With glycosylation now firmly established across both Archaeal and bacterial proteins, a wide array of glycan diversity has become evident from structural analysis and genomic data. These discoveries have been built in part on the development and application of mass spectrometric technologies to the bacterial glycoproteome. This review highlights recent findings using high sensitivity MS of the large variation of glycans that have been reported on flagellin and pilin proteins of bacteria, using both ‘top down’ and ‘bottom up’ approaches to the characterization of these glycoproteins. We summarize current knowledge of the sugar modifications that have been observed on flagellins and pilins, in terms of both the diverse repertoire of monosaccharides observed, and the assemblage of moieties that decorate many of these sugars.

Introduction

Glycosylation, the post-translational modification of proteins by carbohydrates, has long been recognised as a fundamental strategy used by eukaryotes to influence and modulate protein structure and function [1]. It is now evident that protein glycosylation is abundant in prokaryotes, with an ever increasing number of glycoproteins being identified. Sugars present on glycoproteins have traditionally been studied once released from the protein, but new technologies are now permitting the analysis of sugars in situ using MS [2]. The earliest examples of protein glycosylation in prokaryotes were found in the Archaea, which express glycosylated surface (S-layer) proteins [3], and a number of S-layer glycans have now been reported [4]. Subsequently, protein glycosylation was identified in Bacteria such as Flavobacterium meningosepticum [5,6] and Neisseria meningitidis [6,7], but was largely ignored as a rare curiosity until the discovery of a general N-glycosylation system in Campylobacter jejuni following the completion of its genome sequence [8]. It is now firmly established that glycosylation is widespread across both Archaea and bacteria [9]. Indeed, with the progress in sequencing of bacterial genomes, it has become more evident that shared genes among pathogenic bacteria are involved in the glycosylation process and that there is a commonality of sugars across bacteria. An example of commonality is the O-glycosylation of Neisseria pilin and the N-glycosylation of Campylobacter flagellin, where the sugar DATDH (2,4-diacetamido-2,4,6-trideoxyhexose) links the glycan to the protein in both cases [10]. These two pathogens share glycosylation genes (pgl) involved in the synthesis and transfer of DATDH onto the protein backbone [11]. The MS glycosylation studies in Campylobacter recognised that ES-MS (electrospray MS) analysis of glycopeptides yielded abundant signature fragment ions for the nitrogen-containing sugars, and this has facilitated their discovery in other organisms. For example, work in our laboratory has characterized the sugar DATDH in Wolinella succinogenes (P.G. Hitchen and A. Dell), which was predicted from the presence of pgl genes in the genome sequence [12]. MS analysis has revealed the presence of a hexasaccharide attached to the protein via the DATDH residue, which was first noted from the observation of the characteristic signature oxonium ion (m/z 229) for DATDH.

To date, the study of prokaryotic protein glycosylation in bacteria has been best exemplified by the flagellin and pilin glycoproteins [13,14]. Flagellins and pilins are proteins that are arranged into large filamentous structures that extrude from the bacterial surface and are termed the flagellum and pilus respectively. The flagellum is important for bacterial virulence, motility and colonization. Short pilus, termed fimbriae, are used to attach the pathogen to the host surface, whereas type IV pili are integral for adherence and motility. Although the exact function of glycosylation of these appendages has yet to be fully understood, it is evident that glycosylation is functionally important. For example, mutants in glycosylation of Campylobacter flagellin are non-motile and accumulate flagellin intracellularly and have also been shown to display limited flagellin glycoforms that have a significant contributory role in the colonization of chickens [15,16]. The role of pilin glycosylation is less clear. Loss of O-linked glycosylation in Niesseria meningitidis did not support a major role for glycosylation in pilin assembly and subsequent adhesion [17].

In the present article, we aim to highlight the methodologies now being applied to study prokaryotic glycoproteins and we describe how glycoproteomics has revolutionised the discovery of new glycoforms of flagellin and pilin glycoproteins.

Glycoproteomic strategies

The application of MS has proved to be an essential tool over the years for defining carbohydrate structure [2,18]. Advances in MS instrumentation coupled with the success of proteomics has led to the development of new MS strategies for the analysis of bacterial glycoproteins [19]. These techniques have exploited knowledge of genomic information for the functional analysis of genes involved in the glycosylation process. Typical glycoproteomic strategies are highlighted in Figure 1. These techniques can be categorized as ‘top down’ and ‘bottom up’ MS strategies. In a top down approach, ES-MS is typically applied to the analysis of intact glycoproteins. The resulting data are then transformed by the software to ascertain the glycoprotein mass and the extent of glycosylation. In addition, a top down experiment often provides valuable data on the nature of the glycan from the observation of characteristic glycan oxonium ions in the low mass region of the mass spectrum. The formation of these oxonium ions is favoured by amino sugars and it is therefore advantageous that flagellin and pilin proteins usually carry this class of sugars. In a bottom up approach, glycoproteins are proteolytically cleaved to generate glycopeptides, typically using trypsin. These pieces of the glycoprotein are then separated on a liquid chromatography system and analyzed by CAD-MS/MS (collisionally-assisted dissociation tandem MS) in a manner similar to peptide sequencing. From the resulting data, glycopeptides are often identified by examining the data for the same characteristic oxonium ions observed in top down experiments. The CAD-MS/MS data can also give information on the sequence of monosaccharide residues within the glycan attached to the peptide, although often not the specific site of attachment, due to the facile loss of sugars which is inherent to this type of analysis. Recent developments in MS instrumentation have seen the implementation of the complementary technique of ETD-MS/MS (electron transfer dissociation tandem MS), which, owing to the mechanisms involved, favourably fragments the peptide backbone rather than glycosidic bonds, and allows for the identification of specific sites of glycosylation [20]. These MS techniques are invaluable in assigning sugar compositions to flagellin and pilin glycoproteins, as indicated by the masses of the observed glycans. However, owing to the diverse nature of sugars found on bacterial glycoproteins, NMR is often essential as a complementary structural technique for the precise structural determination of novel and unusual sugars that are a common feature of bacterial glycoproteins [21].

