Glycobiology Advance Access originally published online on November 6, 2008
Glycobiology 2009 19(2):172-181; doi:10.1093/glycob/cwn122
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Exposure of 
2,6-sialylated lactosaminic chains marks apoptotic and necrotic death in different cell types
3 Department of Experimental Pathology, University of Bologna, Via S. Giacomo 14
4 Department of Biochemical, Experimental, and Clinical Sciences, University of Bologna, Via Belmeloro 8, 40126 Bologna, Italy
1 To whom correspondence should be addressed: Tel: +39-0512094727; Fax: +39-0512094746; e-mail: fabio.dallolio{at}unibo.it
Received on July 25, 2008; revised on October 24, 2008; accepted on October 31, 2008
 |
Abstract |
|---|
 Top Abstract IntroductionResultsDiscussionMaterial and methodsSupplementary DataFundingConflict of interest statementReferences |
|---|
Many observations have reported glycosylation changes associated with apoptosis in different biological systems, although none of these has shown any general significance. In this work, we show that in cell lines from different histological origin, (colon, breast, pancreas, and bladder cancer) as well as in normal human and mice neutrophils, apoptosis is accompanied by the exposure of sugar chains recognized by the lectin from Sambucus nigra (SNA), specific for Sia
2,6Gal/GalNAc structures. Also, cells undergoing primary necrosis induced by heat treatment (56°C, 30 min) expose specifically binding sites for SNA. While this modification is recognized also by the lectin from the mushroom Polyporus squamosus, which is highly specific for 2,6-sialylated lactosamine, no significant changes were detected in the binding of lectins specific for other carbohydrate structures, such as those from Phaseolus vulgaris, Arachis hypogea, and Maackia amurensis. The binding of SNA to apoptotic/necrotic cells is inhibited by neuraminidase treatment and by 2,6-sialylated compounds. In apoptotic, but not in necrotic SW948 cells, SNA reactivity is specifically associated with 65, 69, and 87 kDa glycoproteins. The exposure of SNA-reactive chains by apoptotic/necrotic cells occurs also in cells not expressing sialyltransferases ST6Gal.1 or ST6Gal.2 and is largely independent of the presence of 2,6-sialylated lactosaminic chains on the surface of preapoptotic cells. In neutrophils from ST6Gal.1 knock-out mice, the apoptosis-related increase in SNA reactivity is reduced but not abolished. These data demonstrate that apoptosis and primary necrosis induce a specific glycosylation change independent of the cell type and nature of the stimulus. Key words: apoptosis / cell death / necrosis / Sambucus nigra agglutinin / sialylation
|
Introduction |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterial and methodsSupplementary DataFundingConflict of interest statementReferences |
|---|
The external surface of cells is covered by a carbohydrate layer, known as the glycocalyx, which mediates many interactions among cells and between cells and the environment. The final steps of the apoptotic program involve the expression of "eat me signals" on the apoptotic bodies; the most important signal is the exposure of phosphatidylserine, which allows prompt recognition and removal of the demised cells by phagocytes. Several studies have reported marked changes in the glycosylation pattern of cells undergoing apoptosis. However, any attempt to identify glycosylation changes unequivocally associated with apoptosis has so far been unsuccessful.
In this paper, we report that apoptotic cells, but also cells undergoing primary necrosis, become strongly reactive with the Sambucus nigra (elderberry) agglutinin (SNA), a lectin specific for 2,6-linked sialic acid. The phenomenon is common to cells of different histological origin, including colon, breast, pancreas, bladder, and neutrophils and appears to be independent of the nature of the apoptotic stimulus. The main carbohydrate structures recognized by SNA are 6'-sialyllactosamine (Sia2,6Galβ1,4GlcNAc) and sialyl-Tn (Sia2,6GalNAc), both of which are often cancer associated (Dall’Olio and Chiricolo 2001
). The recognition of apoptotic/necrotic cells by SNA is dependent on its known sugar-binding activity while independent of 2,6-sialyltransferase expression. The lectin from the mushroom Polyporus squamosus, which has a narrow specificity for 2,6-sialylated lactosaminic chains, also marks apoptotic and necrotic cells, indicating that the sugar chains exposed by apoptotic/necrotic cells are 2,6-sialylated lactosamines.
|
Results |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterial and methodsSupplementary DataFundingConflict of interest statementReferences |
|---|
SNA reactivity increases in apoptotic colon cancer cell lines not expressing ST6Gal sialyltransferases
Our study started from the observation that colon cancer cell lines expressing barely detectable levels of
2,6-sialyltransferases acting on lactosaminic chains, such as SW48 and SW948, or weakly expressing the activity, such as HT-29 (Dall’Olio et al. 1992), became strongly positive to the 2,6-sialyl-specific lectin SNA (Shibuya et al. 1987) when undergoing apoptosis. This is evident from Figure 1A, which shows the flow cytometric profiles of lectin-stained cell lines analyzed 48 h after induction of apoptosis by the intracellular zinc chelator TPEN. Other lectins, such as leukoagglutinin from Phaseolus vulgaris (L-PHA), which recognizes the β1,6 branching of N-linked chains (Hammarstrom et al. 1982), Maackia amurensis agglutinin (MAA), which is specific for 2,3-linked sialic acid (Knibbs et al. 1991), and Arachis hypogea (peanut agglutinin, PNA), which recognizes the unsubstituted T antigen (Galβ1,3GalNAc-Ser/Thr) of O-linked chains (Lotan et al. 1975), displayed a totally different behavior. Reactivity with L-PHA, MAA, and PNA showed little changes in apoptotic cells, compared with control cells. Only in SW48 cells, the bimodal pattern of staining in the normal cell was replaced by a more homogenous pattern in apoptotic cells. In contrast, SNA reactivity increased dramatically in the three cell lines undergoing apoptosis. The occurrence of apoptosis was monitored by annexin-V reactivity, which reacts with phosphatidylserine, while membrane integrity was monitored with 7-amino-actinomycin D (7-AAD). As shown, at this stage of the apoptotic process, the three cell lines showed increased annexin-V reactivity, while the staining with 7-AAD was negative, indicating that the three cell lines were undergoing apoptotic cell death and that the integrity of the plasma membrane was conserved. The level of caspase 3 activation in these cells was 3- to 4-fold higher than that of control cells (data not shown). Other apoptotic stimuli, such as camptothecin, induced very similar effects (supplementary Figure 1), indicating that the increase in SNA reactivity was independent of the nature of the apoptotic stimulus.
