Glycobiology Advance Access originally published online on April 3, 2006
Glycobiology 2006 16(7):623-634; doi:10.1093/glycob/cwj110
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
N-Glycosylation of the MUC1 mucin in epithelial cells and secretions
3 Paediatric Molecular Genetics, Weatherall Institute of Molecular Medicine, University of Oxford, John Radcliffe Hospital, Oxford OX3 9DS, UK; 4 Division of Molecular Biosciences, Imperial College London, South Kensington, London SW7 2AZ, UK; 5 Center of Biochemistry, Medical Faculty and Center for Molecular Medicine Cologne, University of Cologne, Joseph-Stelzmann-Str. 52, 50931 Köln, Germany; and 6 M-SCAN Mass Spectrometry Research and Training Centre, Silwood Park, Ascot SL5 7PZ, UK
1 To whom correspondence should be addressed; e-mails: ann-harris{at}northwestern.edu; akd10{at}uni-koeln.de
2 Present address: Human Molecular Genetics Program, Children’s Memorial Research Center, Northwestern University, 2300 Children’s Plaza, Box 211, Chicago, IL 60614-3394
Received on December 14, 2005; revised on February 28, 2006; accepted on March 26, 2006
 |
Abstract |
|---|
 Top Abstract IntroductionResultsDiscussionMaterials and methodsSupplementary materialConflict of interest statementAcknowledgmentsReferences |
|---|
The MUC1 mucin is an important tumor-associated antigen that shows extensive glycosylation in vivo. The O-glycosylation of this molecule, which has been well characterized in many cell types and tissues, is important in conferring the unusual biochemical and biophysical properties on a mucin. N-Glycosylation is crucial to the folding, sorting, membrane trafficking, and secretion of many proteins. Here, we evaluated the N-glycosylation of MUC1 derived from two sources: endogenous MUC1 isolated from human milk and a recombinant epitope-tagged MUC1F overexpressed in Caco2 colon carcinoma cells. N-Glycans on purified MUC1F/MUC1 were analyzed by matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF-MS), gas chromatography-mass spectrometry (GC-MS), and CAD-ESI-MS/MS. The spectra indicate that MUC1F N-glycans have compositions consistent with high-mannose structures (Hex5-9HexNAc2) and complex/hybrid-type glycans (NeuAc0-3Fuc0-3Hex3-8HexNAc3-7). Many of the N-glycan structures are identical on MUC1F and native MUC1; however, a marked difference is seen between the N-glycans on membrane-bound and secreted forms of the native molecule.
Key words: epithelia / MUC1 / mucin / N-glycosylation
|
Introduction |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsSupplementary materialConflict of interest statementAcknowledgmentsReferences |
|---|
The biochemical and biophysical properties of mucous glycoproteins are in large part determined by the sugar structures that are attached to the tandem repeat (TR) unit, a gene-specific repetitive sequence that is rich in serine, threonine, and proline residues. The O-glycosylation of mucous glycoproteins has received significant attention, as technological advances, particularly in mass spectrometry (MS), have enabled a detailed structural characterization of O-glycans (Dell and Morris, 2001
). Some of the biological properties of mucins correlate with their O-glycosylation. Also of importance in the processing and biological properties of mucins is their N-glycosylation, and this has received much less attention. N-Glycosylation of glycoproteins plays a key role in their folding, sorting, and secretion among other functions. The MUC1 membrane-tethered mucin was the first of this group of glycoproteins to be fully characterized (Gendler et al., 1990; Lan et al., 1990; Ligtenberg et al., 1990; Wreschner et al., 1990). This molecule exists as a heterodimer or heteromultimer following co-translational cleavage (Litvinov and Hilkens, 1993; Parry et al., 2001) that may occur via an autocatalytic mechanism (Levitin et al., 2005). MUC1 subsequently reaches the cell membrane and then undergoes a series of recycling events (Litvinov and Hilkens, 1993). During recycling of the mucin through the trans-Golgi network, its sialylation increases (Litvinov and Hilkens, 1993) and further changes occur in its O-glycosylation. Specifically, core structures change from core 2 on secreted isoforms (which have only passed through the Golgi once) to sialylated core 1 glycans on the transmembrane forms (Engelmann et al., 2005). In some experimental systems, MUC1 molecules with shorter glycans recycle more rapidly than those with longer glycans (Altschuler et al., 2000), although this may not always be the case. A proportion of the membrane-bound MUC1 is released from the cell surface probably following additional cleavage events, which may be protease mediated. The cytoplasmic tail of MUC1 then re-enters the cell and mediates the activation of signaling cascades (reviewed in Hollingsworth and Swanson, 2004). The extracellular domain of MUC1, which protrudes from the cell surface by up to 200 nm (Wesseling et al., 1996), plays an important role in mediating cell adhesion and anti-adhesion. It also acts as a ligand for various molecules and contributes to the configuration of the pericellular space. Both O- and N-glycosylation likely contribute to many of these properties of MUC1, and we and others have previously investigated the O-glycosylation of MUC1 in different cell types (Hanisch et al., 1989; Lloyd et al., 1996; Burdick et al., 1997; Muller et al., 1997, 1999; Silverman et al., 2001, 2002; Muller and Hanisch, 2002). The O-glycans carried by MUC1 are different in various epithelial cell types, and the differences can be of biological significance. We have now evaluated the N-glycosylation of the MUC1 mucin glycoprotein derived from three different sources to determine the spectrum of N-glycan structures and whether different forms of MUC1 carry variant structures. MUC1 has five potential sites of N-glycosylation (Lan et al., 1990). One of these sites is located in the degenerate repeat flanking the C-terminal end of the TR domain, whereas the remaining four are C-terminal of the TR sequences. The most extensive characterization has been carried out on epitope-tagged MUC1 (MUC1F) expressed in the human colon carcinoma cell line Caco2. The N-glycans observed on MUC1 in this expression system have been compared with native MUC1 from human milk (milk fat globule membrane-associated and secreted forms). Although many of the N-glycan structures are the same on MUC1F overexpressed in the Caco2 cell line and in native MUC1, there is a striking difference between the N-glycans on membrane-bound and secreted forms of the native molecule. |
Results |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsSupplementary materialConflict of interest statementAcknowledgmentsReferences |
|---|
Analysis of recombinant MUC1 from human cell lines
Matrix-assisted laser desorption/ionization analysis of N-glycans released from epitope-tagged MUC1F produced in Caco2 cells.
