Glycobiology Advance Access originally published online on May 11, 2005
Glycobiology 2005 15(10):952-964; doi:10.1093/glycob/cwi075
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Identification of the hydrophobic glycoproteins of Caenorhabditis elegans
2 Research Institute, The Hospital for Sick Children, Toronto, Ontario, Canada M5G 1X8; 3 Department of Biochemistry, University of Toronto, Toronto, Ontario, Canada M5S 1A1; and 4 Department of Laboratory Medicine and Pathobiology, University of Toronto, Toronto, Ontario, Canada M5S 1A1
1 To whom correspondence should be addressed; e-mail: hex{at}sickkids.ca
Received on February 25, 2005; revised on April 25, 2005; accepted on May 6, 2005
 |
Abstract |
|---|
 Top Abstract IntroductionResultsDiscussionMaterials and methodsSupplementary dataAcknowledgmentsReferences |
|---|
Hydrophobic proteins such as integral membrane proteins are difficult to separate, and therefore to study, at a proteomics level. However, the Asn-linked (N-linked) carbohydrates (N-glycans) contained in membrane glycoproteins are important in differentiation, embryogenesis, inflammation, cancer and metastasis, and other vital cellular processes. Thus, the identification of these proteins and their sites of glycosylation in a well-characterized model organism is the first step toward understanding the mechanisms by which N-glycans and their associated proteins function in vivo. In this report, a proteomics method recently developed by our group was applied to identify 117 hydrophobic N-glycosylated proteins of Caenorhabditis elegans extracts by analysis of 195 glycopeptides containing 199 Asn-linked oligosaccharides. Most of the proteins identified are involved in cell adhesion, metabolism, or the transport of small molecules. In addition, there are 18 proteins for which no function is known or predictable by sequence homologies and two proteins which were previously predicted to exist only on the basis of genomic sequences in the C. elegans database. Because N-glycosylation is initiated in the lumen of the endoplasmic reticulum (ER), our data can be used to reassess the previously predicted subcellular localizations of these proteins. As well, the identification of N-glycosylation sites helps establish the membrane topology of the associated glycoproteins. Caenorhabditis elegans strains are presently available with mutations in 17 of the genes we have identified. The powerful genetic tools available for C. elegans can be used to make other strains with mutations in genes encoding N-glycosylated proteins and thereby determine N-glycan function.
Key words: Asn-linked oligosaccharide / multidimensional liquid chromatography / proteomics / tandem mass spectrometry
|
Introduction |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsSupplementary dataAcknowledgmentsReferences |
|---|
Glycosylation is one of the most common posttranslational protein modifications. About 0.5–1.0% of the translated mammalian genome participates in oligosaccharide production and function (Varki and Marth, 1995
). In eukaryotic cells, most proteins synthesized on ribosomes bound to the endoplasmic reticulum (ER) undergo this specific posttranslational modification. Asn-linked (N-linked) glycosylation is initiated on the lumenal side of the ER through the oligosaccharyltransferase-catalyzed transfer of the glycan moiety from a lipid-linked oligosaccharide to a target sequence contained in a nascent polypeptide chain. Thus, proteins containing N-linked oligosaccharides (N-glycans) must have contained a functional ER signal peptide, which may be removed cotranslationally (reviewed in Von Heijne, 1990). In mammalian cells, the consensus amino acid sequence needed for the transfer of N-glycans is Asn-X-Ser/Thr (where X is any amino acid except Pro or Asp). Interestingly, in Caenorhabditis elegans, this sequence may also be Asn-X-Cys (Kaji et al., 2003). This sequence is necessary but not sufficient for N-linked glycosylation to occur (reviewed in Kornfeld and Kornfeld, 1985). After synthesis, the glycoproteins are generally first transported to the Golgi complex where their carbohydrate units are altered and elaborated. The proteins are then sorted and sent to the plasma membrane, secretory granules, or lysosomes (reviewed in Helenius and Aebi, 2001). The cytoplasm, mitochondria, peroxisomes, and nucleus are therefore excluded as possible subcellular locations for N-glycosylated proteins. As well, if all or even some of the occupied Asn-X-Ser/Thr sequences are known, the membrane topology of a glycoprotein can be determined.
N-glycans play important roles in protein folding and oligomerization, in the ER-based quality control system (ERAD), in intracellular targeting (particularly to the lysosome) (Helenius and Aebi, 2001), and in carbohydrate-mediated interactions between cells and their environment, for example, differentiation, embryogenesis, inflammation, and metastatic cancer (Varki, 1993; Dennis et al., 1999a,b; Feizi, 2000; Fukuda, 2000; Lowe, 2002). N-glycans involved in the latter functions are likely to be contained in integral plasma membrane proteins.