|
To rule out the possibility that the increased SNA reactivity of apoptotic cells was due to unspecific phenomena, such as the entrance of the lectin in the dying cells, we subjected cell homogenates of untreated and apoptotic SW948 cells, as well as of SW948 cells transduced with a ST6Gal.1 retroviral expression vector to dot blot analysis with digoxigenin-conjugated SNA (Figure 1B). In a typical experiment, the reactivity of apoptotic cells was about double that of untreated cells and very close to that of ST6Gal.1-expressing cells. This datum strongly suggests the de novo exposure of
2,6-sialylated sugar chains in apoptotic cells. The nature of the structures recognized by SNA in apoptotic cells was preliminarily investigated by treating apoptotic SW948 with proteinase K. As shown in Figure 1C, this treatment reduces the level of SNA reactivity by about 100 fluorescence channels, strongly suggesting that the molecules recognized by SNA in apoptotic cells are carried, at least in part, by glycoproteins.
SNA reactivity increases in cells undergoing primary necrosis
Primary necrosis induced by heat treatment (56°C, 30 min) increased SNA reactivity leaving unaltered the reactivity with other lectins (Figure 2A). As shown by FACS analysis, SW48 cells undergoing primary necrosis displayed a near unaltered level of reactivity with L-PHA, MAA, and PNA and a dramatic increase of SNA reactivity.
|
Figure 2B shows the fluorescence microscopy of untreated, primarily necrotic, or apoptotic SW48 cells, stained with the four lectins. In agreement with FACS analysis data (Figures 1A and 2A), SW48 cells displayed a basically low level of reactivity with MAA and SNA and a good, although heterogeneous, level of L-PHA and PNA reactivity. Upon induction of primary necrosis, the staining pattern with L-PHA, MAA, and PNA showed a little or no changes, while SNA reactivity increased dramatically. The staining was localized mainly on the cell membrane, although some diffuse cytoplasmic staining was also visible. In some cells, the reactivity showed a polarized distribution. In apoptotic cells, the reactivity with L-PHA and PNA showed little or no changes, while a slight and heterogeneous increase in MAA reactivity and a dramatic increase in SNA reactivity were evident. The cellular distribution of SNA reactivity in apoptotic cells was not identical to that of necrotic cells. In fact in the former, the reactivity appeared to be more profoundly associated with the cytoplasm with little staining associated with the plasma membrane. Nuclear staining of apoptotic cells with Hoechst 33258 revealed the presence of numerous apoptotic bodies. Taken together, these results indicate that both apoptotic and necrotic death result in a highly specific increase in reactivity with SNA but not with other lectins.
The data reported in Figure 2C have been obtained inducing primary necrosis in the cell line SW948. The necrosis-induced increase in SNA reactivity was similar to that obtained with SW48 cells. Interestingly, if the heat treatment was preceded by formalin fixation, the increase of SNA reactivity was not observed. This strongly suggests that the exposure of SNA binding sites is an active process which can be blocked by the sudden death of the cell induced by formalin fixation. It should be noted that the SNA reactivity appears to be polarized in a restricted area of the cell membrane (Figure 2C, arrows).
SNA treatment does not induce any toxic effect per se
It has been reported that some lectins (Franz et al. 2006), including SNA (Batisse et al. 2004), induce toxic effects on cells. In order to establish whether our conditions for SNA staining were responsible of any toxic effect per se, we took advantage of two physical parameters revealed by the cytometer: forward scatter (FSC), which provides information on the cell size, and side scatter (SSC), which provides information on cell granularity. In both SW48 and SW948 cells, TPEN treatment induced a decrease in the FSC value (Figure 3), which is indicative of the reduced cell size, as expected on the basis of the typical apoptosis-dependent cell shrinkage, whilst heat treatment known to induce primary necrosis induced both a decrease of cell volume and an increase in cellular granularity. On the other hand, the physical parameters of SNA-stained SW48 and SW948 cells were very similar, if not identical, to those of unstained cells, ruling out any relevant toxic effect of this lectin on these cell lines under our conditions of binding.
|
Features of SNA reactivity of apoptotic or necrotic cells
The most obvious explanation for the observed increase in SNA reactivity in apoptotic and primarily necrotic cells is the activation of
2,6-sialyltransferase(s) working on lactosaminic chains. Two sialyltransferases are able to catalyze such reaction: the well-known ST6Gal.1, which is active on glycoproteins and oligosaccharides (Weinstein et al. 1987), and ST6Gal.2, which is active only on free oligosaccharides (Takashima et al. 2002; Krzewinski-Recchi et al. 2003). To measure the activity of both enzymes in a single test, we used the disaccharide lactosamine as an acceptor. While the activity (expressed as pmoles of radioactive sialic acid incorporated per hour mg of protein homogenate) of an homogenate of ST6Gal.1-transduced SW948 cells is around 140, the activity of untreated, apoptotic, and necrotic SW948 cells ranges between 2 and 3 (Figure 4A), ruling out the possibility that the increased SNA reactivity was due to sialyltransferase activation.
|
SNA lectin blot analysis of homogenates of untreated, apoptotic, necrotic, and ST6Gal.1-transduced SW948 cells (Figure 4B) shows that in apoptotic cells, the SNA reactivity is specifically associated with bands showing molecular weights of 65, 69, and 87 kDa. Interestingly, the SNA reactivity in ST6Gal.1-transduced cells appears to be associated mainly with 240 kDa and 140 kDa bands. Apart from a weak band with a molecular weight of about 95 kDa, no SNA-reactive band appears to be specifically present in necrotic cells. These data confirm that in apoptotic cells, the SNA reactive molecules are at least in part glycoproteins and that these glycoproteins are different from those which are normally sialylated by ST6Gal.1. The absence of specific SNA-reactive bands in necrotic cells indicates that the SNA reactive molecules expressed during primary necrosis are different from those exposed during apoptosis and might not be glycoproteins.
Treatment of either apoptotic or necrotic cells with neuraminidase strongly reduces the SNA reactivity (Figure 4C). In a typical experiment, the enzymatic treatment reduced the fluorescence intensity of apoptotic cells from 700 to 190 fluorescence channels (FC) and that of necrotic cells from 1050 to 315 FC. This datum confirms that SNA recognizes sialic acid-containing epitopes on apoptotic/necrotic cells.
