Genetic determinants of genus-level glycan diversity in a bacterial protein glycosylation system

Glycan-Tailored Glycoproteomic Analysis Reveals Serine is the Sole Residue Subjected to O-Linked Glycosylation in Acinetobacter baumannii

Protein glycosylation is a ubiquitous process observed across all domains of life. Within the human pathogen Acinetobacter baumannii, O-linked glycosylation is required for virulence; however, the targets and conservation of glycosylation events remain poorly defined. In this work, we expand our understanding of the breadth and site specificity of glycosylation within A. baumannii by demonstrating the value of strain specific glycan electron-transfer/higher-energy collision dissociation (EThcD) triggering for bacterial glycoproteomics. By coupling tailored EThcD-triggering regimes to complementary glycopeptide enrichment approaches, we assessed the observable glycoproteome of three A. baumannii strains (ATCC19606, BAL062, and D1279779). Combining glycopeptide enrichment techniques including ion mobility (FAIMS), metal oxide affinity chromatography (titanium dioxide), and hydrophilic interaction liquid chromatography (ZIC-HILIC), as well as the use of multiple proteases (trypsin, GluC, pepsin, and thermolysis), we expand the known A. baumannii glycoproteome to 33 unique glycoproteins containing 42 glycosylation sites. We demonstrate that serine is the sole residue subjected to glycosylation with the substitution of serine for threonine abolishing glycosylation in model glycoproteins. An A. baumannii pan-genome built from 576 reference genomes identified that serine glycosylation sites are highly conserved. Combined this work expands our knowledge of the conservation and site specificity of A. baumannii O-linked glycosylation.

Combining FAIMS based glycoproteomics and DIA proteomics reveals widespread proteome alterations in response to glycosylation occupancy changes in Neisseria gonorrhoeae

Protein glycosylation is increasingly recognized as a common protein modification across bacterial species. Within the Neisseria genus O-linked protein glycosylation is conserved yet closely related Neisseria species express O-oligosaccharyltransferases (PglOs) with distinct targeting activities. Within this work, we explore the targeting capacity of different PglOs using Field Asymmetric Waveform Ion Mobility Spectrometry (FAIMS) fractionation and Data-Independent Acquisition (DIA) to allow the characterization of the impact of changes in glycosylation on the proteome of Neisseria gonorrhoeae. We demonstrate FAIMS expands the known glycoproteome of wild type N. gonorrhoeae MS11 and enables differences in glycosylation to be assessed across strains expressing different pglO allelic chimeras with unique substrate targeting activities. Combining glycoproteomic insights with DIA proteomics, we demonstrate that alterations within pglO alleles have widespread impacts on the proteome of N. gonorrhoeae. Examination of peptides known to be targeted by glycosylation using DIA analysis supports alterations in glycosylation occupancy occurs independently of changes in protein levels and that the occupancy of glycosylation is generally low on most glycoproteins. This work thus expands our understanding of the N. gonorrhoeae glycoproteome and the roles that pglO allelic variation may play in governing genus-level protein glycosylation.

Glycoproteomic and proteomic analysis of Burkholderia cenocepacia reveals glycosylation events within FliF and MotB are dispensable for motility

Across the Burkholderia genus O-linked protein glycosylation is highly conserved. While the inhibition of glycosylation has been shown to be detrimental for virulence in Burkholderia cepacia complex species, such as Burkholderia cenocepacia, little is known about how specific glycosylation sites impact protein functionality. Within this study, we sought to improve our understanding of the breadth, dynamics, and requirement for glycosylation across the B. cenocepacia O-glycoproteome. Assessing the B. cenocepacia glycoproteome across different culture media using complementary glycoproteomic approaches, we increase the known glycoproteome to 141 glycoproteins. Leveraging this repertoire of glycoproteins, we quantitively assessed the glycoproteome of B. cenocepacia using Data-Independent Acquisition (DIA) revealing the B. cenocepacia glycoproteome is largely stable across conditions with most glycoproteins constitutively expressed. Examination of how the absence of glycosylation impacts the glycoproteome reveals that the protein abundance of only five glycoproteins (BCAL1086, BCAL2974, BCAL0525, BCAM0505, and BCAL0127) are altered by the loss of glycosylation. Assessing ΔfliF (ΔBCAL0525), ΔmotB (ΔBCAL0127), and ΔBCAM0505 strains, we demonstrate the loss of FliF, and to a lesser extent MotB, mirror the proteomic effects observed in the absence of glycosylation in ΔpglL. While both MotB and FliF are essential for motility, we find loss of glycosylation sites in MotB or FliF does not impact motility supporting these sites are dispensable for function. Combined this work broadens our understanding of the B. cenocepacia glycoproteome supporting that the loss of glycoproteins in the absence of glycosylation is not an indicator of the requirement for glycosylation for protein function.