Immunopurified MUC1F from clone M50 was reduced/carboxymethylated and tryptically digested to facilitate deglycosylation with PNGase F. The released glycans were separated from peptides and glycopeptides and were analyzed by matrix-assisted laser desorption/ionization (MALDI)-MS after permethylation and Sep-Pak purification (Figure 1). The spectrum indicates that MUC1F N-glycans have compositions consistent with high-mannose structures (Hex5-9HexNAc2) and complex/hybrid-type glycans (NeuAc0-3Fuc0-3Hex3-8HexNAc3-7). A second sample (clone M51) was also analyzed, and the compositions of the glycans were very similar (data not shown).
|
Linkage analysis of PNGase F-released glycans.
Data are summarized in Table I and Supplementary Table IA. The complexity of the mixture precludes the allocation of individual components to particular oligosaccharides. Nevertheless, several important conclusions may be drawn from these data: (1) the presence of 3,6-linked Man and 4-linked GlcNAc is in accordance with their being essential constituents of the core of N-glycans; (2) the presence of terminal GlcNAc and 3,4,6-linked Man indicates that some cores have a bisecting GlcNAc; (3) terminal mannose is in accordance with the high-mannose and less abundant hybrid-type glycans; (4) fucose and galactose are the other major terminal sugars; (5) the presence of 2-linked Man is indicative of bi-antennary structures, whereas the 2,4-linked Man and 2,6-linked Man indicate tri- and tetra-antennary structures; and (6) the 4,6-linked GlcNAc residue is consistent with the presence of fucosylated cores.
|
ß-Galactosidase treatment of N-glycans.
To define the anomeric configuration as well as to confirm tentative sequences, we subjected N-glycans released by PNGase F to ß-galactosidase treatment followed by permethylation and analysis by MALDI-MS (Figure 2). The glycans that were previously observed at m/z 2040, 2489, 2605, 2663, 2693, 2850, 2938, 3054, 3112, 3228, 3286, 3299, 3415, 3473, and 3660 in the undigested sample (Figure 1) were significantly reduced or completely absent after ß-galactosidase treatment (Figure 2). This implies that these glycans have a terminal galactose residue in the ß-configuration that is not involved in a Lex/a structure, because the enzyme does not cleave if the subterminal GlcNAc is also linked to a fucose (Khoo et al., 2001). Several major ions in the undigested sample (m/z 2489, 2605, and 2850) were shifted to m/z 2081 (FucHex3HexNAc5), m/z 2401 (NeuAcFucHex4HexNAc4), and m/z 2646 (NeuAcFucHex4HexNAc5), respectively, consistent with them being bi-antennary complex structures with fucosylation predominantly on the core.
|
The ß-galactosidase treatment also revealed the presence of structures with more than two antennae. For example, the ion at m/z 2938 (FucHex6HexNAc6) has shifted to m/z 2326 (FucHex3HexNAc6), which reflects the loss of three galactoses and accordingly is very likely to be from a tri-antennary structure. Similarly, the glycans at m/z 3299 and 3660 (NeuAc1-2FucHex6HexNAc6) have shifted to m/z 2891 (NeuAcFucHex4HexNAc6) and 3456 (NeuAc2FucHex5HexNAc6), respectively, consistent with the loss of galactose residues from sialylated tri-antennary structures.
The ions at m/z 2779 and 3024 (NeuAcFuc2Hex5Hex NAc4 and NeuAcFuc2Hex5HexNAc5, respectively) are abundant components in the spectra of the undigested and digested N-glycans (Figures 1 and 2), suggesting that they are not substrates for ß-galactosidase. As these glycans are likely to be bi-antennary difucosylated species, with a single NeuAc residue, positioning of the second fucose is likely to form a Lex/a or sialyl Lex/a epitope. The resistance of these glycans to digestion implies that the major epitope is Lex/a, as the presence of sialyl Lex/a would leave a galactose residue on the second antenna susceptible to ß-galactosidase digestion. The unsialylated version (m/z 2663) of the glycan at m/z 3024 was susceptible to digestion and resulted in the loss of a single galactose to give the ion at m/z 2459. Difucosylated, bisected tri-antennary glycans at m/z 3112 and 3473 (NeuAc0-1Fuc2 Hex6HexNAc6) were partially susceptible to ß-galactosidase digestion to yield ions at m/z 2704 and 3269, respectively. In each case, a single Lex/a structure was predicted to be present leading to resistance to digestion at this site.
The composition of some glycans is consistent with hybrid structures, and resistance to the ß-galactosidase digest can be used to support this assignment. For example, the abundance of the ions at m/z 2360 and 2390 is not apparently reduced after ß-galactosidase digestion leading us to conclude that there are no susceptible galactose residues. The glycan must be a hybrid structure for this to be true. A similar argument can be made for the glycans at m/z 2431, 2564, and 2809.
ß-1,4-Galactosyltransferase treatment of N-glycans.
To facilitate the assignment of bisecting structures, we subjected the N-glycans to ß-1,4-galactosyltransferase treatment, followed by permethylation, purification by Sep-Pak, and analysis by MALDI-MS (Supplementary Figure 1). Owing to limiting amounts of material from clone M50, the N-glycans from clone M51 were analyzed for this experiment.