‘Top down’ and ‘bottom up’ MS techniques for the analysis of glycoproteins

Figure 1
‘Top down’ and ‘bottom up’ MS techniques for the analysis of glycoproteins

The top down MS approach highlights the application of ES-MS for the analysis of intact glycoproteins, where the glycoprotein is observed as a multiply charged species and is then transformed to give the masses of various glycoforms of the modified protein. In the low mass region, signals may be observed for characteristic glycan oxonium ions that fragment from the glycoprotein during the ES-MS experiment. Additional fragmentation can be performed to gain further insight into the nature of the glycan oxonium ion, using CAD-MS/MS resulting in losses of indicative moieties from the glycan. The bottom up MS approach highlights the LC-MS/MS strategy for the analysis of tryptic glycopeptides generated from a sample. Glycopeptides are separated by LC and selected for CAD-MS/MS analysis for the characterization of glycan oxonium ions and sequence. The bottom right shows the complementary MS technique of ETD-MS/MS, which favorably cleaves the peptide backbone, leaving intact side chain modifications such as glycosylation and enabling the identification of linkage sites. Glycosylation/glycan oxonium ions are represented by filled squares and the filled circle indicates a phosphocholine modification of the polypeptide.

Figure 1
‘Top down’ and ‘bottom up’ MS techniques for the analysis of glycoproteins

The top down MS approach highlights the application of ES-MS for the analysis of intact glycoproteins, where the glycoprotein is observed as a multiply charged species and is then transformed to give the masses of various glycoforms of the modified protein. In the low mass region, signals may be observed for characteristic glycan oxonium ions that fragment from the glycoprotein during the ES-MS experiment. Additional fragmentation can be performed to gain further insight into the nature of the glycan oxonium ion, using CAD-MS/MS resulting in losses of indicative moieties from the glycan. The bottom up MS approach highlights the LC-MS/MS strategy for the analysis of tryptic glycopeptides generated from a sample. Glycopeptides are separated by LC and selected for CAD-MS/MS analysis for the characterization of glycan oxonium ions and sequence. The bottom right shows the complementary MS technique of ETD-MS/MS, which favorably cleaves the peptide backbone, leaving intact side chain modifications such as glycosylation and enabling the identification of linkage sites. Glycosylation/glycan oxonium ions are represented by filled squares and the filled circle indicates a phosphocholine modification of the polypeptide.

Three groups of flagellin and pilin glycans

(i) Acidic monosaccharides that are structurally related to sialic acids and are decorated with a variety of substituents

Flagellins from at least six species of pathogen have been found to be glycosylated with families of ‘sialic acid-like’ monosaccharides. These are derivatives of 5,7-diacetamido-3,5,7,9-tetradeoxy-L-glycero-α-L-manno-nonulosonic acid and 5,7-diacetamido-3,5,7,9-tetradeoxy-D-glycero-β-D-galacto-nonulosonic acid, which are called pseudaminic acid and legionaminic acid respectively. Diversity within each family is conferred by replacement of acetamido with acetamidino, by variations in the N-acyl groups, and by modifications of hydroxy groups. Pseudaminic acid and legionaminic acid derivatives characterized to date are shown in Figure 2(A) together with the masses of their diagnostic oxonium ions, whose presence in both ‘top-down’ and ‘bottom-up’ mass spectra (see Figure 1) is vital for structural assignments. Members of the pseudaminic acid family have been found on flagellins of Campylobacter jejuni, Campylobacter coli, Helicobacter pylori and Aeromonas caviae [2226]. In the latter pathogen, pseudaminic acid is also a component of its LPS (lipopolysaccharide) O-antigen [26]. Some strains of Campylobacter additionally carry legionaminic acid family members on their flagellins and this type of glycosylation is also present on the flagellin of some strains of Clostridium botulinum [27]. The pseudaminic acid and legionaminic acid families include derivatives acylated with 2,3-di-O-methylglyceric acid and amino acids such as N-methylated glutamic acid (see Figure 2A) [27,28].