IMPORTANCE

Burkholderia cenocepacia is an opportunistic pathogen of concern within the Cystic Fibrosis community. Despite a greater appreciation of the unique physiology of B. cenocepacia gained over the last 20 years a complete understanding of the proteome and especially the O-glycoproteome, is lacking. In this study, we utilize systems biology approaches to expand the known B. cenocepacia glycoproteome as well as track the dynamics of glycoproteins across growth phases, culturing media and in response to the loss of glycosylation. We show that the glycoproteome of B. cenocepacia is largely stable across conditions and that the loss of glycosylation only impacts five glycoproteins including the motility associated proteins FliF and MotB. Examination of MotB and FliF shows, while these proteins are essential for motility, glycosylation is dispensable. Combined this work supports that B. cenocepacia glycosylation can be dispensable for protein function and may influence protein properties beyond stability.

Sweet complexity: O-linked protein glycosylation in pathogenic Neisseria

The genus Neisseria, which colonizes mucosal surfaces, includes both commensal and pathogenic species that are exclusive to humans. The two pathogenic Neisseria species are closely related but cause quite different diseases, meningococcal sepsis and meningitis (Neisseria meningitidis) and sexually transmitted gonorrhea (Neisseria gonorrhoeae). Although obvious differences in bacterial niches and mechanisms for transmission exists, pathogenic Neisseria have high levels of conservation at the levels of nucleotide sequences, gene content and synteny. Species of Neisseria express broad-spectrum O-linked protein glycosylation where the glycoproteins are largely transmembrane proteins or lipoproteins localized on the cell surface or in the periplasm. There are diverse functions among the identified glycoproteins, for example type IV biogenesis proteins, proteins involved in antimicrobial resistance, as well as surface proteins that have been suggested as vaccine candidates. The most abundant glycoprotein, PilE, is the major subunit of pili which are an important colonization factor. The glycans attached can vary extensively due to phase variation of protein glycosylation (pgl) genes and polymorphic pgl gene content. The exact roles of glycosylation in Neisseria remains to be determined, but increasing evidence suggests that glycan variability can be a strategy to evade the human immune system. In addition, pathogenic and commensal Neisseria appear to have significant glycosylation differences. Here, the current knowledge and implications of protein glycosylation genes, glycan diversity, glycoproteins and immunogenicity in pathogenic Neisseria are summarized and discussed.

CRISPRi-Mediated Silencing of Burkholderia O-Linked Glycosylation Systems Enables the Depletion of Glycosylation Yet Results in Modest Proteome Impacts

The process of O-linked protein glycosylation is highly conserved across the Burkholderia genus and mediated by the oligosaccharyltransferase PglL. While our understanding of Burkholderia glycoproteomes has increased in recent years, little is known about how Burkholderia species respond to modulations in glycosylation. Utilizing CRISPR interference (CRISPRi), we explored the impact of silencing of O-linked glycosylation across four species of Burkholderia; Burkholderia cenocepacia K56-2, Burkholderia diffusa MSMB375, Burkholderia multivorans ATCC17616, and Burkholderia thailandensis E264. Proteomic and glycoproteomic analyses revealed that while CRISPRi enabled inducible silencing of PglL, this did not abolish glycosylation, nor recapitulate phenotypes such as proteome changes or alterations in motility that are associated with glycosylation null strains, despite inhibition of glycosylation by nearly 90%. Importantly, this work also demonstrated that CRISPRi induction with high levels of rhamnose leads to extensive impacts on the Burkholderia proteomes, which without appropriate controls mask the impacts specifically driven by CRISPRi guides. Combined, this work revealed that while CRISPRi allows the modulation of O-linked glycosylation with reductions up to 90% at a phenotypic and proteome levels, Burkholderia appears to demonstrate a robust tolerance to fluctuations in glycosylation capacity.