ß-1,4-Galactosyltransferase catalyses the transfer of a galactose residue onto terminal GlcNAc residues via a ß-1,4 linkage but does not modify bisecting GlcNAcs. Accordingly, glycans with a terminal GlcNAc on an antenna will become galactosylated, and the mass of the permethylated glycan will increase by 204 Da. In contrast, those glycans with a composition consistent with a terminal GlcNAc that do not get modified after digestion are likely to have bisecting GlcNAcs.
Comparison of MALDI spectra before and after treatment reveals that several components whose compositions are consistent with the presence of a terminal GlcNAc were significantly reduced (compare Figure 1 and Supplementary Figure 1). This provides evidence that the peaks at m/z 1836, 2040, and 2285 (Figure 1 for compositions) have at least one exposed nonreducing GlcNAc. An increase in the intensity of the peak at m/z 2244 supports the galactosylation of the components at m/z 1836 and 2040. Unexpectedly, the component at m/z 2431 was also absent after treatment with ß-1,4-galactosyltransferase, and the peak at m/z 2635 was more intense. Glycans with a mass of m/z 2431 are usually bi-antennary, nonbisected complex structures with terminal NeuAc and galactose residues. Such a structure would not be a substrate for ß-1,4-galactosyltransferase leading us to conclude that it may have a hybrid structure.
Apart from the glycans described above, the profile of the MALDI spectrum was not significantly altered after treatment with ß-1,4-galactosyltransferase, suggesting the presence of a significant amount of bisected GlcNAc. The presence of a peak at m/z 2693 suggests that a proportion of the glycan with m/z 2489 may have a terminal GlcNAc on one of its antennae.
Interestingly, clonal differences were observed between clones M50 and M51. Although the compositions of the glycans were essentially identical, clone M51 apparently had a less active GlcNAc transferase 3—the enzyme that catalyses the addition of the bisecting GlcNAc. Hence, the abundance of the proposed bisected series of glycans (e.g., 2489, 2850, and 3024) was lower in Supplementary Figure 1 relative to that observed in Figure 1.
Characterization of N-glycan structures by CAD-ESI-MS/MS.
Several components observed in the MALDI experiment (Figure 1) were subjected to CAD-ESI-MS/MS to assist sequence assignment. CAD-ESI-MS/MS of NeuAcFucHex5HexNAc5 (m/z 2850) produced a doubly charged fragment ion at m/z 1307 consistent with the presence of a bisecting GlcNAc residue as suggested by the galactosyltransferase experiment (Supplementary Figure 2). The fragmentation pattern in Supplementary Figure 2 also indicates the position of the fucose residue. The presence of a fragment ion at m/z 474 together with the absence of an ion at m/z 300 suggests that the majority of the fucose is on the core rather than on one of the antennae.
The difucosylated counterpart (m/z 3024) of the component at m/z 2850 was also subjected to CAD-ESI-MS/MS to gain information on whether the second fucose residue was present as a Lex/a or sialyl Lex/a epitope (Figure 3A). The fragmentation of the NeuAcFuc2Hex5HexNAc5 (m/z 3024) yielded a complex set of fragment ions consistent with the presence of both Lex/a and sialyl Lex/a structures (see assignments in Figure 3A). The relatively intense signals at m/z 472, 660, 847, and 1111 indicate that the predominant epitope is Lex/a, whereas the lower intensity of the ions at m/z 486 and 1024 confirm the presence of sialyl Lex/a albeit at lower levels. In a previous analysis of the O-glycosylation of MUC1F using antibodies against known carbohydrate epitopes (Silverman et al., 2001), we detected sialyl Lea; hence, it is likely that this structure also exists on MUC1F N-glycans. The fragment ion at m/z 474 confirms core fucosylation of this glycan (Figure 3A).
|
, galactose; 
, mannose; 
, GlcNAc; 
, fucose; 
, NeuAc.
Analysis of endogenous MUC1 from human milk
MALDI-MS analysis of N-glycans released from endogenous MUC1 samples isolated from skimmed milk or human milk fat membranes.
N-Linked glycans on mem-MUC1 and sec-MUC1 were released after trypsin digestion of the protein by deglycosylation with PNGase F. The released glycans were separated from peptides and glycopeptides by solid-phase extraction on C18-reversed-phase columns, and the permethylated glycans were analyzed by reflectron MALDI-MS (Figure 4A and B). The spectra show distinct N-glycosylation patterns in the samples from mem-MUC1 and sec-MUC1 with only minor overlaps. Mem-MUC1-derived glycans (Figure 4A) have compositions consistent with predominating neutral complex-type chains of bi- and tri-antennary structures and one to three fucose residues (NeuAc0-1Fuc0-3Hex4-7HexNAc3-6). The membrane-derived mucin is apparently devoid of high-mannose structures as indicated by the absence of glycans with molecular masses consistent with the presence of only two HexNAc residues. Ions registered at m/z 1621, 1826, and 1999 may represent hybrid-type structures according to their monosaccharide compositions. In contrast, the majority of the N-glycans in the sec-MUC1 sample (Figure 4B) have masses that are compatible with high-mannose structures (Hex5-8HexNAc2), whereas the remaining masses correspond to compositions of acidic complex-type glycans of bi-antennary structure (NeuAc1-2Fuc0-2Hex5HexNAc4).
|
Linkage analysis of PNGase F-released glycans.
Linkage analysis of N-glycans was performed by gas chromatography-MS (GC-MS) after permethylation and generation of the partially methylated alditol acetates. The results are summarized in Table I and Supplementary Table IB. The general core structure of N-glycans was indicated by the presence of 3,6-linked Man and 4-linked GlcNAc. Terminal mannose was found in the sec-MUC1 sample, indicating the presence of high-mannose structures. Trace amounts of terminal mannose may be derived from hybrid-type N-glycans in the mem-MUC1 sample. Terminal fucose and galactose was found in both samples. The presence of 2-linked Man is indicative of bi-antennary structures in both samples, which is in accordance with the MALDI-MS data. Tri-antennary structures are indicated by 2,4-linked Man in the mem-MUC1 sample, which is absent in the sec-MUC1-derived glycans. The 4,6-linked GlcNAc was found in both samples and is consistent with the presence of fucosylated cores.