Diversity of sialic acid analogues in pathogenic flagellins and pilins

Figure 2
Diversity of sialic acid analogues in pathogenic flagellins and pilins

(A) Pseudaminic acid and legionaminic acid, with various R group substituents and characteristic oxonium ions with m/z values shown below. Glycan moiety R groups are as follows: Ac, acetamido; Am, acetamidino; AmNMe, methylacetimidoyl; [3OH]butyryl, 3-hydroxybutyryl; DiOMeGlyA, di-O-methylglycerate; Fm, formyl; GlnNAc, acetylglutamine; GluNMe, methylglutamyl; Pr, dihydroxypropinoyl. (B) The glycan structure characterized for C. difficile 630 strain [42]. (C) The heterogeneous family of oligosaccharides observed in hypervirulent C. difficile 027 strains across distinct geographical locations. deoxyHex, deoxyhexose; Hep, heptose; Me, methyl group.

Figure 2
Diversity of sialic acid analogues in pathogenic flagellins and pilins

(A) Pseudaminic acid and legionaminic acid, with various R group substituents and characteristic oxonium ions with m/z values shown below. Glycan moiety R groups are as follows: Ac, acetamido; Am, acetamidino; AmNMe, methylacetimidoyl; [3OH]butyryl, 3-hydroxybutyryl; DiOMeGlyA, di-O-methylglycerate; Fm, formyl; GlnNAc, acetylglutamine; GluNMe, methylglutamyl; Pr, dihydroxypropinoyl. (B) The glycan structure characterized for C. difficile 630 strain [42]. (C) The heterogeneous family of oligosaccharides observed in hypervirulent C. difficile 027 strains across distinct geographical locations. deoxyHex, deoxyhexose; Hep, heptose; Me, methyl group.

(ii) Oligosaccharides which share sequences with LPS O-antigens

Post-translational modification involving attachment of an O-antigen sequence has so far only been structurally confirmed in the pilin of Pseudomonas aeruginosa strain 1244 [29]. However, results emerging from studies of other bacterial glycoproteins, such as the adhesins of the human oral pathogen Aggregatibacter actinomycetemcomitans, suggest that this type of glycosylation is likely to occur more widely [30]. The O-antigen of P. aeruginosa strain 1244 is a trisaccharide comprising a pseudaminic acid derivative (see Figure 2A) capping a Xyl-FucNAc (xylose-N-acetylfucosamine) disaccharide. This, or a closely related structure, is likely to be also present on the flagellin because its glycosylation has been shown to require one of the O-antigen biosynthetic enzymes [31].

(iii) Glycans whose structures differ from LPS O-antigens

This category embraces a variety of structures of which the best characterised are flagellins from Pseudomonas syringae, Clostridium difficile, C. botulinum and Listeria monocytogenes; pilins from Neisseria meningitidis and Neisseria gonorrhoeae; and pilins and flagellins from several strains of Ps. aeruginosa. Below, we begin with an overview of the glycans that have been identified in several Pseudomonas and Neisseria species whose glycosylation is now quite well understood [29,3234]). We then describe results emerging from research that is providing a wealth of information on O-linked glycosylation in hypervirulent strains of C. difficile.

Pilins from N. meningitidis and N. gonorrhoeae were amongst the first bacterial glycoproteins to be characterized by MS in the mid-1990s using bottom-up technologies. Both species of Neisseria are glycosylated at a single site with a trisaccharide composed of a di-galactose moiety attached to serine via a DATDH residue, a linking sugar that is shared with the N-glycosylation system of Campylobacter [7,35]). Subsequently, top-down analysis was applied with great success to the pilins of N. gonorrhoeae. These experiments have revealed glycan micro-heterogeneity, including O-acetylation of the terminal galactose plus truncation to di- and mono-saccharides, as well as defining phosphocholine and phosphoethanolamine modifications of the polypeptide backbone [7,33,35]. Moreover, MS played a vital role in the recent discovery of a general O-linked glycosylation system in Neisseria [36].

Of all the flagellins and pilins characterized to date, those from Ps. aeruginosa strains exhibit the greatest diversity. As described earlier, the pilin of strain 1244 is decorated with O-antigens. In contrast, the glycans on the pilin of strain Pa5196 are homo-oligomers of α1,5-linked D-arabinofuranose, structures which, interestingly, are shared by cell wall polymers of Mycobacteria [37]. Flagellins have been characterized in two strains, PAK and JJ692. Remarkably, despite both having O-linked rhamnose as the linking sugar, the glycans from these two strains differ substantially [32]. Thus in JJ692, the rhamnose is not further elongated, whereas the PAK flagellin was found to carry an extremely heterogeneous family of rhamnose-linked glycans containing over thirty different glycan compositions. The glycans vary in size from six to eleven residues and all share a peripheral trisaccharide composed of an unknown terminal residue of mass 174 Da linked to a deoxyhexose-deoxyhexosamine disaccharide. The oligosaccharide chains connecting this trisaccharide to the flagellin-linked rhamnose comprise many different combinations of pentose, deoxyhexose, hexose and hexuronic acid.

Similar to the aforementioned Ps. aeruginosa strains, the plant pathogen Ps. syringae has rhamnose-linked glycans on its flagellins. However the latter has relatively homogeneous glycosylation, in contrast to the diverse glycan repertoire that can be expressed by Ps. aeruginosa. Two pathovars have been characterized and they are modified with tri- or tetra-saccharides containing two or three rhamnosyl residues respectively, capped with 4,6-dideoxy-4-(3-hydroxybutanamido)-2-O-methylglucose [38].