To design experiments aimed at understanding the in vivo roles of glycoproteins and their N-glycans, we must first identify these proteins and their oligosaccharide attachment sites in an easily manipulated model organism. Caenorhabditis elegans is the most completely understood metazoan of anatomy, genetics, development, and behavior (Wood, 1988; Epstein and Shakes, 1995; Riddle et al., 1997) and is one of the easiest model organisms to manipulate genetically. Its entire genomic DNA sequence and many well-characterized mutant strains are available. The latter can be preserved and stored by freezing.
The C. elegans DNA databases contain sequences which are similar to a large number of mammalian enzymes involved in N-glycan synthesis. Fine structural analyses of N-glycans, primarily by mass spectrometry, have shown that C. elegans contains a predominance of oligomannose Man5-9GlcNAc2-Asn N-glycans identical to those found in vertebrates (Altmann et al., 2001; Cipollo et al., 2002; Haslam et al., 2002; Natsuka et al., 2002; Schachter et al., 2002; Haslam and Dell, 2003; Zhang et al., 2003; Schachter, 2004). In contrast, the complex and hybrid N-glycans that are highly abundant in vertebrates are either absent in C. elegans or present at very low levels. However, C. elegans contains large amounts of paucimannose Man3-4GlcNAc2-Asn N-glycans not usually present in vertebrates. Because oligomannose and paucimannose N-glycans bind strongly to the lectin Concanavalin A (Con A), whereas complex and hybrid N-glycans do not either bind to Con A or bind to this lectin weakly (Baenziger and Fiete, 1979; Merkle and Cummings, 1987), the capture of C. elegans N-glycans by Con A is predicted to be well over 90%. The ability to use Con A to purify the glycoproteins and/or glycopeptides of C. elegans has been used by Kaji et al. (2003) to isolate predominantly soluble glycoproteins and by our group to isolate primarily integral membrane and other hydrophobic glycoproteins (Fan et al., 2004). The identification of these proteins and their N-glycan attachment sites by mass spectrometry will allow genetic experiments to probe the specific functions of glycoproteins and/or their individual oligosaccharides in C. elegans.
In a previous report (Fan et al., 2004), we described the methodology for identifying hydrophobic glycoproteins and their sites of N-linked glycosylation. We analyzed 22 major peaks from matrix-assisted laser desorption ionization quadrupole time-of-flight mass spectrometry (MALDI-Q-TOF MS) spectra and identified 16 N-glycosylated proteins. N-glycosylation at a specific site was indicated by the presence of an Asp-X-Ser/Thr/Cys sequence at a predicted Asn-X-Ser/Thr/Cys consensus sequence. The Asn to Asp conversion results from the action of glycopeptidase F used to deglycosylate the glycopeptides after isolation by lectin chromatography. In this report, we have used our previous MALDI methodology together with two other MS approaches, (1) off-line reverse phase high performance liquid chromatography (RP-HPLC) followed by MALDI-Q-TOF analyses (LC-MALDI) and (2) on-line ion-exchange/RP-HPLC/MS (2D-LC-MS), to identify 195 glycopeptides containing 199 N-glycans in a total of 117 glycoproteins from C. elegans.
|
Results |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsSupplementary dataAcknowledgmentsReferences |
|---|
Off-line RP-HPLC to fractionate deglycosylated peptides followed by LC-MALDI resulted in the identification of 31 glycoproteins based on the sequences of 37 peptides. This group included all of the 16 proteins we have previously identified by MALDI alone (Fan et al., 2004
) (Table I, Supplementary Table I). One of the proteins identified by LC-MALDI [designated in the C. elegans database as "Putative protein, with at least 13 transmembrane domains, of eukaryotic origin" (M01F1.4)] actually has 14 predicted transmembrane domains (TMDs) and we now know it to be N-glycosylated. This protein contained the highest number of TMDs of all the proteins that we have identified. However, although it was identified by the LC-MALDI technique, it was not found by using the 2D-LC-MS approach. This could have been caused by the different ionization modes of the MALDI-MS and the electrospray ionization mass spectrometry (ESI-MS) (Krutchinsky et al., 2000; She et al., 2001) and underscores the advantage of using both techniques for protein identification. The tandem mass spectrometry (MS/MS) spectrum of the glycopeptide, that is, NN(N
D)FTVAVNQEPR, from this protein is shown in Figure 1a.
|
|
To identify the broader complement of glycoproteins expected at a proteomic scale, we used automated on-line 2D-LC-MS to analyze the deglycosylated peptide mixture. With this method, we identified 111 proteins from 185 glycopeptides; these include all the proteins identified by MALDI and all but four of the proteins identified by off-line LC-MALDI. Figure 1b shows a typical example of the MS/MS spectrum of a glycopeptide from Vacuolar proton ATPase VHA-5 (F35H10.4), identified by on-line 2D-LC-MS. A comparison of some proteins identified by the three approaches (MALDI, off-line LC-MALDI, and on-line 2D-LC-MS) is shown in Figure 2.