Quantitative profiling by normal-phase high-performance liquid chromatography of desialylated 2-aminobenzamide-labelled N-glycans.
To obtain quantitatively reliable data, we treated the native glycans with neuraminidase from Clostridium perfringens, labeled with 2-aminobenzamide and chromatographed on amino-phase high-performance liquid chromatography (HPLC). Identification was based on comparison of their retention times with those of authentic standards, and major components were also identified by MALDI-MS. The profile shown in Figure 5 corresponds to N-glycans derived from mem-MUC1 (upper panel). For comparison, the profiles of asialo fetuin and an oligo-glucose standard are shown (lower panels) to standardize the retention times of A2G2 and A3G3 in glucose units (GU). The mem-MUC1 sample revealed a complex pattern, which was dominated by signals isographic with bi- and tri-antennary complex-type glycans, the fucosylated species being prevalent (8.8 GU : A2G2F). The major signals corresponding to 8.4 GU, 8.8 GU, and 9.7 GU were analyzed by MALDI-MS and revealed molecular masses [M + Na] at m/z 1784, 1930, 2149, and 2295, corresponding to permethylated glycans with m/z of 2070, 2244, 2520, and 2694, respectively (Figure 6). The chromatographic data are in good accordance with the patterns revealed by MALDI-MS of permethylated glycans and confirm that the major species in the mem-MUC1 sample should be identical to the core-fucosylated bi-antennary complex-type glycan A2G2F (m/z 2244 in Figure 6). Minor glycans with GU > 10.5 may represent higher fucosylated species in accordance with the MALDI-MS data (Figure 6).
|
|
, galactose or mannose;
CAD-ESI-MS/MS analysis of major glycans derived from mem-MUC1 and sec-MUC1.
Several of the major components observed in the MALDI-MS experiment (Figure 4A and B) were subjected to CAD-ESI-MS/MS to obtain information on terminal sequences (Figure 3B and C and Supplementary Figure 3A and B, Supplementary Table II). The spectrum of FucHex5HexNAc4 (M + Na at m/z 2244), the dominant N-glycan species in the fraction derived from mem-MUC1, revealed Bi+ fragment ions of the protonated molecular ion at m/z 189 and 464 (and 432 corresponding to the loss of methanol), indicating terminal Fuc and Hex1-4HexNAc, respectively (Supplementary Figure 3A). Bi+, Yi+, Ci+, and Zi+ ions correspond to fragments of the nonreducing (Bi+ and Ci+) or reducing ends (Yi+ and Zi+) generated by fission of the glycosidic bonds at C1-O (Bi+ and Yi+) or O-Cn (Ci+ and Zi+) (Domon and Costello, 1988). The fragment spectrum is consistent with the core-fucosylated (absence of fragment at m/z 638) bi-antennary complex-type glycan A2G2F. The second major species from mem-MUC1, Fuc2Hex5HexNAc4 (M + Na at m/z 2418), revealed in addition the Bi+ series ion at m/z 638 as primary fragment of the protonated molecular ion, indicating that one fucose is linked to a terminal Hex1-4HexNAc unit and forms a Lex epitope (absence of 2-linked Gal in linkage analysis) (Supplementary Table II). A sialylated derivative of the major core-fucosylated bi-antennary complex-type glycan A2G2F with the composition NeuAcFucHex5HexNAc4 (M + Na at m/z 2605) produced Bi+ fragment ion series at m/z 376 (344) and 825 (793), indicating that one of the antennae has the structure NeuAc-Hex1-4HexNAc (absence of the ion at m/z 228) (Figure 3B, Supplementary Table II).
In the fraction of methylated glycans derived from sec-MUC1, the major species Hex5HexNAc2 (M + Na at m/z 1580) revealed only weak fragment ions at m/z 219 (187), indicating terminal Hex and the B4 ion at m/z 1280 (Figure 3C). This is in accordance with high-mannose glycans. The presence of HexNAc in the antennae is always indicated by strong Bi+ ions as shown by one of the complex-type glycans in this fraction, NeuAcFuc2Hex5HexNAc4 (M + Na at m/z 2779), which yielded terminal fragment ion series for M + H at m/z 189, 638 (FucHexHexNAc), at m/z 376, 825 (NeuAcHexHexNAc), and at m/z 464 (HexHexNAc) (Supplementary Figure 3B). Because the sec-MUC-derived fraction contains only bi-antennary complex-type glycans according to the linkage analysis (absence of 2,4-linked mannose) and the composition of glycans, the ion at m/z 464 cannot be interpreted as a primary fragment.
|
Discussion |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsSupplementary materialConflict of interest statementAcknowledgmentsReferences |
|---|
The MUC1 mucous glycoprotein is an important tumor-associated antigen that is currently the focus of extensive translational research for cancer diagnosis and treatment. Hence, it is important to fully understand the structure and molecular biochemistry of the molecule. The proteolytic processing (Ligtenberg et al., 1992
; Parry et al., 2001) and the O-glycosylation of MUC1 (Hanisch et al., 1989; Burdick et al., 1997; Muller et al., 1997; Reid et al., 1999; Silverman et al., 2001, 2002; Muller and Hanisch, 2002) have been investigated in depth, but no studies have defined the structures of the N-glycans present on this molecule. Here, we describe structural studies on the N-glycans carried on an epitope-tagged MUC1 mucin expressed in a colon carcinoma cell line and those found on endogenous MUC1 from human milk secretions.