The recognition of necrotic and apoptotic cells by SNA is dependent on its native structure, in that heat inactivation of the lectin reduced its binding by about 75% (Figure 4D). Moreover, the binding can be inhibited by about 50% by compounds containing 2,6-sialylated lactosaminic chains, such as fetuin or its N-linked glycopeptide, but not by the asialoglycopeptide (Figure 4D). Altogether, these data indicate that the recognition of apoptotic and necrotic cells by SNA depends on its known sugar binding ability.
The sugar chain recognized by SNA on apoptotic and necrotic cells is 2,6-sialylated lactosamine
It is well known that SNA can bind, besides 2,6-sialylated lactosamine, also the sialyl-Tn antigen (Sia2,6GalNAc-Ser/Thr), a sugar structure which is not the product of ST6Gal.1 but mainly of sialyltransferase ST6GalNAc.1. To establish whether the sialyl-Tn antigen was responsible for the recognition of necrotic and apoptotic cells by SNA, the cells were stained with anti sialyl-Tn antibodies and FACS analyzed. As a positive control, the bladder cancer cell line MCR transduced with a retrovirus carrying the ST6GalNAc.1 cDNA was used. While MCR-transduced cells displayed a strong reactivity with the anti sialyl-Tn antibody, both apoptotic and necrotic SW948 cells were found to be negative (data not shown).
The lectin from the mushroom Polyporus squamosus (PSL) has been recently found to be a more specific tool for 2,6-sialylated lactosamine than SNA (Toma et al. 2001). In order to obtain more information about the nature of the sugar chains expressed by apoptotic and necrotic cells and to confirm the observations made with SNA with a lectin of similar specificity but from a totally unrelated species, we analyzed apoptotic, necrotic, and ST6Gal.1-transduced SW948 cells with PSL and compared the reactivity with that of SNA (Figure 5A). FACS analysis and fluorescence microscopy showed that both lectins failed to stain untreated cells while they reacted with necrotic, apoptotic, and ST6Gal.1-transduced SW948 cells. Together, these data indicate that the sugar chains recognized by SNA in necrotic/apoptotic cells were 2,6-sialylated lactosaminic chains.
|
Apoptosis does not change the FACS profile of SNA-stained ST6Gal.1-expressing cells
The cell lines used in the first part of this study express ST6Gal.1 at a very low level and consequently bind poorly to SNA. If apoptosis was induced in a cell line strongly reactive with SNA because of the constitutive expression of ST6Gal.1, such as SW948 cells transduced with the lentiviral ST6Gal.1 vector, the flow cytometric profile did not differ from that of nonapoptotic cells (Figure 5B). However, fluorescence microscopy revealed that the morphology of apoptotic ST6Gal.1-expressing cells was indistinguishable from that of apoptotic SW948 cells, while it was different from that of nonapoptotic ST6Gal.1-expressing cells, because in the latter the reactivity was associated exclusively with the plasma membrane (Figure 5B). This datum indicates that the increased SNA binding displayed by apoptotic cells was independent of preexisting
2,6-sialylated structures on the cell membrane.
SNA marks apoptosis in cells of various tissue origins
Apoptotic cell lines of noncolonic origin such as breast cancer (MCF-7) and pancreatic cancer (Panc1) displayed an increased SNA reactivity very similar to that described in colon cancer cell lines (Figure 6A).
|
Also the bladder cancer cell line HT1376 displayed a strong increase in SNA reactivity when undergoing apoptosis (Figure 6B). The SNA reactivity was acquired in parallel to the membrane reactivity with annexin-V, which is indicative of phosphatidylserine exposure. Both processes started around 24 h after the administration of the apoptotic stimulus and reached a maximum at 48–72 h. It should be noted that during this time, the plasma membrane remained impermeable to small molecules such as propidium iodide (PI).
The relationship between apoptosis and SNA reactivity was further investigated in human neutrophils (or polymorphonuclear leukocytes, PMNs), blood cells which die spontaneously by apoptosis at the end of their short life (48–72 h), thus providing an ideal model of apoptosis in the absence of exogenous stimuli. FACS analysis of cells stained with SNA just after the isolation from the blood (very few apoptotic cells) revealed a low and homogeneous level of reactivity with the lectin (Figure 6C). Fluorescence microscopy revealed that these cells had the typical polylobate nuclei, characteristic of PMNs. Twenty hours after the isolation, a percentage of cells became annexin-V positive; this was accompanied by an increased SNA reactivity (mean fluorescence channels 298 versus 195). The level of caspase 3 activation of these cells was about 5-fold higher than that of freshly isolated cells (data not shown). Morphologically, the nuclei of apoptotic PMNs lose the polylobate shape and acquire a round morphology (Akgul et al. 2001). The fluorescence microscopy of the 20 h sample revealed that cells showing a round nuclear morphology (arrow) were also strongly SNA positive. At 44 h after the blood isolation, the vast majority of PMNs became both annexin-V and SNA positive. Together, these results indicate that the increase in SNA reactivity accompanies the apoptotic process in cells of different histological origin, regardless of the nature of the apoptotic stimulus.
SNA marks apoptotic PMNs of both wild-type and ST6Gal.1 knock-out mice, although at a different intensity
Mice knocked out for Siat1, the gene encoding ST6Gal.1 (Hennet et al. 1998) (Siat1 null), provide an ideal model to study the role played by the enzyme and/or by an 2,6-sialylated environment in the expression of the apoptosis-associated SNA reactivity. Leukocyte migration was elicited according to the thioglycollate model of inflammatory response (Nasirikenari et al. 2006). PMNs obtained by peritoneal lavage were SNA stained and FACS analyzed, just after the isolation or after 24 h of culture to allow the developing of spontaneous apoptosis. A comparison of the SNA staining patterns of cells from WT mice, incubated for 0 or 24 h (Figure 7 top left panel), reveals in both the presence of a SNA-negative cell population and of a cell population with intermediate reactivity (between 50 and 500 FC). However, in 24 h-incubated cells only, a strongly SNA-reactive population (FC >1000) appears; this population is also annexin-V positive (Figure 7 bottom-left panel), indicating that also in a murine model, spontaneous apoptosis of PMNs is associated with a strong increase in SNA reactivity. In 24 h incubated cells from Siat1 null mice, the cell population with SNA reactivity >1000 FC is lacking (Figure 7 top right). However, a population of apoptotic cells displaying a level of SNA reactivity between 200 and 1000 FC is still present. This datum indicates that apoptotic PMNs acquire a SNA-positive phenotype, albeit at a lower level, even in an ST6Gal.1-negative background. A similar pattern is seen when analysis is restricted to cells expressing Ly6G+, a murine granulocyte-specific antigen (Figure 7 middle panels). In both WT and Siat1 null mice, the SNA- and annexin-V-positive populations coincide (Figure 7 bottom panels).