|
Removal of overlapping proteins from the three methods resulted in 117 distinct N-glycosylated proteins in C. elegans identified by the sequences of 195 glycopeptides, containing 199 N-glycosylation sites. Table I lists the amino acid number for each Asn in a given protein that contains an N-glycan, and Supplementary Table I lists the complete amino acid sequence of each glycopeptide. The "identity score" assigned by the Mascot program for 139 of the glycopeptides was >33, p < 0.05 (Supplementary Table I). For peptides with scores <33, the peptide sequences were also carefully interpreted manually (see Materials and methods section). Additionally, each glycopeptide was confirmed to have a predicted Asn converted to Asp in one or more consensus N-glycosylation site(s). All but one of the N-glycosylation sites that we identified corresponded to the conventional sequence Asn-X-Thr/Ser. The single exception was an Asn-X-Cys-containing peptide (Kaji et al., 2003) from C25B8.3 (Asn201). Of the glycopeptides predicted to have an N-terminal Cys residue, all four were modified to Pyro-CamC (cyclization of N-terminal carbamidomethyl-Cys), due to a common modification observed in our preliminary study (Fan et al., 2004). This modification is caused by the reduction and alkylation of the proteins before trypsin digestion.
The Mr of the identified glycoproteins range from 15 kDa (Transthyretin-like family member, Y5F2A.1) to 1370 kDa (Mesocentin, K07E12.1), with 8 (6.7%) having Mr of >200,000 (Supplementary Table I). Among the 117 proteins, 73 (62.4%) have at least one predicted TMD and 40 (34.2%) were predicted to have two or more TMDs. Among the 44 proteins with no predicted TMD, 36 were predicted to contain a signal peptide (SOSUIsignal, Beta version), which could result in a permanent TMD being generated if it were not cleaved during import into the ER. Among the eight proteins with neither a predicted signal peptide nor a TMD, four are known basement membrane proteins, that is, fibrillin, and three laminin-related proteins [in humans laminin A is known to have a short 17 amino acid signal peptide (Haaparanta et al., 1991)] (Table I). The distribution of TMDs contained within the identified proteins is shown in Figure 3.
|
Because proteins containing N-glycans are synthesized in the ER, they are limited in their possible subcellular localizations. We used these limitations to evaluate the accuracy of the PENCE Proteome Analyst program (http://www.cs.ualberta.ca/~bioinfo/PA/) in predicting subcellular locations de novo (Tables I and II). Table II compares the predicted percent distribution of the entire C. elegans proteome with our 117 glycoproteins. The program clearly mislocalized only 5% of our proteins, that is, five to the cytosol and one to the nucleus. Interestingly, this program successfully predicted that none of the proteins we identified as glycoproteins were localized to the mitochondria or peroxisome. Of the five proteins predicted to be cytosolic, three have more than two predicted TMDs and one is galectin, a known extracellular protein. As well, some proteins predicted by this program to be extracellular are actually more likely to be in the plasma membrane, based on either their known plasma membrane location or the fact that they are predicted to have more than one TMD, for example, HMR-1 (W02D8.6) and a predicted AMOP domain-containing family member (R09E10.5) (Table I).
|
Of the 117 proteins we identified, only 18 (15.4%) could not be assigned a putative function based on significant sequence homology with known proteins and/or previously identified functional domains (Table I). The remaining 99 (84.6%) proteins could be sorted into four main functional groups. The two largest groups are predicted to be involved in, or related to, cell adhesion (38.4%: cell adhesion/receptors, 26.3%, and extracellular matrix proteins, 12.1%), and those involved in metabolism (39.4%: proteolysis and/or peptidolysis, 11.1%; carbohydrate metabolism and protein glycosylation, 18.2%; and other enzymes, 10.1%). Transporters of small molecules (8.1%) and proteins with "other" miscellaneous functions (14.1%) complete the groupings (Table I).
Many proteins identified in this work are designated as hypothetical in the database, that is, proteins predicted from genomic DNA. However, cDNAs have been cloned from all but two of these proteins. A function for both these proteins can be predicted from their predicted sequences (Table I): R09E10.5 with 2 TMD is predicted to be an AMOP domain-containing protein family member and M01F12.6 with 10 TMD is predicted to be a class W protein surpentine receptor 1 like family member.
A search of WormBase has revealed that there are mutant C. elegans strains available for each of 17 (14.5%) of the 117 genes (Table I), 13 of which produce an abnormal phenotype whereas the other strains are either normal or have an as yet unidentified phenotype (Table I).
|
Discussion |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsSupplementary dataAcknowledgmentsReferences |
|---|
Owing to the large variation in the expression levels of proteins, general proteomic analyses of whole cells or tissues fail to identify low abundance proteins, for example, plasma membrane receptors. It is now recognized that some simple fractionation step(s) must be carried out before the proteomic study to enrich for lower abundance proteins (Brunet et al., 2003
). In this study, we have used two such fractionation steps: (1) the isolation of the total hydrophobic protein fraction from C. elegans and (2) the purification of a trypsin-digested glycopeptide subfraction by lectin chromatography. Thus, although the group of 117 glycoproteins we have identified may seem modest by comparisons with general proteomic analyses, the glycopeptide fraction we analyzed represented only 
0.2% of the total starting protein we obtained from sonicated worms (Fan et al., 2004).