Top down MS has uncovered two distinct families of glycans in the flagellins of C. botulinum [27]. As described earlier, some strains are decorated with legionaminic acid derivatives. Others exhibit oxonium ions whose masses suggest the presence of di-N-acetylhexuronic acid and its O-acetylated counterpart (Table 1). It is not yet known whether this uronic acid family is present as monosaccharides and their derivatives, similar to the legionaminic acid family, or whether they occur within oligosaccharides.

Table 1
Diversity of sugars observed by MS on flagellin and pilin proteins of bacteria

m/z values are shown as mass increments for each monosaccharide. Ac, acetyl; Hep, heptose; HexA, hexuronic acid; HexN, hexosamine; Leg, legionaminic acid; Pent, pentose; Pse, pseudaminic acid.

m/z Putative assignment Glycan characterized in: 
116 Deoxypent Linked to Pse in C. jejuni [22,45
132 Pent Ps. aeruginosa [29,32
145 DeoxyhexN Ps. aeruginosa [32], Ps. syringae [38
146 Deoxyhexose Ps. aeruginosa [32], Ps. syringae [38], C. difficile [42
160 Methyl-deoxyhexose C. difficile [42
162 Hexose N. meningitidis, N. gonorrhoeae, Ps. aeruginosa [7,32
174 Unassigned, possibly diaminoHexA Ps. aeruginosa [32
176 HexA Ps. aeruginosa [32
187 Deoxy-N-acetylhexose Ps. aeruginosa [29,32
192 Hep C. difficile [42
203 HexNAc L. monocytogenes, C. difficile [39,42
228 DATDH N. meningitidis, N. gonorrhoeae [7,35
246 DideoxyHex4N(3-hydroxy-1-oxobutyl)2Me Ps. syringae [38
258 Di-N-acetylHexA C. botulinum [27
300 O-Ac-di-N-acetylHexA C. botulinum [27
346 Pse5NβOHC47NFm Ps. aeruginosa [29
359 Unassigned; probably a Pse derivative C. jejuni [46
373 Unassigned; probably a Pse derivative A. caviae, C. jejuni [25,46
417 Unassigned; probably a Leg derivative C. botulinum [27
431 Unassigned; possibly Pse5Ac7Am plus deoxypent C. botulinum [27
432 Unassigned; possibly Pse5Ac7Ac plus deoxypent C. botulinum [27
696 Unassigned C. botulinum [27
m/z Putative assignment Glycan characterized in: 
116 Deoxypent Linked to Pse in C. jejuni [22,45
132 Pent Ps. aeruginosa [29,32
145 DeoxyhexN Ps. aeruginosa [32], Ps. syringae [38
146 Deoxyhexose Ps. aeruginosa [32], Ps. syringae [38], C. difficile [42
160 Methyl-deoxyhexose C. difficile [42
162 Hexose N. meningitidis, N. gonorrhoeae, Ps. aeruginosa [7,32
174 Unassigned, possibly diaminoHexA Ps. aeruginosa [32
176 HexA Ps. aeruginosa [32
187 Deoxy-N-acetylhexose Ps. aeruginosa [29,32
192 Hep C. difficile [42
203 HexNAc L. monocytogenes, C. difficile [39,42
228 DATDH N. meningitidis, N. gonorrhoeae [7,35
246 DideoxyHex4N(3-hydroxy-1-oxobutyl)2Me Ps. syringae [38
258 Di-N-acetylHexA C. botulinum [27
300 O-Ac-di-N-acetylHexA C. botulinum [27
346 Pse5NβOHC47NFm Ps. aeruginosa [29
359 Unassigned; probably a Pse derivative C. jejuni [46
373 Unassigned; probably a Pse derivative A. caviae, C. jejuni [25,46
417 Unassigned; probably a Leg derivative C. botulinum [27
431 Unassigned; possibly Pse5Ac7Am plus deoxypent C. botulinum [27
432 Unassigned; possibly Pse5Ac7Ac plus deoxypent C. botulinum [27
696 Unassigned C. botulinum [27

Top down MS also played a key role in the discovery of O-GlcNAc (β-O-linked N-acetylglucosamine) in the flagellins of Listeria monocytogenes. O-GlcNAc is a ubiquitous sugar on nuclear and cytoplasmic proteins of eukaryotes where it is implicated in many essential biological processes such as signalling and transcriptional regulation. MS showed that HexNAc (N-acetylhexosamine) moieties are attached to up to six serine and threonine residues in the central surface-exposed region of the Listeria flagellin [39]. These HexNAc residues were assigned as β-linked GlcNAc because the Listeria flagellin was found to be immunoreactive with a monoclonal antibody that recognises O-GlcNAc in eukaryotic proteins.