Genetic, Functional, and Immunogenic Analyses of the O-Linked Protein Glycosylation System in Neisseria meningitidis Serogroup A ST-7 Isolates

Neisseria meningitidis exhibits a general O-linked protein glycosylation system in which pili and other extracytoplasmic proteins are glycosylated. To investigate glycan antigenicity in humans and the significance of high glycan diversity on immune escape mechanisms, we exploited serogroup A meningococcal strains and serum samples obtained from laboratory-confirmed Ethiopian patients with meningococcal disease. The 37 meningococcal isolates were sequenced, and their protein glycosylation (pgl) genotypes and protein glycosylation phenotypes were investigated in detail. An insertion sequence (IS1655) element in pglH reduced glycan variability in the majority of isolates, while phase variation strengthened glycan variability and microheterogeneity. Homologous recombination events within the pgl genes were identified in eight of the 37 isolates, and the phenotypic consequences ranged from none detected to altered glycoforms in two of the isolates in which the whole pgl locus was exchanged. Immunoblotting of sera against a complete panel of glycan-expressing mutant strains demonstrated that most of these patient sera had IgG antibodies against various neisserial protein glycan antigens. Furthermore, using a bactericidal assay comparing a wild-type meningococcal A strain and a glycosylation-null variant strain, we showed that these protein glycan antigens interfere with bactericidal killing by antibodies in patient sera. Altogether, we were largely able to link pgl genotype with glycosylation phenotype. Our study reveals that protein glycans seem to contribute to the ability of N. meningitidis to resist the bactericidal activity of human serum, possibly by masking protein epitopes important for bactericidal killing and thus protection against meningococcal disease. IMPORTANCE Bacterial meningitis is a serious global health problem, and one of the major causative organisms is Neisseria meningitidis. Extensive variability in protein glycan structure and antigenicity is due to phase variation of protein glycosylation genes and polymorphic gene content and function. The exact role(s) of glycosylation in Neisseria remains to be determined, but increasing evidence, supported by this study, suggests that glycan variability can be a strategy to escape the human immune system. The complexity of the O-linked protein glycosylation system requires further studies to fully comprehend how these bacteria utilize variation in pgl genes to produce such high glycoform diversity and to evade the human immune response.

Sculpting the Bacterial O-Glycoproteome: Functional Analyses of Orthologous Oligosaccharyltransferases with Diverse Targeting Specificities

Protein glycosylation systems are widely recognized in bacteria, including members of the genus Neisseria. In most bacterial species, the molecular mechanisms and evolutionary contexts underpinning target protein selection and the glycan repertoire remain poorly understood. Broad-spectrum O-linked protein glycosylation occurs in all human-associated species groups within the genus Neisseria, but knowledge of their individual glycoprotein repertoires is limited. Interestingly, PilE, the pilin subunit of the type IV pilus (Tfp) colonization factor, is glycosylated in Neisseria gonorrhoeae and Neisseria meningitidis but not in the deeply branching species N. elongata subsp. glycolytica. To examine this in more detail, we assessed PilE glycosylation status across the genus and found that PilEs of commensal clade species are not modified by the gonococcal PglO oligosaccharyltransferase. Experiments using PglO oligosaccharyltransferases from across the genus expressed in N. gonorrhoeae showed that although all were capable of broad-spectrum protein glycosylation, those from a deep-branching group of commensals were unable to support resident PilE glycosylation. Further glycoproteomic analyses of these strains using immunoblotting and mass spectrometry revealed other proteins differentially targeted by otherwise remarkably similar oligosaccharyltransferases. Finally, we generated pglO allelic chimeras that begin to localize PglO protein domains associated with unique substrate targeting activities. These findings reveal previously unappreciated differences within the protein glycosylation systems of highly related bacterial species. We propose that the natural diversity manifest in the neisserial protein substrates and oligosaccharyltransferases has significant potential to inform the structure-function relationships operating in these and related bacterial protein glycosylation systems. IMPORTANCE Although general protein glycosylation systems have been well recognized in prokaryotes, the processes governing their distribution, function, and evolution remain poorly understood. Here, we have begun to address these gaps in knowledge by comparative analyses of broad-spectrum O-linked protein glycosylation manifest in species within the genus Neisseria that strictly colonize humans. Using N. gonorrhoeae as a well-defined model organism in conjunction with comparative genomics, intraspecies gene complementation, and glycoprotein phenotyping, we discovered clear differences in both glycosylation susceptibilities and enzymatic targeting activities of otherwise largely conserved proteins. These findings reveal previously unappreciated differences within the protein glycosylation systems of highly related bacterial species. We propose that the natural diversity manifest within Neisseria species has significant potential to elucidate the structure-function relationships operating in these and related systems and to inform novel approaches to applied glycoengineering strategies.