Identifying glycoproteins in a model organism, characterizing their N-linked glycosylation site(s), and determining their subcellular localization are major steps toward understanding the in vivo roles played by N-glycans. We and others have chosen to do this in C. elegans, the most completely understood metazoan and one of the easiest model organisms to manipulate genetically. The focus of this work is the identification of the hydrophobic glycoproteins, particularly the integral membrane proteins. Kaji et al. (2003) have identified 242 primarily soluble glycoproteins in C. elegans. The largest group of proteins (49) that were found in both studies were those predicted to have either 0 or 1 TMD and likely represent relatively abundant proteins in these two categories. Together these studies have identified 304 proteins containing N-glycans.
Many of the hydrophobic proteins we have identified are important for cell adhesion or are part of the extracellular matrix (Table I). Cell adhesion is one of the defining features of multicellular organisms and underlies the organization of cells into discrete tissues and organs during metazoan development. These processes require adhesion both between neighboring cells and between cells and extracellular matrix substrates. Glycoproteins on cell surfaces are important for communication between cells (or between cells and extracellular matrix), for maintaining cell structure and for self-recognition by the immune system. In this report, we demonstrate that many of the important cell adhesion receptors in C. elegans are glycoproteins (Table I), and we identify some of their N-linked glycosylation sites. These proteins include cell–cell adhesion receptors (1) HMR-1, one of the cadherin genes which is calcium-dependent and regulates morphogenesis, pattern formation, and cell migration (Tepass, 1999); (2) LAD-1, one of the immunoglobulin superfamily cell adhesion molecules (IgCAMs); and (3) ITX-1, a neurexin family protein, and cell-extracellular matrix adhesion receptors: (1) integrins; (2) dystroglycan (dyn-1), the central component of the dystrophin-glycoprotein complex; and (3) the nematode-specific molecules, MUP-4 and myotactin.
In C. elegans, the putative cell adhesion receptors MUP-4 and myotactin have no clear nonnematode homologs and are components of fibrous organelles that are analogous to vertebrate hemidesmosomes (Cox and Hardin, 2004; Cox et al., 2004). MUP-4 is a single-span TMD receptor that localizes to the apical epidermal surface and probably binds cuticular collagens to maintain adhesion between the epidermis and cuticle (Hong et al., 2001). Its extracellular domain includes epidermal growth factor-like repeats, a von Willebrand factor A domain which mediates collagen binding in other proteins. Its cytoplasmic domain has some sequence similarity to filaggrins, which are intermediate filament-binding proteins. Myotactin (LET-805) is a large, single-span TMD protein in the basal epidermal membrane. Our data identify myotactin as a highly glycosylated protein, with at least 13 N-linked glycosylation sites (Table I). It is required for adhesion between the epidermis and muscle (Hresko et al., 1999). Its longest splice form encodes a protein with at least 32 fibronectin type III repeats in the extracellular domain and a novel cytoplasmic domain.
Integrins are the major metazoan receptors for cell adhesion to extracellular matrix proteins. In vertebrates, they also play important roles in certain cell–cell adhesions, make transmembrane connections to the cytoskeleton, and activate many intracellular signaling pathways. Integrins are 
ß heterodimeres which form multiprotein complexes (Zamir and Geiger, 2001). The genome of the nematode C. elegans encodes two subunits, PAT-2::F54F2.1 and INA-1::F54G8.3, and one ß subunit, PAT-3::ZK1058.2. The integrin formed by INA-1/ßPAT-3 is predicted to bind laminins, whereas the PAT-2/ßPAT-3 form is predicted to bind ligands with RGD motifs (the tripeptide Arg-Gly-Asp) frequently found in matrix proteins (Hutter et al., 2000; Cox et al., 2004). Glycosylation sites (five from PAT-2, one from INA-1, and two from PAT-3) were identified for these three integrin gene products in this report (Table I).
Laminins are the major noncollagenous components of basement membranes that mediate cell adhesion, growth migration, and differentiation. On the surface of cells, laminins are known to bind several receptors and receptor-like molecules, including integrins, /ß-dystroglycan, and syndecans (Colognato et al., 1999; Huang et al., 2003). Laminins are heterotrimeric (/ß/
) proteins. Analysis of the C. elegans genome has revealed the presence of two , one ß, and one subunits, a:: T22A3.8, b::K08C7.3 (epi-1), ß::w03F8.5 (lam-1), and ::C54D1.5, suggesting that only aß and bß heterotrimers are present (Hutter et al., 2000; Huang et al., 2003). The glycosylation sites of all four laminin subunit gene products were identified in this study: 5 for , 10 for b, 3 for a, and 6 for ß.