N-Linked core glycosylation is the first event in the addition of sugar moieties to protein backbones that occurs in the lumen of the endoplasmic reticulum (ER) after protein synthesis. The oligosaccharide Glc3Man9GlcNAc2 is transferred to an acceptor asparagine residue in the tripeptide Asn-Xaa-Ser/Thr. The initial mannose-rich N-glycans are subsequently trimmed by mannosidases in the cis-Golgi and further modified by another mannosidase and a series of glycosidases/glycosyltransferases in the medial and trans-Golgi. N-Glycosylation is crucial for the function, stability, folding, transport, and secretion of glycoproteins. Of particular relevance to the biology of MUC1 is the role of N-glycosylation in the sorting and apical expression of glycoproteins in polarized cells. This has been studied extensively for sucrase-isomaltase, aminopeptidase N, and intestinal dipeptidyl peptidase IV (DPPIV) (Naim et al., 1999; Alfalah et al., 2002). The processes of N- and O-glycosylation were shown to be temporarily associated, because O-glycosylation of sucrase-isomaltase and DPPIV was drastically reduced when the processing of the mannose-rich N-linked glycans was blocked. Incomplete O-glycan processing subsequently resulted in an alteration of the polarized sorting of the two proteins. MUC1 is normally expressed on the apical surface of differentiated epithelial cells, although its localization may be aberrant in tumor cells, and this may be important in tumor invasion and metastasis (reviewed in Hollingsworth and Swanson, 2004).
The structures of the N-glycans derived from the epitope-tagged MUC1 produced in Caco2 cells were a mixture of high-mannose, complex, and hybrid types (Figure 6 and Supplementary Figure 4). The extensive range of structures identified may in part be because of the material being purified from whole cells rather than secreted or membrane-bound forms being isolated separately. This was necessary to obtain sufficient material for immunoprecipitation and subsequent MS. The fact that the epitope-tagged MUC1 was expressed in a colon carcinoma cell line (Caco2) may also have influenced the N-glycan structures present. Also, endogenous MUC1 was produced under strong hormonal stimuli in the mammary gland, which may influence the rate of synthesis and the glycosylation. Further organ and/or cell type-specific variation cannot be excluded. However, many of the same structures were seen on endogenous MUC1 derived from human milk, although the spectra for mem-MUC1 (membrane-derived) and sec-MUC1 (secreted) showed distinct N-glycosylation patterns with only minor overlaps. Mem-MUC1-derived glycans had compositions, linkage patterns, and terminal sequences consistent with predominating neutral complex-type chains of bi- and tri-antennary structure and one to three fucose residues and lacked high-mannose structures. In contrast, the majority of the N-glycans in the sec-MUC1 sample revealed structural data that are compatible with high-mannose structures, whereas the remaining species apparently correspond to acidic complex-type glycans of bi-antennary structure. No bisected glycans were detected in the milk-derived fragments, because neither 3,4,6-linked mannose nor terminal GlcNAc was found in the linkage analysis.
Many secreted glycoproteins express both high-mannose- and complex-type N-glycans, which reflect the different stages of N-glycan processing in the ER and Golgi. However, membrane-tethered glycoproteins, such as MUC1, re-enter the cell by endocytosis after exposure at the cell surface and recycle through the trans-Golgi network (TGN) and possibly even earlier compartments as shown for other membrane-tethered glycoproteins (Volz et al., 1995). This recycling enables these glycoproteins to pass through the compartments of the Golgi and trans-Golgi network several times facilitating the processing of residual high-mannose glycans. These considerations could explain the distinct N-glycosylation profiles of secreted and transmembrane glycoproteins seen here. The same hypothesis was proposed to account for recent observations that transmembrane forms of recombinant MUC1 express distinct O-glycan core profiles compared with their secretory isoforms (Engelmann et al., 2005). The membrane-tethered isoforms were shown to recycle through the Golgi/TGN compartments and to undergo further O-glycosylation. The shift from core 2 glycans on secretory isoforms to core 1 on membrane isoforms may result from the addition of new glycan chains in the late Golgi compartments or from the reconstruction of core 2 glycans in early Golgi compartments involving the activity of glycosidases and glycosyltransferases similar to the processing of N-linked chains. Further experiments with transmembrane forms of recombinant MUC1 and mutant derivatives that are not able to recycle via clathrin-mediated endocytosis are required. These might reveal that recycling through Golgi compartments could actually form the mechanistic basis of the observed structural differences in MUC1 N-glycosylation between secretory and membrane-tethered forms.
There are many similarities between the N-glycan profiles observed for MUC1F purified from the Caco2 cell line and the MUC16 (CA125) mucin secreted from the epithelial ovarian tumor cell line OVCAR-3 (Kui Wong et al., 2003). Both mucins have relatively high levels of monofucosylated bi-antennary structures (m/z 2244, 2285, 2489, and 2850) and high molecular mass compounds consistent with tri- and tetra-antennary N-glycans (Supplementary Figure 4). Fucose is present on the core of the glycans on MUC1F and MUC16, and in some cases, a second fucose can be observed on the antenna. There are also significant amounts of high-mannose glycans with the compositions Man5-9GlcNAc2 in both mucins. In contrast to these similarities, MUC1F has higher levels of sialylation than MUC16 with up to three sialic acids present in the glycan at m/z 4399; however, this may simply reflect specific sialidase or sialyltransferase activities in the Caco2 and OVCAR cell lines.
The similarity of the N-glycan profile from MUC1F and MUC16 is of interest, as these mucins were purified from different cell lines. Previously, we have shown that MUC1F expressed in Caco2 and the airway cell line 16HBE14o– has different O-glycans (Silverman et al., 2002), but the N-glycans were not examined. Further experimentation is required to determine the biological significance of the similarities in N-glycan profiles between MUC1F and MUC16.
|
Materials and methods |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsSupplementary materialConflict of interest statementAcknowledgmentsReferences |
|---|
Cell culture and immunopurification of epitope-tagged MUC1F
The generation of the epitope (FLAG)-tagged MUC1F was described previously (Burdick et al., 1997
). Briefly, the MUC1F construct is derived from a pancreatic MUC1 cDNA with 42 TRs and contains a FLAG epitope adjacent to the TR region. Cell clones M50 and M51 are Caco2 colon carcinoma-derived cells that were stably transfected with MUC1F characterized previously (Reid et al., 1999). Cell lysates were prepared from 4-day post-confluent cultures of clones M50 and M51 and MUC1F immunopurified using M2-agarose beads as described previously (Silverman et al., 2001). MUC1F was concentrated and purified by centricon centrifugation and lyophilized.