C. difficile is an emerging opportunistic pathogen which is the leading cause of nosocomial diarrhoea worldwide. Severity and mortality of C. difficile infection has increased and are correlated with the emergence of C. difficile PCR-ribotype 027 strains, causing several antibiotic-associated nosocomial outbreaks in North America and Europe [40,41]. The origin and development of these hypervirulent strains remains largely unknown despite efforts to identify the genetic elements and resulting pathogenicity determinants which enable their increased virulence. One avenue being explored by Logan and co-workers in Ottawa is the possibility that changes in flagellin structure contribute to pathogenicity [42]. Their elegant MS analyses have uncovered remarkable differences in the glycosylation of flagellins from recent clinical isolates from the U.S.A. and Canada compared with the ‘historic’ 630 strain which was isolated from a hospital patient in Zurich approx. 30 years ago [42]. The 630 flagellin was found to be glycosylated at up to seven sites with an O-linked HexNAc carrying aspartic acid attached via a phosphodiester moiety (Figure 2B, diagnostic oxonium ion at 399 Da). C. difficile strains, including ribotype-027 strains, were examined and the glycans were found to be similarly attached to the flagellin backbone via a HexNAc but none exhibited the 399 Da oxonium ion that is characteristic of the 630 strain. Instead, the hypervirulent C. difficile strains carry a heterogeneous family of oligosaccharides containing up to five sugar residues arranged in various combinations of deoxyhexose, methylated deoxyhexose, heptose and HexNAc (see Figure 2C). In our laboratory we have examined hypervirulent C. difficile ribotype-027 strains isolated in France in 1985 (strain CD196) and in England in 2006 (strain R20291), the latter being associated with a hospital epidemic that resulted in 127 deaths over three years. We observed similar structures (P.G. Hitchen and A. Dell, unpublished work) to those 027 strains characterized by the Logan laboratory, suggesting that this glycan family is shared by hypervirulent strains irrespective of geographical location. It remains to be seen whether the glycans play a role in C. difficile colonization and/or virulence.

Partially characterized pilin and flagellin glycosylation

Within the past few years, top down MS has transformed the analysis of pilins and flagellins, and has uncovered a wealth of residue masses whose molecular make-up requires complementary techniques such as NMR and metabolomics for rigorous assignment [43]. The glycans described in previous sections attest to the very considerable progress that has already been made in defining these glycan structures. Nevertheless, detailed structural analysis is often very challenging, and an increasing number of oxonium ions observed in top down experiments, as well as mass increments observed in bottom up experiments, await rigorous structural assignment. Current understanding of pathogen glycosylation suggests that these sugars and their derivatives are likely to occur in more than one species. To facilitate the future discovery of shared glycan moieties, Table 1 contains a summary, with possible assignments, of fragment ion masses that have not yet been attributed to defined glycan structures in pilins and flagellins.

Density of glycosylation

Although a discussion of glycan function is not within the scope of the present article, it is, nevertheless, pertinent to consider whether the overall architectures of pilin and flagellin glycosylation are consistent with glycans playing a direct role in host–pathogen engagement. Many studies have documented the importance of multivalency in the interaction of glycans of diverse bacterial, plant and animal origin with their receptors (lectins) [1]. Indeed, it is becoming increasingly evident that appropriate glycan density is likely to be a pre-requisite for functional outcomes of glycan–lectin recognition [44]. So it is relevant to note that pilins and flagellins have very different densities of glycosylation. Thus flagellins are characterized by multiple sites of glycosylation as well as considerable glycan heterogeneity. For example, top down MS has revealed up to 19 glycosylation sites on flagellins of C. jejuni, H. pylori and A. caviae [25]. It is conceivable, therefore, that flagellins have the potential to be involved in processes that are regulated, at least in part, by glycan recognition. In contrast, characterized pilins such as those from Neisseria [33] have only a single glycosylation site, although serine-liked phosphocholine and phosphoethanolamine are adjacent to the glycosylation site, and could conceivably play a role in strengthening putative inter-molecular interactions mediated by the glycans.

Closing remarks

We have highlighted the relatively limited repertoire of underlying sugars present on flagellin and pilin glycoproteins of pathogenic bacteria, and also the extensive diversity that can result from the large cassettes of sugar modification genes that populate the glycosylation loci. Thus the study of this vast array of glycostructures presented by pathogens is inherently complex, but ideally suited to the powerful application of MS. The inherent sensitivity of MS and its ability to generate valuable information on very small sample quantities, coupled to its capacity to distinguish between very similar glycoforms within a population, makes MS an essential tool in the analysis of these biopolymers. In turn, the nature of the family of amino sugars so far identified on these appendages facilitates their study by MS, the glycans being an ideal size for detection which, combined with their ability to yield oxonium ions, enhances their discovery. With such a diverse range of sugar modifications across pathogenic bacteria and even strains, a more systematic approach is required to better grasp the complex network of glycosylation gene functions combined with the elaborate repertoire of glycostructures cloaking the bacteria that are perceived by the host. We are using a systems biology approach in the CISBIC (Centre for Integrative Systems Biology at Imperial College) to better understand the complexity of the interaction between bacterial glycomes and the host immune response during infection. We are developing models to enable a better understanding of the complexity of the Campylobacter glycome using logic programming in combination with wet lab experimentation [47]. The diversity of the Campylobacter glycome presents an excellent model on which a systems approach can be extended to include the study of glycome variation across strains.

Systems Approaches to Health and Disease: A Biochemical Society Focused Meeting held at University of York, U.K., 22–24 March 2010, as part of the Systems Biochemistry Linked Focused Meetings. Organized and Edited by David Fell (Oxford Brookes, U.K.), Hans Westerhoff (Manchester, U.K., and Amsterdam, The Netherlands) and Michael White (Liverpool, U.K.).