In addition to the cell adhesion receptors, we found nine basement membrane proteins (9.1%) in our identified hydrophobic protein list, including laminins, collagens, nidogen::F54F3.1 (nid-1), and fibrillin. It is not surprising that these basement membrane proteins were present since the sample we used to isolate the membrane-bound N-glycosylated proteins was the insoluble fraction from sonicated worms.
Another major class of N-glycosylated proteins identified in this study is a group of proteins involved in metabolism. Many of these are either known to be located in lysosomes, for example, ß-galactosidase protective protein, -amylase, and acid phosphatase; or involved in protein glycosylation, for example, nine UDP-glucuronosyl/UDP-glucosyl transferase family members (Table I).
If one wishes to determine the functions of glycoproteins and/or their glycans in development and morphogenesis, it is clear that an animal model such as C. elegans must be used. Although mutant strains are presently available for only 17 of the 117 genes we have identified as encoding N-glycosylated proteins (Table I), these strains were not developed with an eye toward investigating glycoprotein and/or N-glycan function. With the knowledge gained from this study, the powerful genetic tools available for C. elegans can be used to make strains with mutations in genes encoding N-glycosylated proteins and thereby to determine a wide variety of N-glycan functions, for example, targeted mutagenesis of the 21 glycoproteins we identified as likely to reside in the plasma membrane will provide information on the role of N-glycans in cell adhesion.
|
Materials and methods |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsSupplementary dataAcknowledgmentsReferences |
|---|
Sample preparation
The sample preparation was carried out, as previously described (Fan et al., 2004
). Briefly, mixed-stage worms (8 g, wet weight) were sonicated (yielding 170 mg of protein) and washed in phosphate buffered saline (PBS). The insoluble fraction (60 mg) was separated and solubilized in 6 M guanidine–HCl, reduced by dithiothreitol (DTT), and alkylated with iodoacetamide. Proteins were then precipitated with ethanol and digested with trypsin (1:100, w/w). After proteolysis, the sample was centrifuged at 110,000 x g for 1 h at 4°C, and the supernatant containing soluble tryptic peptides (40 mg) was loaded onto a 1-mL Con A column. The Con A-bound glycopeptides (0.4 mg) were eluted with 15% (w/v) methyl--d-mannoside in PBS and desalted with a C18 Sep-Pak cartridge (Waters, Milford, MA). Glycopeptidase F was then used to remove the N-glycans, and the deglycosylated peptides were analyzed by MS/MS.
MALDI-Q-TOF MS
The deglycosylated peptide mixture (10 µL) was purified by using a C18 Zip-Tip (Millipore, Bedford, MA). Then the sample (1 µL, 2 µg) was mixed with 1 µL of matrix 2,5-dihydroxybenzoic acid (DHB) and deposited on a MALDI target. Following peptide mapping, the individual deglycosylated peptide ions were selected for MS/MS analyses on a QStar XL MALDI-Q-TOF mass spectrometer (Applied Biosystems, Foster City, CA). The peptide sequences derived from MS/MS spectra were examined manually (taking into account the 1 Da change at the Asn-X-Ser/Thr site due to the AsnAsp conversion during deglycosylation). Protein identification was performed by using PeptideSearch from the EMBL Bioanalytical Research Group (http://www.mann.embl-heidelberg.de), ProteinProspector at the University of California, San Francisco (http://prospector.ucsf.edu), or Mascot (http://www.matrixscience.com).
Off-line LC-MALDI MS
The deglycosylated peptide mixture was concentrated and acidified with trifluoroacetic acid (TFA) to 0.1%, and then 5 µL of the sample (40 µg) was loaded onto a 0.5 mm x 15 cm Zorbax SB-C18 column (Agilent Technologies, Palo Alto, CA) connected to an Agilent 1100 capillary LC system. The gradients used are as follows (buffer A: 5% acetonitrile with 0.1% TFA; buffer B: 70% acetonitrile with 0.1% TFA; and flow rate: 5 µL/min): for 0–10 min, 5% B (95% A); for 10–75 min, 5–80% B; for 75–80 min, 80–95% B; for 80–85 min, 95% B; and for 85–90 min, 95–5% B. Fractions were automatically collected at 5-min intervals, and each fraction (N = 18, 25 µL each) was dried by Speed-Vac (UniEquip, Martinsried, Germany). Each fraction was dissolved in 1 µL of distilled water and mixed with 1 µL of matrix DHB and deposited on a MALDI target. LC-MALDI and protein identification were then performed in the same manner, as described above.