|
The SNA-reactive sugar chains of apoptotic cells are not derived from soluble glycoproteins of the extracellular environment
A possible mechanism explaining the presence of SNA-reactive molecules on the surface of apoptotic cells poorly expressing ST6Gal.1 involves the transfer of
2,6-sialylated sugar chains or of 2,6-linked sialic acid from 2,6-sialylated glycoproteins present in the extracellular milieu (e.g., serum) to the surface of the cells. Different mechanisms can be hypothesized to be responsible for this transfer and will be discussed in the next section. In any case, regardless of the mechanism, to ascertain whether there is a transfer of sialylated chains or of sialic acid to the apoptotic cells, the glycoprotein fetuin, previously chemically radiolabeled in the sialic acid residues, was included in the medium of cells undergoing apoptosis. Apoptotic and control cells were harvested at 24, 48, and 72 h after the administration of the apoptotic stimulus; a small aliquot of the cells was used to measure the amount of cell-associated radioactivity while the rest were analyzed by electrophoresis, followed by autoradiography. The amount of radioactivity taken up by the cells never exceeded 0.05% of the total radioactive fetuin added. Analysis of the cell-associated radioactivity revealed only one band, having the same molecular weight of fetuin (supplementary Figure 2). A transfer of either 2,6-sialylated chains or 2,6-linked sialic acid to cells would have given rise to multiple bands. This datum makes unlikely the possibility that some kind of transfer of 2,6-linked sialic acid from the glycoproteins of the environment to cells occurred during the apoptotic process. The small amount of cell-associated fetuin was probably nonspecific. |
Discussion |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterial and methodsSupplementary DataFundingConflict of interest statementReferences |
|---|
The identification and removal by phagocytes of apoptotic cells is based on their recognition through the exposure of "eat me signals" on the cell membrane. The best known of these signals is certainly represented by phosphatidylserine, which is "flopped" from the internal to the external membrane of apoptotic cells (Fadok et al. 1992
). However, other molecules are exposed and contribute to the formation of the so-called phagocytic synapse (Vandivier et al. 2006), a complex system of signals which regulate the phagocytosis of apoptotic cells. A contribution of carbohydrate molecules to this network of signals is likely because the glycocalyx forms the most external barrier of the cell (Duvall et al. 1985). Among the molecules which are exposed on the surface of apoptotic cells, there are calreticulin and mannose-binding lectin, both of which are sugar binding molecules. Sialylated compounds expressed by apoptotic bodies have been suggested to be involved in recognition by macrophages through sialic acid binding Ig-like lectins (siglecs) (Rapoport et al. 2005), while apoptotic neutrophils have been reported to be recognized and phagocytized by fibroblasts through a mechanism involving a mannose-fucose-specific lectin (Hall et al. 1994).
It is well established that the apoptotic process can induce profound alterations in the lectin-staining pattern of cell membranes (Rapoport and Le Pendu 1999; Franz et al. 2006; Sarter et al. 2007). However, no single glycosylation change has been so far unequivocally associated with apoptosis. Several apoptosis-related changes in cell glycosylation has been described so far, including a decrease (Azuma et al. 2000; Hart et al. 2000; Batisse et al. 2004) or an increase in cell sialylation (Kim et al. 2007), an increased expression of Lewis X and Y antigens (Azuma et al. 2007), an altered biosynthesis of O-glycans (Bleesing et al. 2001; Brockhausen et al. 2002), a general decrease in lectin binding (Morris et al. 1984), or increased fucosylation (Russell et al. 1998).
In this work, we show for the first time that apoptotic cells of different histological origin expose sugar chains which can be recognized by the sugar-binding activity of the 2,6-sialyl-specific lectin SNA. The reactivity is not limited to the plasma membrane but involves the cytoplasm. In cells expressing basic low levels of SNA reactivity, this phenomenon results in a dramatic change of the FACS profile. In cells expressing basic high levels of SNA reactivity, the FACS profile shows little or no changes (this could explain why this phenomenon has not been recognized previously), although SNA reactivity changes from being restricted to the cell surface to an involvement of deeper cytoplasmic structures. In this way, it is very similar to that observed in apoptotic cells characterized by low SNA-binding activity, thus indicating that the phenomenon is general. During the course of this study, we have found no exception to the rule of high SNA reactivity expression by apoptotic cells. The observation that the exposure of SNA binding sites is also detectable by dot blot analysis, together with the fact that other lectins do not show the same behavior and that SNA is a huge molecule with a MW of 240 kDa (Van Damme et al. 1996), rules out the possibility that SNA passively crosses the membranes of apoptotic cells. The specificity of the phenomenon is confirmed by the fact that neuraminidase treatment strongly reduces the SNA binding. Furthermore, the lectin PSL, which has a narrow specificity for 2,6-sialylated lactosamine and is obtained from an organism belonging to a different life kingdom (fungi instead of plants), recognizes the same modification, strongly indicating that both lectins recognize 2,6-sialylated lactosamines in apoptotic cells.
Important clues on the nature of the phenomenon have been obtained by the experiments with Siat1 null mice. These animals lack a functional ST6Gal.1 and, as a consequence, do not express 2,6-sialylated lactosaminic chains in tissues and body fluids (Hennet et al. 1998). Apoptotic neutrophils isolated from wild-type animals display a strong increase in SNA reactivity, further confirming that the phenomenon is of general significance. In neutrophils of Siat1 null mice, the phenomenon is still present, although at a reduced intensity. Thus, in agreement with the data obtained with the ST6Gal.1-negative colon cancer cell lines, SW48 and SW948, ST6Gal.1 is not strictly required for the increased SNA reactivity by apoptotic cells. However, the phenomenon is boosted in an environment rich in 2,6-sialylated sugar chains.