Abbreviations

     
  • CAD-MS/MS

    collisionally assisted dissociation tandem MS

  •  
  • DATDH

    2,4-diacetamido-2,4,6-trideoxyhexose, ES-MS, electrospray MS

  •  
  • ETD-MS/MS

    electron transfer dissociation tandem MS

  •  
  • HexNAc

    N-acetylhexosamine

  •  
  • LPS

    lipopolysaccharide

  •  
  • O-GlcNAc

    β-O-linked N-acetylglucosamine

Funding

This work was funded by the UK Biotechnology and Biological Science Research Council (BBSRC) Core Support Grant [grant number B19088 (to A.D.)] and Integrative Systems Biology Grant [grant number BBC5196701 (to P.G.H., B.W.W. and A.D.)].

References

References
1
Taylor
M.T.
Drickamer
K.
Introduction to Glycobiology
2002
Oxford
Oxford University Press
2
Dell
A.
Morris
H.R.
Glycoprotein structure determination by mass spectrometry
Science
2001
, vol. 
291
 (pg. 
2351
-
2356
)
3
Mescher
M.F.
Strominger
J.L.
Purification and characterization of a prokaryotic glucoprotein from the cell envelope of Halobacterium salinarium
J. Biol. Chem.
1976
, vol. 
251
 (pg. 
2005
-
2014
)
4
Messner
P.
Prokaryotic protein glycosylation is rapidly expanding from “curiosity” to “ubiquity”
ChemBioChem
2009
, vol. 
10
 (pg. 
2151
-
2154
)
5
Plummer
T.H.
Jr
Tarentino
A.L.
Hauer
C.R.
Novel, specific O-glycosylation of secreted Flavobacterium meningosepticum proteins. Asp-Ser and Asp-Thr-Thr consensus sites
J. Biol. Chem.
1995
, vol. 
270
 (pg. 
13192
-
13196
)
6
Reinhold
B.B.
Hauer
C.R.
Plummer
T.H.
Reinhold
V.N.
Detailed structural analysis of a novel, specific O-linked glycan from the prokaryote Flavobacterium meningosepticum
J. Biol. Chem.
1995
, vol. 
270
 (pg. 
13197
-
13203
)
7
Stimson
E.
Virji
M.
Makepeace
K.
Dell
A.
Morris
H.R.
Payne
G.
Saunders
J.R.
Jennings
M.P.
Barker
S.
Panico
M.
, et al. 
Meningococcal pilin: a glycoprotein substituted with digalactosyl 2,4-diacetamido-2,4,6-trideoxyhexose
Mol. Microbiol.
1995
, vol. 
17
 (pg. 
1201
-
1214
)
8
Young
N.M.
Brisson
J.R.
Kelly
J.
Watson
D.C.
Tessier
L.
Lanthier
P.H.
Jarrell
H.C.
Cadotte
N.
St Michael
F.
Aberg
E.
Szymanski
C.M.
Structure of the N-linked glycan present on multiple glycoproteins in the Gram-negative bacterium, Campylobacter jejuni
J. Biol. Chem.
2002
, vol. 
277
 (pg. 
42530
-
42539
)
9
Abu-Qarn
M.
Eichler
J.
Sharon
N.
Not just for Eukarya anymore: protein glycosylation in Bacteria and ArchaeaCurr
Opin. Struct. Biol.
2008
, vol. 
18
 (pg. 
544
-
550
)
10
Szymanski
C.M.
Wren
B.W.
Protein glycosylation in bacterial mucosal pathogens
Nat. Rev. Microbiol.
2005
, vol. 
3
 (pg. 
225
-
237
)
11
Power
P.M.
Jennings
M.P.
The genetics of glycosylation in Gram-negative bacteria
FEMS Microbiol. Lett.
2003
, vol. 
218
 (pg. 
211
-
222
)
12
Baar
C.
Eppinger
M.
Raddatz
G.
Simon
J.
Lanz
C.
Klimmek
O.
Nandakumar
R.
Gross
R.
Rosinus
A.
Keller
H.
, et al. 
Complete genome sequence and analysis of Wolinella succinogenes
Proc. Natl. Acad. Sci. U.S.A.
2003
, vol. 
100
 (pg. 
11690
-
11695
)
13
Logan
S.M.
Flagellar glycosylation - a new component of the motility repertoire?
Microbiology
2006
, vol. 
152
 (pg. 
1249
-
1262
)
14
Schmidt
M.A.
Riley
L.W.
Benz
I.
Sweet new world: glycoproteins in bacterial pathogens
Trends Microbiol.
2003
, vol. 
11
 (pg. 
554
-
561
)
15
Goon
S.
Kelly
J.F.
Logan
S.M.
Ewing
C.P.
Guerry
P.
Pseudaminic acid, the major modification on Campylobacter flagellin, is synthesized via the Cj1293 gene
Mol. Microbiol.
2003
, vol. 
50
 (pg. 
659
-
671
)
16
Howard
S.L.
Jagannathan
A.
Soo
E.C.
Hui
J.P.
Aubry
A.J.
Ahmed
I.
Karlyshev
A.
Kelly
J.F.
Jones
M.A.
Stevens
M.P.
, et al. 
Campylobacter jejuni glycosylation island important in cell charge, legionaminic acid biosynthesis, and colonization of chickens
Infect. Immun.
2009
, vol. 
77
 (pg. 
2544
-
2556
)
17
Marceau
M.
Forest
K.
Beretti
J.L.
Tainer
J.
Nassif
X.
Consequences of the loss of O-linked glycosylation of meningococcal type IV pilin on piliation and pilus-mediated adhesion
Mol. Microbiol.
1998
, vol. 
27
 (pg. 
705
-
715
)
18
North
S.J.
Hitchen
P.G.
Haslam
S.M.
Dell
A.
Mass spectrometry in the analysis of N-linked and O-linked glycans
Curr. Opin. Struct. Biol.
2009
, vol. 
19
 (pg. 
498
-
506
)
19
Hitchen
P.G.
Dell
A.
Bacterial glycoproteomics
Microbiology
2006
, vol. 
152
 (pg. 
1575
-
1580
)
20
Wiesner
J.
Premsler
T.
Sickmann
A.
Application of electron transfer dissociation (ETD) for the analysis of posttranslational modifications
Proteomics
2008
, vol. 
8
 (pg. 
4466
-
4483
)
21
Kelly
J.
Brisson
J.-R.
Young
M.N.
Jarrel
H.C.
Szymanski
C.M.
Chan
R.V.L.
Sherman
P.M.
Bourke
B.
Campylobacter: from glycome to pathogenesis
Bacterial Genomes and Infectious Diseases