On-line 2D-LC-ESI MS
The deglycosylated peptide mixture was concentrated and purified by using a C18 Zip-Tip. Then 6.4 µL of the sample (130 µg) was subjected to on-line 2D-LC-ESI MS analyses. The peptide mixture was first fractionated by 12 chromatographic steps (0, 20, 40, 60, 80, 120, 140, 160, 180, 250, 350, and 1000 mM KCl in 0.1% TFA) on a Waters SCX 5 µm OPTI-PAK Trap column (0.35 x 5 mm) (Waters, Milford, MA). After desalting with a C18 column, the peptides in each ion-exchange fraction were further fractionated on a PicoFrit C18 column (10.2 cm x 75 µm) (New Objective, Woburn, MA) by using a linear gradient of 5–80% mobile phase B (A: 0.1% TFA; and B: 90% acetonitrile in 0.1% TFA) with a flow rate of 200 nL/min for 60 min. The eluted peptides were analyzed by a Q-ToF mass spectrometer (Waters, Milford, MA).
MS and MS/MS data were acquired and processed automatically by using MassLynx 3.5 software. Database searching was carried out with MS/MS ion searches of the MASCOT (http://www.matrixscience.com) and NCBInr databases. The parameters for Mascot searches were selected as follows: two missed cleavages; 0.2 Da mass tolerance allowed for both peptide and fragment ions; carbamidomethyl as fixed modification. For variable modifications, apart from oxidized methionine and N-terminal glutamine modification (PyroGlu), we also selected PyroCamC as it has recently been documented as a common modification when using iodoacetamide to alkylate proteins (Geoghegan et al., 2002; Fan et al., 2004). In addition, we created the modification entry for glycosylated asparagines (Glyc–Asn) in the Unimod (a database of protein modifications used by Mascot) so that it could also be used as one of the variable modifications, that is, the conversion of glycosylated Asn to Asp upon deglycosylation with glycopeptidase F, resulting in 1 Da mass increment. All the proteins that scored as significant database "hits" by Mascot (scores >33, p < 0.05) were verified by manual inspection of their MS/MS data; this procedure also made sure that the Glyc–Asn modification (AsnAsp) had occurred at one or more NXT/S/C sites. For peptides with scores <33, the peptide sequences were also carefully interpreted manually, and the following criteria were checked to confirm or cancel the Mascot suggestion that (1) the MS/MS spectra must be of good quality with fragment ions clearly above baseline noise; (2) there should be a continuous stretch of the peptide sequence covered by either the y- or the b-series ions; (3) the y ions that correspond to a proline residue should be intensive ions; and (4) the modified Asn (N) should be at a consensus sequence NXT/S/C.
Database searches
Predictions for membrane-spanning regions and N-terminal signal peptides were performed by using SOSUIsignal (Beta version) software (http://sosui.proteome.bio.tuat.ac.jp/sosuisignal/sosuisignal_submit.html). Subcellular locations were predicted by using PENCE Proteome Analyst (http://pagosub.cs.ualberta.ca:8080/Blast) (Lu et al., 2004). The functional classifications of the proteins were based on NCBI annotation (http://www.ncbi.nlm.nih.gov/). The genes for which mutant strains (with or without known phenotypes) are available were determined by searching WormBase (WS 130) (http://www.wormbase.org).
|
Supplementary data |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsSupplementary dataAcknowledgmentsReferences |
|---|
Supplementary data are available at Glycobiology online (http://glycob.oupjournals.org/).
|
Acknowledgments |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsSupplementary dataAcknowledgmentsReferences |
|---|
This work was funded by a PENCE (Protein Engineering Network Centres of Excellence, Canada) grant to H.S., D.M., and J.C., and an O.G.I. (Ontario Genome Initiative, Canada) grant to D.M. and J.C.
|
Abbreviations |
|---|
ConA, Concanavalin A; 2D-LC-MS, ion-exchange/RP-HPLC/MS; ER, endoplasmic reticulum; LC-MALDI, liquid chromatography-matrix assisted laser desorption ionization; MALDI-Q-TOF MS, matrix-assisted laser desorption ionization quadrupole time-of-flight mass spectrometry; MS/MS, tandem mass spectrometry; N-glycan, N-linked oligosaccharide; N-linked, Asn-linked; RP-HPLC, reverse phase high performance liquid chromatography; TFA, trifluoroacetic acid; TMD, transmembrane domain
|
References |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsSupplementary dataAcknowledgmentsReferences |
|---|
Altmann, F., Fabini, G., Ahorn, H., and Wilson, I.B. (2001) Genetic model organisms in the study of N-glycans. Biochimie, 83, 703–712.[Medline]
Baenziger, J.U. and Fiete, D. (1979) Structural determinants of Concanavalin A specificity for oligosaccharides. J. Biol. Chem., 254, 2400–2407.