The molecular mechanisms underlying this phenomenon are far from clear. Although in some cases the expression of sugar epitopes in apoptotic cells has been shown to be dependent on the modulation of the cognate glycosyltransferase (Akamatsu et al. 1996; Zhang et al. 2003; Azuma et al. 2007; Kim et al. 2007), in this case there is evidence that no activation of the 2,6-sialyltransferases acting on lactosamine has occurred. A mechanism that has been proposed to explain the altered lectin binding pattern of apoptotic cells involves the exposure on the cell membrane of molecules resident in the endoplasmic reticulum, including incompletely processed sugar chains (Franz et al. 2007). This could explain the altered binding pattern to lectins that are specific for incomplete sugar structures. However, this mechanism can hardly explain the appearance of a terminally processed structure such as 2,6-sialylated lactosaminic chains.
To explain the appearance of SNA-reactive molecules in ST6Gal.1-negative cells, we considered the possibility that 2,6-sialylated sugar chains are taken up from the extracellular environment (including serum glycoproteins), a mechanism that could conceivably be mediated by different enzymatic activities. For example, endo-β-N-acetylglucosaminidase HS, an enzyme which cleaves the glycosidic bond between the two innermost N-acetylglucosamine residues of N-linked chains, can transfer the released sugar chains to molecules containing hydroxyl groups, including sugars (Ito et al. 2006). Other mechanisms could explain the transfer of sialic acid from sialylated donors to acceptors (Chandrasekaran et al. 2008). We failed to observe any evidence of the transfer of sialylated compounds from [3H]NeuAc-labeled fetuin to cell components. This datum makes unlikely that the transfer of sialylated compounds from serum glycoproteins causes the increased SNA reactivity of apoptotic ST6Gal.1-negative cells. However, the possibility of a transfer from other sugar-containing molecules remains open.
SNA-negative cell lines develop a strong SNA reactivity when induced to undergo primary necrosis by heat treatment. The cellular distribution of the reactivity appears to be different from that of apoptotic cells, being mainly associated with the plasma membrane, with a tendency to cluster at one pole of the cell. Interestingly, this phenomenon is completely blocked if the cells are fixed before heat treatment, thus strongly suggesting that the clustering is an active process which is prevented by the chemical crosslinking caused by fixation. Like in apoptotic cells, the recognition of cells undergoing primary necrosis by SNA is dependent on its sugar-binding activity, is not dependent on sialyltransferase activation, and is inhibited by neuraminidase treatment. While in apoptotic cells SNA reactivity appears to be associated with a few glycoproteins of discrete molecular weight, there are no SNA-reactive glycoproteins specifically associated with primary necrosis. This strongly suggests that the two processes activate the expression of 2,6-sialylated lactosaminic chains on different class compounds.
Collectively, the data presented in this paper indicate that cells undergoing either apoptosis or primary necrosis undergo cellular modifications resulting in the accessibility 2,6-sialylated lactosaminic structures by specific lectins, such as SNA and PSL. Unlike other glycosylation changes reported in different experimental systems, the exposure of 2,6-sialylated lactosaminic structures appears to be a general phenomenon, independent of the nature of the apoptotic stimulus and on the tissue origin of the cell. Even if the molecular bases of this change and its biological implications remain to be elucidated, it is tempting to speculate that the exposure of 2,6-sialylated lactosaminic structures is involved in the network of signals regulating the disposal of dying cells.
|
Material and methods |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterial and methodsSupplementary DataFundingConflict of interest statementReferences |
|---|
Induction and control of apoptosis and of primary necrosis
Colon cancer cell lines SW48 and SW948 were grown in the Leibovitz's L15 medium; cell lines HT29 (colon), MCF7 (breast), Panc1 (pancreas), and HT1376 (bladder) were grown in DMEM. Media were supplemented with 10% FCS and antibiotics. In routine experiments, apoptosis was induced by the intracellular zinc chelator NNN'N'-tetrakis-(2-pyridylmethyl)ethylenediamine (TPEN) treatment (10 µM) for 48 h (Chai et al. 2000
; Marini et al. 2001). In some experiments, apoptosis was induced by 5 µM camptothecin for 72 h. Apoptotic cells were collected from the medium and harvested by centrifugation. Mock-treated cells were released by trypsin treatment. The occurrence of apoptotic death was monitored by the following four parameters: the expression of phosphatidylserine on the cell membrane by binding with annexin-V (Annexin V-PE detection kit, BD Pharmingen, Milan, Italy); caspase 3 activation (CPP32 colorimetric assay kit, MBL, Woburn, MA); analysis of physical parameters by flow cytometry; and analysis of nuclei morphology in Hoechst 33258-stained cells. Primary necrosis was induced by treating trypsin-released cells at 56°C for 30 min (Franz et al. 2006). Integrity of cell membranes was monitored in FACS analysis by 7-AAD, included in the Annexin V-PE detection kit, or with PI. Staining with annexin-V and 7-AAD was performed according to manufacturer's instructions.
FACS and fluorescence microscopy analysis
Lectins SNA, L-PHA, PNA, and MAA were purchased from Sigma (St. Louis, MO), while lectin from Polyporus squamosus was purchased from EY laboratories (San Matteo, CA) and conjugated to fluorescein isothiocyanate (FITC) as described (Hoebeke et al. 1978). Cells were stained with fluorescent lectins (50 µg/mL) for 20 min in ice and FACS analyzed as previously described (Chiricolo et al. 2006). After analysis, cells were fixed in 4% formaldehyde; an aliquot of stained cells was collected by centrifugation in a minifuge and resuspended in 10 µL of phosphate buffered saline (PBS, 20 mM Tris–HCl, pH 7.4, 0.15 M NaCl) containing 100x diluted Hoechst 33258 and incubated for 20 min at room temperature. Cells were then washed with 1 mL of PBS, collected by centrifugation, resuspended in 10 µL of mounting solution, spotted onto a microscope glass, covered with a glass coverslip, and observed using a fluorescence microscope.