2006
New York
Springer
(pg. 
63
-
90
)
22
Logan
S.M.
Kelly
J.F.
Thibault
P.
Ewing
C.P.
Guerry
P.
Structural heterogeneity of carbohydrate modifications affects serospecificity of Campylobacter flagellins
Mol. Microbiol.
2002
, vol. 
46
 (pg. 
587
-
597
)
23
Thibault
P.
Logan
S.M.
Kelly
J.F.
Brisson
J.R.
Ewing
C.P.
Trust
T.J.
Guerry
P.
) Identification of the carbohydrate moieties and glycosylation motifs in Campylobacter jejuni flagellin
J. Biol. Chem.
2001
, vol. 
276
 (pg. 
34862
-
34870
)
24
Schirm
M.
Soo
E.C.
Aubry
A.J.
Austin
J.
Thibault
P.
Logan
S.M.
Structural, genetic and functional characterization of the flagellin glycosylation process in Helicobacter pylori
Mol. Microbiol.
2003
, vol. 
48
 (pg. 
1579
-
1592
)
25
Schirm
M.
Schoenhofen
I.C.
Logan
S.M.
Waldron
K.C.
Thibault
P.
Identification of unusual bacterial glycosylation by tandem mass spectrometry analyses of intact proteins
Anal. Chem.
2005
, vol. 
77
 (pg. 
7774
-
7782
)
26
Tabei
S.M.
Hitchen
P.G.
Day-Williams
M.J.
Merino
S.
Vart
R.
Pang
P.C.
Horsburgh
G.J.
Viches
S.
Wilhelms
M.
Tomas
J.M.
, et al. 
An Aeromonas caviae genomic island is required for both O-antigen lipopolysaccharide biosynthesis and flagellin glycosylation
J. Bacteriol.
2009
, vol. 
191
 (pg. 
2851
-
2863
)
27
Twine
S.M.
Paul
C.J.
Vinogradov
E.
McNally
D.J.
Brisson
J.R.
Mullen
J.A.
McMullin
D.R.
Jarrell
H.C.
Austin
J.W.
Kelly
J.F.
Logan
S.M.
Flagellar glycosylation in Clostridium botulinum
FEBS J.
2008
, vol. 
275
 (pg. 
4428
-
4444
)
28
Logan
S.M.
Hui
J.P.
Vinogradov
E.
Aubry
A.J.
Melanson
J.E.
Kelly
J.F.
Nothaft
H.
Soo
E.C.
Identification of novel carbohydrate modifications on Campylobacter jejuni 11168 flagellin using metabolomics-based approaches
FEBS J.
2009
, vol. 
276
 (pg. 
1014
-
1023
)
29
Castric
P.
Cassels
F.J.
Carlson
R.W.
Structural characterization of the Pseudomonas aeruginosa 1244 pilin glycan
J. Biol. Chem.
2001
, vol. 
276
 (pg. 
26479
-
26485
)
30
Tang
G.
Mintz
K.P.
Glycosylation of the collagen adhesin EmaA of Aggregatibacter actinomycetemcomitans is dependent upon the lipopolysaccharide biosynthetic pathway
J. Bacteriol.
2010
, vol. 
192
 (pg. 
1395
-
1404
)
31
Miller
W.L.
Matewish
M.J.
McNally
D.J.
Ishiyama
N.
Anderson
E.M.
Brewer
D.
Brisson
J.R.
Berghuis
A.M.
Lam
J.S.
Flagellin glycosylation in Pseudomonas aeruginosa PAK requires the O-antigen biosynthesis enzyme WbpO
J. Biol. Chem.
2008
, vol. 
283
 (pg. 
3507
-
3518
)
32
Schirm
M.
Arora
S.K.
Verma
A.
Vinogradov
E.
Thibault
P.
Ramphal
R.
Logan
S.M.
Structural and genetic characterization of glycosylation of type a flagellin in Pseudomonas aeruginosa
J. Bacteriol.
2004
, vol. 
186
 (pg. 
2523
-
2531
)
33
Aas
F.E.
Vik
A.
Vedde
J.
Koomey
M.
Egge-Jacobsen
W.
Neisseria gonorrhoeae O-linked pilin glycosylation: functional analyses define both the biosynthetic pathway and glycan structure
Mol. Microbiol.
2007
, vol. 
65
 (pg. 
607
-
624
)
34
Power
P.M.
Roddam
L.F.
Rutter
K.
Fitzpatrick
S.Z.
Srikhanta
Y.N.
Jennings
M.P.
Genetic characterization of pilin glycosylation and phase variation in Neisseria meningitidis
Mol. Microbiol.
2003
, vol. 
49
 (pg. 
833
-
847
)
35
Hegge
F.T.
Hitchen
P.G.
Aas
F.E.
Kristiansen
H.
Lovold
C.
Egge-Jacobsen
W.
Panico
M.
Leong
W.Y.
Bull
V.
Virji
M.
, et al. 
Unique modifications with phosphocholine and phosphoethanolamine define alternate antigenic forms of Neisseria gonorrhoeae type IV pili
Proc. Natl. Acad. Sci. U.S.A.
2004
, vol. 
101
 (pg. 
10798
-
10803
)
36
Vik
A.
Aas
F.E.
Anonsen
J.H.
Bilsborough
S.
Schneider
A.
Egge-Jacobsen
W.
Koomey
M.
Broad spectrum O-linked protein glycosylation in the human pathogen Neisseria gonorrhoeae
Proc. Natl. Acad. Sci. U.S.A.
2009
, vol. 
106
 (pg. 
4447
-
4452
)
37
Kus
J.V.
Kelly
J.
Tessier
L.
Harvey
H.
Cvitkovitch
D.G.
Burrows
L.L.
Modification of Pseudomonas aeruginosa Pa5196 type IV pilins at multiple sites with D-Araf by a novel GT-C family arabinosyltransferase, TfpW
J. Bacteriol.
2008
, vol. 
190
 (pg. 
7464
-
7478
)
38
Takeuchi
K.
Ono
H.
Yoshida
M.
Ishii
T.
Katoh
E.
Taguchi
F.
Miki
R.
Murata
K.
Kaku
H.
Ichinose
Y.
Flagellin glycans from two pathovars of Pseudomonas syringae contain rhamnose in D and L configurations in different ratios and modified 4-amino-4,6-dideoxyglucose
J. Bacteriol.
2007
, vol. 
189
 (pg. 
6945
-
6956
)
39
Schirm
M.
Kalmokoff
M.
Aubry
A.
Thibault
P.
Sandoz
M.
Logan
S.M.
Flagellin from Listeria monocytogenes is glycosylated with β-O-linked N-acetylglucosamine
J. Bacteriol.
2004
, vol. 
186
 (pg. 
6721
-
6727
)
40
Mooney
H.
Annual incidence of MRSA falls in England, but C. difficile continues to rise
BMJ
2007
, vol. 
335
 pg. 
958
 