Brunet, S., Thibault, P., Gagnon, E., Kearney, P., Bergeron, J.J., and Desjardins, M. (2003) Organelle proteomics: looking at less to see more. Trends Cell Biol., 13, 629–638.[CrossRef][ISI][Medline]
Cipollo, J.F., Costello, C.E., and Hirschberg, C.B. (2002) The fine structure of Caenorhabditis elegans N-glycans. J. Biol. Chem., 277, 49143–49157.
Colognato, H., Winkelmann, D.A., and Yurchenco, P.D. (1999) Laminin polymerization induces a receptor-cytoskeleton network. J. Cell Biol., 145, 619–631.
Cox, E.A. and Hardin, J. (2004) Sticky worms: adhesion complexes in Caenorhabditis elegans. J. Cell Sci., 117, 1885–1897.
Cox, E.A., Tuskey, C., and Hardin, J. (2004) Cell adhesion receptors in C. elegans. J. Cell Sci., 117, 1867–1870.
Dennis, J.W., Granovsky, M., and Warren, C.E. (1999a) Glycoprotein glycosylation and cancer progression. Biochim. Biophys. Acta., 1473, 21–34.[Medline]
Dennis, J.W., Granovsky, M., and Warren, C.E. (1999b) Protein glycosylation in development and disease. Bioessays, 21, 412–421.[CrossRef][ISI][Medline]
Epstein, H.F. and Shakes, D.C. (1995) Caenorhabditis elegans: modern biological analysis of an organism. Academic Press, San Diego, CA.
Fan, X., She, Y.M., Bagshaw, R.D., Callahan, J.W., Schachter, H., and Mahuran, D.J. (2004) A method for proteomic identification of membrane-bound proteins containing Asn-linked oligosaccharides. Anal. Biochem., 332, 178–186.[CrossRef][ISI][Medline]
Feizi, T. (2000) Carbohydrate-mediated recognition systems in innate immunity. Immunol. Rev., 173, 79–88.[CrossRef][ISI][Medline]
Fukuda, M. (2000) Cell surface carbohydrates: cell type-specific expression. In Fukuda, M. and Hindsgaul, O. (eds), Molecular and Cellular Glycobiology. Oxford University Press, Oxford, pp. 1–61.
Geoghegan, K.F., Hoth, L.R., Tan, D.H., Borzilleri, K.A., Withka, J.M., and Boyd, J.G. (2002) Cyclization of N-terminal S-carbamoylmethylcysteine causing loss of 17 Da from peptides and extra peaks in peptide maps. J. Proteome Res., 1, 181–187.[CrossRef][ISI][Medline]
Haaparanta, T., Uitto, J., Ruoslahti, E., and Engvall, E. (1991) Molecular cloning of the cDNA encoding human laminin A chain. Matrix, 11, 151–160.[ISI][Medline]
Haslam, S.M. and Dell, A. (2003) Hallmarks of Caenorhabditis elegans N-glycosylation: complexity and controversy. Biochimie, 85, 25–32.[Medline]
Haslam, S.M., Gems, D., Morris, H.R., and Dell, A. (2002) The glycomes of Caenorhabditis elegans and other model organisms. Biochem. Soc. Symp., 69, 117–134.
Helenius, A. and Aebi, M. (2001) Intracellular function of N-linked glycans. Science, 291, 2364–2369.
Hong, L., Elbl, T., Ward, J., Franzini-Armstrong, C., Rybicka, K.K., Gatewood, B.K., Baillie, D.L., and Bucher, E.A. (2001) MUP-4 is a novel transmembrane protein with functions in epithelial cell adhesion in Caenorhabditis elegans. J. Cell Biol., 154, 403–414.
Hresko, M.C., Schriefer, L.A., Shrimankar, P., and Waterston, R.H. (1999) Myotactin, a novel hypodermal protein involved in muscle-cell adhesion in Caenorhabditis elegans. J. Cell Biol., 146, 659–672.
Huang, C.C., Hall, D.H., Hedgecock, E.M., Kao, G., Karantza, V., Vogel, B.E., Hutter, H., Chisholm, A.D., Yurchenco, P.D., and Wadsworth, W.G. (2003) Laminin a subunits and their role in C. elegans development. Development, 130, 3343–3358.
Hutter, H., Vogel, B.E., Plenefisch, J.D., Norris, C.R., Proenca, R.B., Spieth, J., Guo, C., Mastwal, S., Zhu, X., Scheel, J., and Hedgecock, E.M. (2000) Conservation and novelty in the evolution of cell adhesion and extracellular matrix genes. Science, 287, 989–994.