Transduction of the SW948 cell line with human ST6Gal.1 cDNA
The cDNA of the whole coding sequence of human ST6Gal.1 from Caco2 cells was PCR amplified using primer pair hST6Gal1L.1 (CACCATGATTCACACCAACCTGAAGAAA AAGTTCAGCTGCTGC) (the underlined sequence is necessary for cloning in the TOPO vectors and is not gene specific) and hST6Gal1.R1 (TTAGCAGTGAATGGTCCGGAAGCCA GGCAGTGTG). The PCR product was cloned in the pLenti6/V5 directional TOPO vector (Invitrogen, Carlsbad, CA) and used to transfect the 293FT cell line, according to the instruction manual of the ViraPower Lentiviral Expression System (Invitrogen). The conditioned medium, containing the virions, was centrifuged to remove cell debris, filtered on a 0.45 µM membrane, diluted with 9 volumes of fresh culture medium, added to SW948 cells, and left for 48 h. Cells were selected by the presence of 10 µg/mL blasticidin. Resistant cells were SNA-FITC stained, FACS analyzed, and found to be more than 85% positive. A mock transduction was run in parallel with a retrovirus lacking the ST6Gal.1 insert.
Determination of ST6Gal activities
Cell pellets were homogenized in water and the protein concentration of the homogenates was determined by the Lowry method. The activity of the two enzymes which could add sialic acid in 2,6-linkage to lactosaminic chains, namely ST6Gal. 1 and ST6Gal.2, was assayed using the common acceptor N-acetyllactosamine according to published procedures (Dall’Olio et al. 2006).
Dot and lectin blot analysis
Dot blots and SNA-lectin blots were probed with digoxigenin-conjugated SNA. Conditions for electrophoretic separation and blotting were as previously described (Dall’Olio et al. 2006).
Protease and neuraminidase treatments
Apoptosis or primary necrosis was induced in SW948 cells with TPEN for 48 h or by heat treatment as described above. Protease treatment was performed with 0.5 mg/mL of proteinase-K from Tritirachium album (Sigma) in serum-free DMEM for 30 min at 37°C. Cells were then washed twice by centrifugation, formalin fixed, SNA-FITC stained, and FACS analyzed as above. Neuraminidase treatment was performed with 1.3 U/100 µL of neuraminidase from Clostridium perfrigens (Sigma) in PBS for 1 h at 37°C. Then, cells were fixed and analyzed as above.
Isolation and analysis of human PMNs
Human PMNs were isolated from the blood of healthy donors as follows: the buffy coat obtained from 40 mL of EDTA-treated blood was stratified on Ficoll-Paque (Pharmacia, Piscataway, NJ) and centrifuged at 400 x g for 30 min at 20°C. The granulocytes stratified on the packed red cells were collected and diluted to 40 mL with PBS. Erythrocytes were sedimented by adding 2.5 mL 4% (w/v) dextran 500 (Pharmacia) and the PMNs were spun down from the supernatant at 250 x g for 5 min at room temperature. Contaminating red cells were removed with an osmotic lysis (1 min in 1.5 mL 0.2% NaCl), followed by the addition of 3.5 mL of 1.2% NaCl. PMNs were pelleted at 200 x g for 5 min, washed twice with PBS containing 1% (w/v) BSA, and resuspended in 1 mL of the same buffer. An aliquot was fixed with 4% formaldehyde just after the isolation (0 time). The rest of the preparation was suspended in RPMI containing 10% FCS and incubated at 37°C for 20 or 44 h. At the indicated times, cells were fixed. The fixed cells were SNA-FITC stained and FACS analyzed as described above.
Isolation and analysis of murine neutrophils
Inflammatory response and leukocyte emigration were elicited in wild-type or Siat1 null mice (Hennet et al. 1998) by the thioglycollate model of induced peritonitis (Nasirikenari et al. 2006). Briefly, 1 mL of 4% (wt/vol) sterile solution of brewer's yeast thioglycollate (Becton Dickinson Microbiology, Baltimore, MD) was administered intraperitoneally in recipient animals. Twenty-two hours later, animals were killed by CO2 asphyxiation, and peritoneal cells were recovered by peritoneal lavage with 6 mL of ice cold PBS. Lavage cells were recovered by centrifugation and resuspended in RPMI containing 10% FCS. Macrophage population was allowed to adhere to the bottom of the plastic dish by incubation for 2 h. The nonadherent fraction was transferred to the fresh culture dish, incubated for another 22 h and stained with annexin-V (Biolegend, San Diego, CA) and biotinylated SNA (Vector laboratories, Burlingame, CA). The biotin signal was monitored using streptavidin-allophycocyanin (Biolegend).
Treatment with [3H]NeuAc fetuin
Radiolabeled fetuin was prepared as described elsewhere (Van Lenten and Ashwell 1971). Briefly, to fetuin, dissolved at 10 mg/mL in buffer A (0.1 M Na acetate, pH 5.6, 0.15 M NaCl), 0.64 volumes of 12 mM NaIO4 were added. The glycoprotein solution was incubated for 10 min in ice, and then oxidation was blocked by the addition of 2 µL ethylene glycol/mL of solution. The glycoprotein solution was dialyzed against PBS and 15 µL tritiated Na borohydride (Perkin Elmer, Boston, MA) with a specific activity of 35 GBq/mmol, 45 MBq/µL, in NaOH per mL oxidized glycoprotein solution was added in a cap closed tube and incubated for 30 min. Then, 0.5 mg/mL unlabeled NaBH4 was added and after 30 min, the solution was extensively dialyzed against buffer A. The specific activity obtained was 25,000 dpm/µg protein. About 212 x 106 dpm radiolabeled fetuin in 1.5 mL of medium was added to cells seeded in 6-well plates contemporarily with the apoptotic stimulus. Cells were harvested 24, 48, or 72 h later using a cell scraper after three washes with PBS. Cell pellets were resuspended in the electrophoresis buffer, one-tenth of the sample was counted for radioactivity while the rest was subjected to electrophoretic separation and autoradiography. An aliquot of labeled fetuin was electrophoresed in parallel.
|
Supplementary Data |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterial and methodsSupplementary DataFundingConflict of interest statementReferences |
|---|
Supplementary data for this article is available online at http://glycob.oxfordjournals.org/.
|
Funding |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterial and methodsSupplementary DataFundingConflict of interest statementReferences |
|---|
Pallotti Legacy for Cancer Research and the University of Bologna.
|
Conflict of interest statement |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterial and methodsSupplementary DataFundingConflict of interest statementReferences |
|---|
None declared.
|
Acknowledgements |
|---|
We thank Professor Maurizio Brigotti and Dr. C. M. Betts (Department of Experimental Pathology of the University of Bologna) for the help with neutrophils experiments and for the critical reading of the manuscript. We gratefully thank Dr. Joseph Lau and Mehrab Nasirikenari of the Roswell Park Cancer Institute, Buffalo, NY for the help with mice experiments.