Isolation of native mucin samples
Two endogenous MUC1 samples were isolated from human sources: mucin secreted into the fraction of soluble milk proteins (sec-MUC1) and mucin associated with the milk fat globule membranes (mem-MUC1). A detailed description of mem-MUC1 purification from human milk fat globule membranes was published in Muller et al. (1997). Sec-MUC1 was isolated from skimmed milk according to Hanisch et al. (1985).
Tryptic digestion
Lyophilized MUC1F or endogenous MUC1 samples were incubated with trypsin (Sigma, Poole, UK; or Promega, Mannheim, Germany; sequencing grade) at a 50:1 ratio (w/w) in 50 mM ammonium bicarbonate, pH 8.5, for 16 h at 37°C. The digestion was terminated by incubation at 100°C for 3 min, followed by lyophilization or vacuum centrifugation.
PNGase F digestion of tryptic glycopeptides
Digestions were carried out in 50 mM ammonium bicarbonate, pH 8.5, for 16 h at 37°C with 3 U of recombinantly expressed Flavobacterium PNGase F (Roche Molecular Biochemicals, Lewes, UK, and Roche, Mannheim, Germany) for MUC1F and 10 U of enzyme for other mucin samples. The reaction was terminated by lyophilization or vacuum centrifugation, and the released N-glycans were separated from peptides and O-glycopeptides by solid-phase extraction on Sep-Pak C18 (Waters, Elstree, UK) or Bakerbond SPE-C18 (Mallinckrodt Baker, Deventer, The Netherlands) as described (Dell et al., 1993).
ß-Galactosidase digestion
Glycans were incubated with 10 mU of ß-galactosidase (bovine testes; Sigma) in 100 µL of 50 mM ammonium acetate buffer, pH 4.6, at 37°C for 48 h with a fresh aliquot of enzyme added after 24 h.
ß-1,4-Galactosyltransferase treatment
Lyophilized N-glycans were incubated with 45 µM Uridine diphospho-galactose, 20 mM manganese chloride, 50 mM 3-(N-Morpholino)-propanesulfonic acid, pH 7.4, and ß-1,4-galactosyltransferase (bovine recombinant; Calbiochem, Nottingham, UK) for 18 h at 37°C.
Chemical derivatization of glycans for MALDI-TOF-MS, GC-MS, and CAD-ESI-MS/MS analysis
The permethylation of glycans from endogenous MUC1 samples was performed using the sodium hydroxide procedure of Ciacanu and Kerek as modified by Anumula and Taylor (1992), whereas the permethylation of N-glycans from recombinant MUC1F was performed using the sodium hydroxide procedure as described by Dell and colleagues (1993). Partially methylated alditol acetates were prepared from permethylated samples by the method of Albersheim et al. (1967).
2. -Aminobenzamide labeling and normal-phase chromatography of MUC1 from ascites and milk
Quantitative N-glycan profiling was described previously (Wuttke et al., 2001; Hambrock et al., 2004; Kaufmann et al., 2004).
MALDI-TOF mass spectrometric analysis of permethylated glycans
For MUC1F MALDI time-of-flight MS (MALDI-TOF-MS), data were acquired by using a Voyager-DE sSTR mass spectrometer (PerSeptive Biosystems, Framingham, MA) in the reflectron mode with delayed extraction. Permethylated samples were dissolved in 10 µL of 80% (v/v) methanol in water, and 1 µL of dissolved sample was premixed with 1 µL of matrix (2,5-dihydroxybenzoic acid) before loading onto a metal plate. MALDI-TOF-MS analysis of native MUC1 samples was performed similarly by using a Reflex IV instrument (Bruker Daltonic, Bremen, Germany) (Muller and Hanisch, 2002).
GC-MS of partially methylated alditols
For MUC1F, the linkage analysis of partially methylated alditol acetates was carried out on a Perkin-Elmer Clarus 500 instrument fitted with a RTX-5 fused silica capillary column (30 m x 0.32 mm internal diameter; Restek, Bad Homburg, Germany). The sample was dissolved in hexane and injected into the column at 65°C. The column was maintained at this temperature for 1 min and then heated to 290°C at a rate of 8°C per minute. For milk mucins, the analysis was performed on an MD800 (Thermo Electron, Dreieich, Germany) using a temperature gradient that started at 60°C followed by heating the column (15 m RTX5-SILMS column from Restek to 100°C at 40°C/min and then to 280°C with 10°C/min.
Collisionally activated dissociation electrospray tandem mass spectrometry of permethylated glycans
Collisionally activated dissociation electrospray tandem MS (CAD-ESI-MS/MS) spectra were acquired using a Q-TOF or Q-TOF2 (Micromass, Manchester, UK) instrument. The permethylated glycans were dissolved in methanol or in 80% (v/v) aqueous methanol with 1% acetic acid before loading into a nanospray capillary coated with a thin layer of gold/palladium, inner diameter 2 µm (Proxeon, Odense, Denmark). A potential of 1.5 kV (Q-TOF) or 0.8 kV (Q-TOF2) was applied to a nanoflow tip to produce a flow rate of 10–30 nL/min. The drying gas used was N2, and the collision gas was argon. The cone voltage was set at 50 V. Collision energies varied depending on the size of the carbohydrate and the type of pseudomolecular ion (H+ or Na+ adducts) but were typically between 30 and 90 eV.
|
Supplementary material |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsSupplementary materialConflict of interest statementAcknowledgmentsReferences |
|---|
Supplementary data are available at Glycobiology online (http://glycob.oxfordjournals.org/).
|
Conflict of interest statement |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsSupplementary materialConflict of interest statementAcknowledgmentsReferences |
|---|
None declared.
|
Acknowledgments |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsSupplementary materialConflict of interest statementAcknowledgmentsReferences |
|---|
F.G.H. thanks Mrs. Kirsten Ottenberg for skillful technical assistance and Dr. Stefan Müller of the Central Bioanalytics for ESI-MS support in Köln.