A general approach to explore prokaryotic protein glycosylation reveals the unique surface layer modulation of an anammox bacterium

The enormous chemical diversity and strain variability of prokaryotic protein glycosylation makes their large-scale exploration exceptionally challenging. Therefore, despite the universal relevance of protein glycosylation across all domains of life, the understanding of their biological significance and the evolutionary forces shaping oligosaccharide structures remains highly limited. Here, we report on a newly established mass binning glycoproteomics approach that establishes the chemical identity of the carbohydrate components and performs untargeted exploration of prokaryotic oligosaccharides from large-scale proteomics data directly. We demonstrate our approach by exploring an enrichment culture of the globally relevant anaerobic ammonium-oxidizing bacterium Ca. Kuenenia stuttgartiensis. By doing so we resolve a remarkable array of oligosaccharides, which are produced by two seemingly unrelated biosynthetic routes, and which modify the same surface-layer protein simultaneously. More intriguingly, the investigated strain also accomplished modulation of highly specialized sugars, supposedly in response to its energy metabolism-the anaerobic oxidation of ammonium-which depends on the acquisition of substrates of opposite charges. Ultimately, we provide a systematic approach for the compositional exploration of prokaryotic protein glycosylation, and reveal a remarkable example for the evolution of complex oligosaccharides in bacteria.

Glycopeptide-Centric Approaches for the Characterization of Microbial Glycoproteomes

Protein glycosylation is increasingly recognized as a common class of modifications within microbial species that can shape protein functions and the proteome at large. Due to this, there is an increasing need for robust analytical methods, which allow for the identification and characterization of microbial glycopeptides from proteome samples in a high-throughput manner. Using affinity-based enrichment (either hydrophilicity or antibody-based approaches) glycopeptides can easily be separated from non-glycosylated peptides and analyzed using mass spectrometry. By utilizing multiple mass spectrometry fragmentation approaches and open searching-based bioinformatic techniques, novel glycopeptides can be identified and characterized without prior knowledge of the glycans used for glycosylation. Using these approaches, glycopeptides within samples can rapidly be identified as well as quantified to understand how glycosylation changes in response to stimuli or how changes in glycosylation systems impact the glycoproteome. This chapter outlines a set of robust protocols for the initial preparation, enrichment, and analysis of microbial glycopeptides for both qualitative and quantitative glycoproteomic studies. Using these approaches, glycosylation events can be easily identified by researchers without the need for extensive manual analysis of proteomic datasets.

Modelling evolutionary pathways for commensalism and hypervirulence in Neisseria meningitidis

Neisseria meningitidis, the meningococcus, resides exclusively in humans and causes invasive meningococcal disease (IMD). The population of N. meningitidis is structured into stable clonal complexes by limited horizontal recombination in this naturally transformable species. N. meningitidis is an opportunistic pathogen, with some clonal complexes, such as cc53, effectively acting as commensal colonizers, while other genetic lineages, such as cc11, are rarely colonizers but are over-represented in IMD and are termed hypervirulent. This study examined theoretical evolutionary pathways for pathogenic and commensal lineages by examining the prevalence of horizontally acquired genomic islands (GIs) and loss-of-function (LOF) mutations. Using a collection of 4850 genomes from the BIGSdb database, we identified 82 GIs in the pan-genome of 11 lineages (10 hypervirulent and one commensal lineage). A new computational tool, Phaser, was used to identify frameshift mutations, which were examined for statistically significant association with genetic lineage. Phaser identified a total of 144 frameshift loci of which 105 were shown to have a statistically significant non-random distribution in phase status. The 82 GIs, but not the LOF loci, were associated with genetic lineage and invasiveness using the disease carriage ratio metric. These observations have been integrated into a new model that infers the early events of the evolution of the human adapted meningococcus. These pathways are enriched for GIs that are involved in modulating attachment to the host, growth rate, iron uptake and toxin expression which are proposed to increase competition within the meningococcal population for the limited environmental niche of the human nasopharynx. We surmise that competition for the host mucosal surface with the nasopharyngeal microbiome has led to the selection of isolates with traits that enable access to cell types (non-phagocytic and phagocytic) in the submucosal tissues leading to an increased risk for IMD.