|
Footnotes |
|---|
2 These authors equally contributed to this study.
|
Abbreviations |
|---|
7-AAD, 7-amino-actinomycin D; Ann-V, annexin-V; DMEM, Dulbecco's modified Eagle medium; FC, fluorescence channels; FCS, fetal calf serum; FITC, fluorescein isothiocyanate; L-PHA, leukoagglutinin from Phaseolus vulgaris; MAA, Maackia amurensis agglutinin; PBS, phosphate buffered saline; PI, propidium iodide; PMN, polymorphonuclear leukocyte; PNA, peanut agglutinin; PSL, Polyporus squamosus lectin; SNA, Sambucus nigra agglutinin; ST6Gal.1, β-galactoside
2,6-sialyltransferase 1; ST6Gal.2, β-galactoside 2,6-sialyltransferase 2; TPEN, NNN'N'-tetrakis-(2-pyridylmethyl)ethylendiamine |
References |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterial and methodsSupplementary DataFundingConflict of interest statementReferences |
|---|
Akamatsu S, Yazawa S, Zenita K, Matsumoto H, Tachikawa T, Kannagi R. Elevation of an alpha(1,3)fucosyltransferase activity correlated with apoptosis in the human colon adenocarcinoma cell line, HT-29. Glycoconj J (1996) 13:1021–1029.[CrossRef][Web of Science][Medline]
Akgul C, Moulding DA, Edwards SW. Molecular control of neutrophil apoptosis. FEBS Lett (2001) 487:318–322.[CrossRef][Web of Science][Medline]
Azuma Y, Kurusu Y, Sato H, Higai K, Matsumoto K. Increased expression of Lewis X and Y antigens on the cell surface and FUT 4 mRNA during granzyme B-induced Jurkat cell apoptosis. Biol Pharm Bull (2007) 30:655–660.[Medline]
Azuma Y, Taniguchi A, Matsumoto K. Decrease in cell surface sialic acid in etoposide-treated Jurkat cells and the role of cell surface sialidase. Glycoconj J (2000) 17:301–306.[CrossRef][Medline]
Batisse C, Marquet J, Greffard A, Fleury-Feith J, Jaurand MC, Pilatte Y. Lectin-based three-color flow cytometric approach for studying cell surface glycosylation changes that occur during apoptosis. Cytometry A (2004) 62:81–88.[Medline]
Bleesing JJ, Morrow MR, Uzel G, Fleisher TA. Human T cell activation induces the expression of a novel CD45 isoform that is analogous to murine B220 and is associated with altered O-glycan synthesis and onset of apoptosis. Cell Immunol (2001) 213:72–81.[CrossRef][Medline]
Brockhausen I, Lehotay M, Yang JM, Qin W, Young D, Lucien J, Coles J, Paulsen H. Glycoprotein biosynthesis in porcine aortic endothelial cells and changes in the apoptotic cell population. Glycobiology (2002) 12:33–45.
Chai F, Truong-Tran AQ, Evdokiou A, Young GP, Zalewski PD. Intracellular zinc depletion induces caspase activation and p21 Waf1/Cip1 cleavage in human epithelial cell lines. J Infect Dis (2000) 182(Suppl_1):S85–S92.[CrossRef][Web of Science][Medline]
Chandrasekaran EV, Xue J, Xia J, Locke RD, Matta KL, Neelamegham S. Reversible sialylation: Synthesis of cytidine 5'-monophospho-N-acetylneuraminic acid from cytidine 5'-monophosphate with alpha2,3-sialyl O-glycan-, glycolipid-, and macromolecule-based donors yields diverse sialylated products. Biochemistry (2008) 47:320–330.[CrossRef][Web of Science][Medline]
Chiricolo M, Malagolini N, Bonfiglioli S, Dall’Olio F. Phenotypic changes induced by expression of β-galactoside 2,6-sialyltransferase I in the human colon cancer cell line SW948. Glycobiology (2006) 16:146–154.
Dall’Olio F, Chiricolo M. Sialyltransferases in cancer. Glycoconj J (2001) 18:841–850.[CrossRef][Medline]
Dall’Olio F, Malagolini N, Chiricolo M. β-Galactoside 2,6-sialyl- transferase and the sialyl 2,6-galactosyl-linkage in tissues and cell lines. Methods Mol Biol (2006) 347:157–170.[Medline]
Dall’Olio F, Malagolini N, Serafini-Cessi F. Enhanced CMP-NeuAc:Gal β-1,4GlcNAc-R 2,6-sialyltransferase activity of human colon cancer xenografts in athymic nude mice and of xenograft-derived cell lines. Int J Cancer (1992) 50:325–330.[Medline]
Duvall E, Wyllie AH, Morris RG. Macrophage recognition of cells undergoing programmed cell death (apoptosis). Immunology (1985) 56:351–358.[Web of Science][Medline]
Fadok VA, Voelker DR, Campbell PA, Cohen JJ, Bratton DL, Henson PM. Exposure of phosphatidylserine on the surface of apoptotic lymphocytes triggers specific recognition and removal by macrophages. J Immunol (1992) 148:2207–2216.[Abstract]
Franz S, Frey B, Sheriff A, Gaipl US, Beer A, Voll RE, Kalden JR, Herrmann M. Lectins detect changes of the glycosylation status of plasma membrane constituents during late apoptosis. Cytometry A (2006) 69:230–239.[Medline]
Franz S, Herrmann K, Fuhrnrohr B, Sheriff A, Frey B, Gaipl US, Voll RE, Kalden JR, Jack HM, Herrmann M. After shrinkage apoptotic cells expose internal membrane-derived epitopes on their plasma membranes. Cell Death Differ (2007) 14:733–742.[CrossRef][Medline]
Hall SE, Savill JS, Henson PM, Haslett C. Apoptotic neutrophils are phagocytosed by fibroblasts with participation of the fibroblast vitronectin receptor and involvement of a mannose/fucose-specific lectin. J Immunol (1994) 153:3218–3227.[Abstract]
Hammarstrom S, Hammarstrom ML, Sundblad G, Arnarp J, Lonngren J. Mitogenic leukoagglutinin from Phaseolus vulgaris binds to a pentasaccharide unit in N-acetyllactosamine-type glycoprotein glycans. Proc Natl Acad Sci USA (1982) 79:1611–1615.