41
O'Connor
J.R.
Johnson
S.
Gerding
D.N.
Clostridium difficile infection caused by the epidemic BI/NAP1/027 strain
Gastroenterology
2009
, vol. 
136
 (pg. 
1913
-
1924
)
42
Twine
S.M.
Reid
C.W.
Aubry
A.
McMullin
D.R.
Fulton
K.M.
Austin
J.
Logan
S.M.
Motility and flagellar glycosylation in Clostridium difficile
J. Bacteriol.
2009
, vol. 
191
 (pg. 
7050
-
7062
)
43
Twine
S.M.
Tessier
L.
Kelly
J.
Li
J.
Glycoprotein characterization
Functional Glycomics
2010
New York
Springer
(pg. 
111
-
132
)
44
Dam
T.K.
Brewer
C.F.
Lectins as pattern recognition molecules: the effects of epitope density in innate immunity
Glycobiology
2010
, vol. 
20
 (pg. 
270
-
279
)
45
Guerry
P.
Szymanski
C.M.
Campylobacter sugars sticking out
Trends Microbiol.
2008
, vol. 
16
 (pg. 
428
-
435
)
46
van Alphen
L.B.
Wuhrer
M.
Bleumink-Pluym
N.M.
Hensbergen
P.J.
Deelder
A.M.
van Putten
J.P.
A functional Campylobacter jejuni maf4 gene results in novel glycoforms on flagellin and altered autoagglutination behaviour
Microbiology
2008
, vol. 
154
 (pg. 
3385
-
3397
)
47
Kay
E.
Lesk
V.I.
Tamaddoni-Nezhad
A.
Hitchen
P.G.
Dell
A.
Sternberg
M.J.
Muggleton
S.
Wren
B.W.
Systems analysis of bacterial glycomes
Biochem. Soc. Trans.
2010
, vol. 
38
 (pg. 
1290
-
1293
)