Burkholderia PglL enzymes are Serine preferring oligosaccharyltransferases which target conserved proteins across the Burkholderia genus

Glycosylation is increasingly recognised as a common protein modification within bacterial proteomes. While great strides have been made in identifying species that contain glycosylation systems, our understanding of the proteins and sites targeted by these systems is far more limited. Within this work we explore the conservation of glycoproteins and glycosylation sites across the pan-Burkholderia glycoproteome. Using a multi-protease glycoproteomic approach, we generate high-confidence glycoproteomes in two widely utilized B. cenocepacia strains, K56-2 and H111. This resource reveals glycosylation occurs exclusively at Serine residues and that glycoproteins/glycosylation sites are highly conserved across B. cenocepacia isolates. This preference for glycosylation at Serine residues is observed across at least 9 Burkholderia glycoproteomes, supporting that Serine is the dominant residue targeted by PglL-mediated glycosylation across the Burkholderia genus. Combined, this work demonstrates that PglL enzymes of the Burkholderia genus are Serine-preferring oligosaccharyltransferases that target conserved and shared protein substrates.

Hayes et al provide a glycosylation site focused analysis of the glycoproteome of two widely utilized B. cenocepacia strains, K56-2 and H111. This team demonstrates that within these glycoproteomes Serine is the sole residue targeted for protein glycosylation and that glycoproteins/glycosylation sites are highly conserved across B. cenocepacia isolates.

What Are We Missing by Using Hydrophilic Enrichment? Improving Bacterial Glycoproteome Coverage Using Total Proteome and FAIMS Analyses

Hydrophilic interaction liquid chromatography (HILIC) glycopeptide enrichment is an indispensable tool for the high-throughput characterization of glycoproteomes. Despite its utility, HILIC enrichment is associated with a number of shortcomings, including requiring large amounts of starting materials, potentially introducing chemical artifacts such as formylation when high concentrations of formic acid are used, and biasing/undersampling specific classes of glycopeptides. Here, we investigate HILIC enrichment-independent approaches for the study of bacterial glycoproteomes. Using three Burkholderia species (Burkholderia cenocepacia, Burkholderia Dolosa, and Burkholderia ubonensis), we demonstrate that short aliphatic O-linked glycopeptides are typically absent from HILIC enrichments, yet are readily identified in whole proteome samples. Using high-field asymmetric waveform ion mobility spectrometry (FAIMS) fractionation, we show that at high compensation voltages (CVs), short aliphatic glycopeptides can be enriched from complex samples, providing an alternative means to identify glycopeptide recalcitrant to hydrophilic-based enrichment. Combining whole proteome and FAIMS analyses, we show that the observable glycoproteome of these Burkholderia species is at least 25% larger than what was initially thought. Excitingly, the ability to enrich glycopeptides using FAIMS appears generally applicable, with the N-linked glycopeptides of Campylobacter fetus subsp. fetus also being enrichable at high FAIMS CVs. Taken together, these results demonstrate that FAIMS provides an alternative means to access glycopeptides and is a valuable tool for glycoproteomic analysis.