Hart SP, Ross JA, Ross K, Haslett C, Dransfield I. Molecular characterization of the surface of apoptotic neutrophils: Implications for functional downregulation and recognition by phagocytes. Cell Death Differ (2000) 7:493–503.[CrossRef][Web of Science][Medline]
Hennet T, Chui D, Paulson JC, Marth JD. Immune regulation by the ST6Gal sialyltransferase. Proc Natl Acad Sci USA (1998) 95:4504–4509.
Hoebeke J, Foriers A, Schreiber AB, Strosberg AD. Equilibrium and kinetic studies of the binding of Lens culinaris lectin to rabbit erythrocytes by a quantitative fluorometric method. Biochemistry (1978) 17:5000–5005.
Ito K, Miyagawa K, Matsumoto M, Yabuno S, Kawakami N, Hamaguchi T, Iizuka M, Minamiura N. Evidence for the transglycosylation of complex type oligosaccharides of glycoproteins by endo-beta-N-acetylglucosaminidase HS. Arch Biochem Biophys (2006) 454:89–99.[Medline]
Kim SM, Lee JS, Lee YH, Kim WJ, Do SI, Choo YK, Park YI. Increased alpha2,3-sialylation and hyperglycosylation of N-glycans in embryonic rat cortical neurons during camptothecin-induced apoptosis. Mol Cells (2007) 24:416–423.[Medline]
Knibbs RN, Goldstein IJ, Ratcliffe RM, Shibuya N. Characterization of the carbohydrate binding specificity of the leukoagglutinating lectin from Maackia amurensis. Comparison with other sialic acid-specific lectins. J Biol Chem (1991) 266:83–88.
Krzewinski-Recchi MA, Julien S, Juliant S, Teintenier-Lelievre M, Samyn-Petit B, Montiel MD, Mir AM, Cerutti M, Harduin-Lepers A, Delannoy P. Identification and functional expression of a second human β-galactoside 2,6-sialyltransferase, ST6Gal II. Eur J Biochem (2003) 270:950–961.[Web of Science][Medline]
Lotan R, Skutelsky E, Danon D, Sharon N. The purification, composition, and specificity of the anti-T lectin from peanut (Arachis hypogaea). J Biol Chem (1975) 250:8518–8523.
Marini M, Frabetti F, Canaider S, Dini L, Falcieri E, Poirier GG. Modulation of caspase-3 activity by zinc ions and by the cell redox state. Exp Cell Res (2001) 266:323–332.[CrossRef][Web of Science][Medline]
Morris RG, Hargreaves AD, Duvall E, Wyllie AH. Hormone-induced cell death: 2. Surface changes in thymocytes undergoing apoptosis. Am J Pathol (1984) 115:426–436.[Abstract]
Nasirikenari M, Segal BH, Ostberg JR, Urbasic A, Lau JT. Altered granulopoietic profile and exaggerated acute neutrophilic inflammation in mice with targeted deficiency in the sialyltransferase ST6Gal I. Blood (2006) 108:3397–3405.
Rapoport E, Le Pendu J. Glycosylation alterations of cells in late phase apoptosis from colon carcinomas. Glycobiology (1999) 9:1337–1345.
Rapoport EM, Sapot’ko YB, Pazynina GV, Bojenko VK, Bovin NV. Sialoside-binding macrophage lectins in phagocytosis of apoptotic bodies. Biochemistry (Mosc) (2005) 70:330–338.[CrossRef][Medline]
Russell L, Waring P, Beaver JP. Increased cell surface exposure of fucose residues is a late event in apoptosis. Biochem Biophys Res Commun (1998) 250:449–453.[CrossRef][Web of Science][Medline]
Sarter K, Mierke C, Beer A, Frey B, Fuhrnrohr BG, Schulze C, Franz S. Sweet clearance: Involvement of cell surface glycans in the recognition of apoptotic cells. Autoimmunity (2007) 40:345–348.
Shibuya N, Goldstein IJ, Broekaert WF, Nsimba-Lubaki M, Peeters B, Peumans WJ. The elderberry (Sambucus nigra L.) bark lectin recognizes the Neu5Ac( 2-6)Gal/GalNAc sequence. J Biol Chem (1987) 262:1596–1601.
Takashima S, Tsuji S, Tsujimoto M. Characterization of thesecond type of human β-galactoside 2,6-sialyltransferase (ST6Gal II), which sialylates Galβ1,4GlcNAc structures on oligosaccharides preferentially. Genomic analysis of human sialyltransferase genes. J Biol Chem (2002) 277:45719–45728.
Toma V, Zuber C, Winter HC, Goldstein IJ, Roth J. Application of a lectin from the mushroom Polysporus squamosus for the histochemical detection of the NeuAc2,6Galβ1,4Glc/GlcNAc sequence of N-linked oligosaccharides: A comparison with the Sambucus nigra lectin. Histochem Cell Biol (2001) 116:183–193.[Web of Science][Medline]
Van Damme EJ, Barre A, Rouge P, Van Leuven F, Peumans WJ. The NeuAc2,6Gal/GalNAc-binding lectin from elderberry (Sambucus nigra) bark, a type-2 ribosome-inactivating protein with an unusual specificity and structure. Eur J Biochem (1996) 235:128–137.[Web of Science][Medline]
Van Lenten L, Ashwell G. Studies on the chemical and enzymatic modification of glycoproteins. A general method for the tritiation of sialic acid-containing glycoproteins. J Biol Chem (1971) 246:1889–1894.
Vandivier RW, Henson PM, Douglas IS. Burying the dead: The impact of failed apoptotic cell removal (efferocytosis) on chronic inflammatory lung disease. Chest (2006) 129:1673–1682.
Weinstein J, Lee EU, McEntee K, Lai PH, Paulson JC. Primary structure of β-galactoside 2,6-sialyltransferase. Conversion of membrane-bound enzyme to soluble forms by cleavage of the NH2-terminal signal anchor. J Biol Chem (1987) 262:17735–17743.
Zhang S, Cai M, Zhang SW, Hu Y, Gu JX. Involvement of beta 1,4-galactosyltransferase 1 and Gal beta1
4GlcNAc groups in human hepatocarcinoma cell apoptosis. Mol Cell Biochem (2003) 243:81–86.[Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||