): Apoptosis: molecular aspects of cell death and disease. Lab. Invest. 78, 893-913.

Hall, S., Savill, J. S., Henson, P.M., Haslett, C. (1994): Apoptotic neutrophils are phagocytosed by fibroblasts with participation of the fibroblast vitronectin receptor and involvement of a mannose/fucose-specific lectin. J. Immunol. 153, 3218-3227.

Hemmerich, S., Leffler, H., Rosen, S. D. (1995): Structure of the Oglycans in GlyCAM-1, an endothelial-derived ligand for L-selectin. J. Biol. Chem. 270, 12035-12047.

Kannan, V. M., Appukuttan, P. S. (1993): Sugar-specific interaction of bovine brain beta-galactoside-binding lectin with endogenous gangliosides. Indian J. Biochem. Biophys. 30, 111-116.

Kannan, V. M., Appukuttan, P. S. (1997): Relative affinities of bovine brain 14-kDa galactose lectin binding to individual endogenous gangliosides. Indian J. Biochem. Biophys. 34, 249-252.

Karsten, U., Butschak, G., Cao, Y., Goletz, S., Hanisch, F.-G. (1995): A new monoclonal antibody (A78-G/A7) to the Thomsen- Friedenreich pan-tumor antigen. Hybridoma 14, 37-44.

Kelm, S. (2001): Ligands for siglecs. In: P. Crocker (ed.): Mammalian carbohydrate recognition systems. Results and problems in cell differentiation. Springer-Verlag, Berlin, pp. 154-176.

Kopitz, J., von Reitzenstein, C., Burchert, M., Cantz, M., Gabius, H.-J. (1998): Galectin-1 is a major receptor for ganglioside GM1, a product of the growth controlling activity of a cell surface ganglioside sialidase, on human neuron cells in culture. J. Biol. Chem. 273, 11205-11211.

Lahm, H., Andre, S., Hoeflich, A., Fischer, J. R., Sordat, B., Kaltner, H., Wolf, E., Gabius, H.-J. (2001): Comprehensive galectin fingerprinting in a panel of 61 human tumor cell lines by RT-PCR and its implications for diagnostic and theraupetic procedures. J. Cancer Res. Clin. Oncol. 127, 375-386.

Langermans, J. A. M., Hazenbos, W. L. W., van Furth, R. (1994): Antimicrobial functions of mononuclear phagocytes. J. Immunol. Methods 17, 185-194.

Leffler, H. (2001): Galectins structure and function – a synopsis. In: P. Crocker (ed.): Mammalian carbohydrate recognition systems. Results and problems in cell differentiation. Springer- Verlag, Berlin, pp. 57-83.

Liu, Y., Misulovin, Z., Bjorkman, P. J. (2001): The molecular mechanism of sulfated carbohydrate recognition by the cysteine-rich domain of mannose receptor. J. Mol. Biol. 305, 481-490.

Munday, J., Floyd, H., Crocker, P. (1999): Sialic acid binding receptors (siglecs) expressed by macrophages. J. Leukoc. Biol. 66, 705-710.

Nicoll, G., Ni, J., Liu, D., Klenerman, P., Munday, J., Dubock, S., Mattei, M.-G., Crocker, P. (1999): Identification and characterization of a novel siglec, siglec-7, expressed by human natural killer cells and monocytes. J. Biol. Chem. 274, 34089- 34095.

Perillo, N. L., Marcus, M. E., Baum, L. G. (1998): Galectins: versatile modulators of cell adhesion, cell proliferation and cell death. J. Mol. Med. 76, 402-412.

Rapoport, E., LePendu, J. (1999): Glycosylation alterations of cells in late phase apoptosis from colon carcinomas. Glycobiology 9, 1337-1343.

Reichert, F., Saada, A., Rotshenker, S. (1994): Peripheral nerve injury induces Schwann cells to express two macrophage phenotypes: phagocytosis and the galactose-specific lectin MAC-2. J. Neurosci. 14, 3231-3245.

Rigotti, A.,. Acton, S. L., Krieger, M. (1995): The class B scavenger receptors SR-BI and CD 36 are receptors for anionic phospholipids. J. Biol. Chem. 270, 16221-16224.

Roseman, D. S., Baenziger, J. U. (2001): The mannose/N- acetylgalactosamine-4-SO^sub 4^ receptor displays greater specificity for multivalent than monovalent ligands. J. Biol. Chem. 276, 17052-17057.

Rosen, S. D. (1993): L-selectin and its biological ligands. Histochemistry 100, 185-191.

Sakamaki, T, Imai, Y., Irimura, T. (1995): Enhancement in accessibility to macrophages by modification of mucin-type carbohydrate chains on a tumor cell line: role of a C-type lectin of macrophages. J. Leukoc. Biol. 57, 407-414.

Scubitz, K. M., Pessano, S., Rottero, L., Ferrero, D., Rovera, J., August, J. T (1983): Human granulocyte surface molecules identified by murine monoclonal antibodies. J. Immunol. 131, 1882- 1888.

Stahl, P., Ezekowitz, R. A. B. (1998): The mannose receptor is a pattern recognition receptor involved in host defense. Curr. Opin. Immunol. 10, 50-55.

Tsuchiya, S., Yamabe, M., Yamaguchi, Y., Kobayashi, Y., Konno, T. (1980): Establishment and characterization of a human acute monocytic leukemia cell line (THP-1). Int. J. Cancer 26, 171-176.

Vlasova, E., Byramova, N., Tuzikov, A., Zhigis, L., Rapoport, E., Khaidukov, S., Bovin, N. (1994): Monoclonal antibodies directed to the synthetic carbohydrate antigen Le^sup y^. Hybridoma 4, 295-301.

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 alpha2,6-sialyl residues: a comparison with siglec-9. J. Biol. Chem. 277, 6324-6332.

Eugenia Rapoport(a), Sergei Khaidukov(a), Olga Baidina(a), Vladimir Bojenko(b), Ekaterina Moiseeva(a)”, Galina Pasynina(a), Uwe Karsten(c), Nikolay Nifant’ev(d), Jacques LePendu(e), Nicolai Bovin1)a

a Shemyakin and Ovchinnikov Institute of Bioorganic Chemistry RAS, Moscow/Russia

b Russian Scientific Center for Roentgen Radiology, Moscow/ Russia

c Max Delbruck Centre for Molecular Medicine, Berlin/Germany

d Zelinsky Institute of Organic Chemistry RAS, Moscow/Russia

e INSERM U419, Institut de Biologie, Nantes/France

Received October 15, 2002

Received in revised version January 6, 2003

Accepted January 18, 2003

1) Dr. Nicolai Bovin, Shemyakin and Ovchinnikov Institute of Bioorganic Chemistry RAS, ul. Miklukho-Maklaya 16/10, 117997 Moscow/ Russia, e- mail: bovin@carb.siobc.ras.ru, Fax: +70953305592.

Copyright Urban & Fischer Verlag Jun 2003