This work was supported by the Cystic Fibrosis Trust (A.H.), the Biotechnology and Biological Sciences Research Council (BBSRC), the Wellcome Trust (Bair) (A.D.), the Deutsche Forschungsgemeinschaft grant Ha 2092 /10–2, and the NIH grant 1RO1 CA84106 (F.G.H.). A.D. is a BBSRC Professorial Research Fellow.
|
Abbreviations |
|---|
CAD-ESI-MS/MS, collisionally activated dissociation electrospray tandem mass spectrometry; GC-MS, gas chromatography-mass spectrometry; GU, glucose units; HPLC, high-performance liquid chromatography; MALDI-TOF, matrix-assisted laser desorption ionization time-of-flight; MS, mass spectrometry; TR, tandem repeat
|
References |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsSupplementary materialConflict of interest statementAcknowledgmentsReferences |
|---|
Albersheim, P., Nevins, D.J., English, P.D., and Karr, A. (1967) A method for the analysis of sugars in plant cell wall polysaccharides by gas-liquid chromatography. Carbohydr. Res., 5, 340–345.[CrossRef]
Alfalah, M., Jacob, R., and Naim, H.Y. (2002) Intestinal dipeptidyl peptidase IV is efficiently sorted to the apical membrane through the concerted action of N- and O-glycans as well as association with lipid microdomains. J. Biol. Chem., 277, 10683–10690.
Altschuler, Y., Kinlough, C., Poland, P., Bruns, J., Apodaca, G., Weisz, O., and Hughey, R. (2000) Clathrin-mediated endocytosis of MUC1 is modulated by its glycosylation state. Mol. Biol. Cell, 11, 819–831.
Anumula, K.R. and Taylor, P.B. (1992) A comprehensive procedure for preparation of partially methylated alditol acetates from glycoprotein carbohydrates. Anal. Biochem., 203, 101–108.[CrossRef][Web of Science][Medline]
Burdick, M.D., Harris, A., Reid, C.J., Iwamura, T., and Hollingsworth, M.A. (1997) Oligosaccharides expressed on MUC1 produced by pancreatic and colon tumor cell lines. J. Biol. Chem., 272, 24198–24202.
Dell, A., Khoo, K.-H., Panico, M., McDowell, R.A., Etienne, A.T., Reason, A.J., and Morris, H.R. (1993) FAB-MS and ES-MS of glycoproteins. In Fukuda, M. and Kobata, A. (eds.), Glycobiology: A Practical Approach, Oxford University Press, Oxford, pp. 187–222.
Dell, A. and Morris, H.R. (2001) Glycoprotein structure determination by mass spectrometry. Science, 291, 2351–2356.
Domon, B. and Costello, C.E. (1988) A systematic nomenclature for carbohydrate fragmentations in FAB-MS/MS spectra of glycoconjugates. Glycoconj. J., 5, 397–409.[CrossRef][Web of Science]
Engelmann, K., Kinlough, C.L., Muller, S., Razawi, H., Baldus, S.E., Hughey, R.P., and Hanisch, F.-G. (2005) Transmembrane and secreted MUC1 probes show trafficking-dependent changes in O-glycan core profiles. Glycobiology, 15, 1111–1124.
Gendler, S.J., Lancaster, C.A., Taylor, P.J., Duhig, T., Peat, N., Burchell, J., Pemberton, L., Lalani, E.N. and Wilson, D. (1990) Molecular cloning and expression of human tumor-associated polymorphic epithelial mucin. J. Biol. Chem., 265, 15286–15293.
Hambrock, H.O., Kaufmann, B., Muller, S., Hanisch, F.G., Nose, K., Paulsson, M., Maurer, P., and Hartmann, U. (2004) Structural characterization of TSC-36/Flik: analysis of two charge isoforms. J. Biol. Chem., 279, 11727–11735.
Hanisch, F.G., Uhlenbruck, G., Dienst, C., Stottrop, M., and Hippauf, E. (1985) Ca 125 and Ca 19–9: two cancer-associated sialylsaccharide antigens on a mucus glycoprotein from human milk. Eur. J. Biochem., 149, 323–330.[Web of Science][Medline]
Hanisch, F.G., Uhlenbruck, G., Peter-Katalinic, J., Egge, H., Dabrowski, J., and Dabrowski, U. (1989) Structures of neutral O-linked polylactosaminoglycans on human skim milk mucins. A novel type of linearly extended poly-N-acetyllactosamine backbones with Gal beta(1–4)GlcNAc beta(1–6) repeating units. J. Biol. Chem., 264, 872–883.
Hollingsworth, M.A. and Swanson, B. (2004) Mucins in cancer: protection and control of the cell surface. Nat. Rev. Cancer, 4, 45–60.[CrossRef][Web of Science][Medline]
Kaufmann, B., Muller, S., Hanisch, F.G., Hartmann, U., Paulsson, M., Maurer, P., and Zaucke, F. (2004) Structural variability of BM-40/SPARC/osteonectin glycosylation: implications for collagen affinity. Glycobiology, 14, 609–619.
Khoo, K.H., Huang, H.H., and Lee, K.M. (2001) Characteristic structural features of schistosome cercarial N-glycans: expression of Lewis X and core xylosylation. Glycobiology, 11, 149–163.
Kui Wong, N., Easton, R.L., Panico, M., Sutton-Smith, M., Morrison, J.C., Lattanzio, F.A., Morris, H.R., Clark, G.F., Dell, A., and Patankar, M.S. (2003) Characterization of the oligosaccharides associated with the human ovarian tumor marker CA125. J. Biol. Chem., 278, 28619–28634.