Kaji, H., Saito, H., Yamauchi, Y., Shinkawa, T., Taoka, M., Hirabayashi, J., Kasai, K., Takahashi, N., and Isobe, T. (2003) Lectin affinity capture, isotope-coded tagging and mass spectrometry to identify N-linked glycoproteins. Nat Biotechnol., 21, 667–672.[CrossRef][ISI][Medline]
Kornfeld, R. and Kornfeld, S. (1985) Assembly of asparagine-linked oligosaccharides. Ann. Rev. Biochem., 54, 631–664.[CrossRef][ISI][Medline]
Krutchinsky, A.N., Zhang, W., and Chait, B.T. (2000) Rapidly switchable matrix-assisted laser desorption/ionization and electrospray quadrupole-time-of flight mass spectrometry for protein identification. J. Am. Soc. Mass Spectrom., 11, 493–504.[CrossRef][ISI][Medline]
Lowe, J.B. (2002) Glycosylation in the control of selectin counter-receptor structure and function. Immunol. Rev., 186, 19–36.[CrossRef][ISI][Medline]
Lu, Z., Szafron, D., Greiner, R., Lu, P., Wishart, D.S., Poulin, B., Anvik, J., Macdonell, C., and Eisner, R. (2004) Predicting subcellular localization of proteins using machine-learned classifiers. Bioinformatics, 20, 547–556.
Merkle, R.K. and Cummings, R.D. (1987) Lectin affinity chromatography of glycopeptides. Methods Enzymol., 138, 232–259.[ISI][Medline]
Natsuka, S., Adachi, J., Kawaguchi, M., Nakakita, S., Hase, S., Ichikawa, A., and Ikura, K. (2002) Structure analysis of N-linked glycans in Caenorhabditis elegans. J. Biochem. (Tokyo), 131, 807–813.
Riddle, D.L., Blumenthal, T., Meyer, B.J., and Priess, J.R. (1997) C. Elegans II. Cold Spring Harbor Laboratory Press, Plainview, NY.
Schachter, H. (2004) Protein glycosylation lessons from Caenorhabditis elegans. Curr. Opin. Struct. Biol., 14, 607–616.[CrossRef][ISI][Medline]
Schachter, H., Chen, S., Zhang, W., Spence, A.M., Zhu, S., Callahan, J.W., Mahuran, D.J., Fan, X., Bagshaw, R.D., She, Y.M., and others. (2002) Functional post-translational proteomic approach to study the role of N-glycans in the development of Caenorhabditis elegans. Biochem. Soc. Symp., 69, 1–21.
She, Y.M., Haber, S., Seifers, D.L., Loboda, A., Chernushevich, I., Perreault, H., Ens, W., and Standing, K.G. (2001) Determination of the complete amino acid sequence for the coat protein of brome mosaic virus by time-of-flight mass spectrometry. J. Biol. Chem., 276, 20039–20047.
Tepass, U. (1999) Genetic analysis of cadherin function in animal morphogenesis. Curr. Opin. Cell Biol., 11, 540–548.[CrossRef][ISI][Medline]
Varki, A. (1993) Biological roles of oligosaccharides: all of the theories are correct. Glycobiology, 3, 97–130.
Varki, A. and Marth, J. (1995) Oligosaccharides in vertebrate development. Seminars Dev. Biol., 6, 127–138.
Von Heijne, G. (1990) The signal peptide. J. Membr. Biol., 115, 195–201.[CrossRef][ISI][Medline]
Wood, W.B. (1988) The Nematode Caenorhabditis Elegans. Cold Spring Harbor Laboratory Press, Plainview, NY.
Zamir, E. and Geiger, B. (2001) Molecular complexity and dynamics of cell-matrix adhesions. J. Cell Sci., 114, 3583–3590.
Zhang, W., Cao, P., Chen, S., Spence, A.M., Zhu, S., Staudacher, E., and Schachter, H. (2003) Synthesis of paucimannose N-glycans by Caenorhabditis elegans requires prior actions of UDP-GlcNAc: alpha-3-D-mannoside beta 1,2-N-acetylglycosaminyltransferase I, alpha-3,6-mannosidase II and a specific membrane-bound beta-N-acetylglucosaminidase. Biochem. J., 372, 53–64.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  K. Koles, J.-M. Lim, K. Aoki, M. Porterfield, M. Tiemeyer, L. Wells, and V. Panin Identification of N-Glycosylated Proteins from the Central Nervous System of Drosophila Melanogaster Glycobiology, December 1, 2007; 17(12): 1388 - 1403. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Gutternigg, D. Kretschmer-Lubich, K. Paschinger, D. Rendic, J. Hader, P. Geier, R. Ranftl, V. Jantsch, G. Lochnit, and I. B. H. Wilson Biosynthesis of Truncated N-Linked Oligosaccharides Results from Non-orthologous Hexosaminidase-mediated Mechanisms in Nematodes, Plants, and Insects J. Biol. Chem., September 21, 2007; 282(38): 27825 - 27840. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Ju, Q. Zheng, and R. D. Cummings Identification of core 1 O-glycan T-synthase from Caenorhabditis elegans Glycobiology, October 1, 2006; 16(10): 947 - 958. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||