Allelic polymorphisms in a glycosyltransferase gene shape glycan repertoire in the O-linked protein glycosylation system of Neisseria

Glycosylation of multiple proteins via O-linkage is well documented in bacterial species of Neisseria of import to human disease. Recent studies of protein glycosylation (pgl) gene distribution established that related protein glycosylation systems occur throughout the genus including nonpathogenic species. However, there are inconsistencies between pgl gene status and observed glycan structures. One of these relates to the widespread distribution of pglG, encoding a glycosyltransferase that in Neisseria elongata subsp. glycolytica is responsible for the addition of di-N-acetyl glucuronic acid at the third position of a tetrasaccharide. Despite pglG residing in strains of N. gonorrhoeae, N. meningitidis and N. lactamica, no glycan structures have been correlated with its presence in these backgrounds. Moreover, PglG function in N. elongata subsp. glycolytica minimally requires UDP-glucuronic acid (GlcNAcA), and yet N. gonorrhoeae, N. meningitidis and N. lactamica lack pglJ, the gene whose product is essential for UDP-GlcNAcA synthesis. We examined the functionality of pglG alleles from species spanning the Neisseria genus by genetic complementation in N. elongata subsp. glycolytica. The results indicate that select pglG alleles from N. meningitidis and N. lactamica are associated with incorporation of an N-acetyl-hexosamine at the third position and reveal the potential for an expanded glycan repertoire in those species. Similar experiments using pglG from N. gonorrhoeae failed to find any evidence of function suggesting that those alleles are missense pseudogenes. Taken together, the results are emblematic of how allelic polymorphisms can shape bacterial glycosyltransferase function and demonstrate that such alterations may be constrained to distinct phylogenetic lineages.

Open Database Searching Enables the Identification and Comparison of Bacterial Glycoproteomes without Defining Glycan Compositions Prior to Searching

Mass spectrometry has become an indispensable tool for the characterization of glycosylation across biological systems. Our ability to generate rich fragmentation of glycopeptides has dramatically improved over the last decade yet our informatic approaches still lag behind. Although glycoproteomic informatics approaches using glycan databases have attracted considerable attention, database independent approaches have not. This has significantly limited high throughput studies of unusual or atypical glycosylation events such as those observed in bacteria. As such, computational approaches to examine bacterial glycosylation and identify chemically diverse glycans are desperately needed. Here we describe the use of wide-tolerance (up to 2000 Da) open searching as a means to rapidly examine bacterial glycoproteomes. We benchmarked this approach using N-linked glycopeptides of Campylobacter fetus subsp. fetus as well as O-linked glycopeptides of Acinetobacter baumannii and Burkholderia cenocepacia revealing glycopeptides modified with a range of glycans can be readily identified without defining the glycan masses before database searching. Using this approach, we demonstrate how wide tolerance searching can be used to compare glycan use across bacterial species by examining the glycoproteomes of eight Burkholderia species (B. pseudomallei; B. multivorans; B. dolosa; B. humptydooensis; B. ubonensis, B. anthina; B. diffusa; B. pseudomultivorans). Finally, we demonstrate how open searching enables the identification of low frequency glycoforms based on shared modified peptides sequences. Combined, these results show that open searching is a robust computational approach for the determination of glycan diversity within bacterial proteomes.

Open Database Searching Enables the Identification and Comparison of Bacterial Glycoproteomes without Defining Glycan Compositions Prior to Searching

Mass spectrometry has become an indispensable tool for the characterization of glycosylation across biological systems. Our ability to generate rich fragmentation of glycopeptides has dramatically improved over the last decade yet our informatic approaches still lag behind. Although glycoproteomic informatics approaches using glycan databases have attracted considerable attention, database independent approaches have not. This has significantly limited high throughput studies of unusual or atypical glycosylation events such as those observed in bacteria. As such, computational approaches to examine bacterial glycosylation and identify chemically diverse glycans are desperately needed. Here we describe the use of wide-tolerance (up to 2000 Da) open searching as a means to rapidly examine bacterial glycoproteomes. We benchmarked this approach using N-linked glycopeptides of Campylobacter fetus subsp. fetus as well as O-linked glycopeptides of Acinetobacter baumannii and Burkholderia cenocepacia revealing glycopeptides modified with a range of glycans can be readily identified without defining the glycan masses before database searching. Using this approach, we demonstrate how wide tolerance searching can be used to compare glycan use across bacterial species by examining the glycoproteomes of eight Burkholderia species (B. pseudomallei; B. multivorans; B. dolosa; B. humptydooensis; B. ubonensis, B. anthina; B. diffusa; B. pseudomultivorans). Finally, we demonstrate how open searching enables the identification of low frequency glycoforms based on shared modified peptides sequences. Combined, these results show that open searching is a robust computational approach for the determination of glycan diversity within bacterial proteomes.