Lan, M.S., Batra, S.K., Qi, W.N., Metzgar, R.S., and Hollingsworth, M.A. (1990) Cloning and sequencing of a human pancreatic tumor mucin cDNA. J. Biol. Chem., 265, 15294–15299.
Levitin, F., Stern, O., Weiss, M., Gil-Henn, C., Ziv, R., Prokocimer, Z., Smorodinsky, N.I., Rubinstein, D.B., and Wreschner, D.H. (2005) The MUC1 SEA module is a self-cleaving domain. J. Biol. Chem., 280, 33374–33386.
Ligtenberg, M.J., Kruijshaar, L., van Buijs, F.M.M., Litvinov, S.V., and Hilkens, J. (1992) Cell-associated episialin is a complex containing two proteins derived from a common precursor. J. Biol. Chem., 267, 6171–6177.
Ligtenberg, M.J., Vos, H.L., Gennissen, A.M., and Hilkens, J. (1990) Episialin, a carcinoma-associated mucin, is generated by a polymorphic gene encoding splice variants with alternative amino termini, J. Biol. Chem., 265, 5573–5578.
Litvinov, S.V. and Hilkens, J. (1993) The epithelial sialomucin, episialin, is sialylated during recycling. J. Biol. Chem., 268, 21364–21371.
Lloyd, K.O., Burchell, J., Kudryashov, V., Yin, B.W.T., and Taylor-Papadimitriou, J. (1996) Comparison of O-linked carbohydrate chains in MUC-1 mucin from normal breast epithelial cell lines and breast carcinoma cell lines. J. Biol. Chem., 271, 33325–33334.
Muller, S., Alving, K., Peter Katalinic, J., Zachara, N., Gooley, A.A., and Hanisch, F.G. (1999) High density O-glycosylation on tandem repeat peptide from secretory MUC1 of T47D breast cancer cells. J. Biol. Chem., 274, 18165–18172.
Muller, S., Goletz, S., Packer, N., Gooley, A., Lawson, A.M., and Hanisch, F.G. (1997) Localization of O-glycosylation sites on glycopeptide fragments from lactation-associated MUC1. All putative sites within the tandem repeat are glycosylation targets in vivo. J. Biol. Chem., 272, 24780–24793.
Muller, S. and Hanisch, F.G. (2002) Recombinant MUC1 probe authentically reflects cell-specific O-glycosylation profiles of endogenous breast cancer mucin. High density and prevalent core 2-based glycosylation. J. Biol. Chem. 277, 26103–26112.
Naim, H.Y., Joberty, G., Alfalah, M., and Jacob, R. (1999) Temporal association of the N- and O-linked glycosylation events and their implication in the polarized sorting of intestinal brush border sucrase-isomaltase, aminopeptidase N, and dipeptidyl peptidase IV. J. Biol. Chem., 274, 17961–17967.
Parry, S., Silverman, H.S., McDermott, K., Willis, A., Hollingsworth, M.A., and Harris, A. (2001) Identification of MUC1 proteolytic cleavage sites in vivo. Biochem. Biophys. Res. Commun., 283, 715–720.[CrossRef][Web of Science][Medline]
Reid, C.J., Burdick, M.D., Hollingsworth, M.A., and Harris, A. (1999) CFTR expression does not influence glycosylation of an epitope-tagged MUC1 mucin in colon carcinoma cell lines. Glycobiology, 9, 389–398.
Silverman, H.S., Parry, S., Sutton-Smith, M., Burdick, M.D., McDermott, K., Reid, C.J., Batra, S.K., Morris, H.R., Hollingsworth, M.A., Dell, A., and Harris, A. (2001) In vivo glycosylation of mucin tandem repeats. Glycobiology, 11, 459–471.
Silverman, H.S., Sutton-Smith, M., Heal, P., Parry, S., Palmai-Pallag, T., Leir, S.H., Morris, H.R., Dell, A., and Harris, A. (2002) In vivo glycosylation of MUC1 in airway epithelial cells. Glycoconj. J., 19, 379–384.[CrossRef][Web of Science][Medline]
Volz, B., Orberger, G., Porwoll, S., Hauri, H.P., and Tauber, R. (1995) Selective reentry of recycling cell surface glycoproteins to the biosynthetic pathway in human hepatocarcinoma HepG2 cells. J. Cell Biol., 130, 537–551.
Wesseling, J., van der Valk, S.W., and Hilkens, J. (1996) A mechanism for inhibition of E-cadherin-mediated cell-cell adhesion by the membrane-associated mucin episialin/MUC1. Mol. Biol. Cell, 7, 565–577.[Abstract]
Wreschner, D.H., Hareuveni, M., Tsarfaty, I., Smorodinsky, N., Horev, J., Zaretsky, J., Kotkes, P., Weiss, M., Lathe, R., Dion, A., and others (1990) Human epithelial tumor antigen cDNA sequences. Differential splicing may generate multiple protein forms. Eur. J. Biochem. 189, 463–473.[Web of Science][Medline]
Wuttke, M., Muller, S., Nitsche, D.P., Paulsson, M., Hanisch, F.G., and Maurer, P. (2001) Structural characterization of human recombinant and bone-derived bone sialoprotein. Functional implications for cell attachment and hydroxyapatite binding. J. Biol. Chem., 276, 36839–36848.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  Y. He, Y. Li, Z. Peng, H. Yu, X. Zhang, L. Chen, Q. Ji, W. Chen, and R. Wang Role of N-glycosylation of the SEA module of rodent Muc3 in posttranslational processing of its carboxy-terminal domain Glycobiology, October 1, 2009; 19(10): 1094 - 1102. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Yue, I. J. Goldstein, M. A. Hollingsworth, K. Kaul, R. E. Brand, and B. B. Haab The Prevalence and Nature of Glycan Alterations on Specific Proteins in Pancreatic Cancer Patients Revealed Using Antibody-Lectin Sandwich Arrays Mol. Cell. Proteomics, July 1, 2009; 8(7): 1697 - 1707. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||