Despite the fact that Campylobacter coli a major foodborne pathogen, its glycobiology has been largely neglected. The genetic makeup of the C. coli lipooligosaccharide biosynthesis locus was largely unknown until recently. C. coli harbors a large set of genes associated with lipooligosaccharide biosynthesis, including genes for several putative glycosyltransferases involved in the synthesis of sialylated lipooligosaccharide in Campylobacter jejuni. In the present study, C. coli was found to express lipooligosaccharide structures containing sialic acid and other nonulosonate acids. These findings have a strong impact on our understanding of C. coli ecology, host-pathogen interaction, and pathogenesis.
KEYWORDS: Campylobacter coli, GT-42, glycosyltransferases, lipooligosaccharides, nonulosonic acid, sialyltransferase
ABSTRACT
Campylobacter jejuni and Campylobacter coli are the most common causes of bacterial gastroenteritis in the world. Ganglioside mimicry by C. jejuni lipooligosaccharide (LOS) is the triggering factor of Guillain-Barré syndrome (GBS), an acute polyneuropathy. Sialyltransferases from glycosyltransferase family 42 (GT-42) are essential for the expression of ganglioside mimics in C. jejuni. Recently, two novel GT-42 genes, cstIV and cstV, have been identified in C. coli. Despite being present in ∼11% of currently available C. coli genomes, the biological role of cstIV and cstV is unknown. In the present investigation, mutation studies with two strains expressing either cstIV or cstV were performed and mass spectrometry was used to investigate differences in the chemical composition of LOS. Attempts were made to identify donor and acceptor molecules using in vitro activity tests with recombinant GT-42 enzymes. Here we show that CstIV and CstV are involved in C. coli LOS biosynthesis. In particular, cstV is associated with LOS sialylation, while cstIV is linked to the addition of a diacetylated nonulosonic acid residue.
IMPORTANCE Despite the fact that Campylobacter coli a major foodborne pathogen, its glycobiology has been largely neglected. The genetic makeup of the C. coli lipooligosaccharide biosynthesis locus was largely unknown until recently. C. coli harbors a large set of genes associated with lipooligosaccharide biosynthesis, including genes for several putative glycosyltransferases involved in the synthesis of sialylated lipooligosaccharide in Campylobacter jejuni. In the present study, C. coli was found to express lipooligosaccharide structures containing sialic acid and other nonulosonate acids. These findings have a strong impact on our understanding of C. coli ecology, host-pathogen interaction, and pathogenesis.
INTRODUCTION
Nonulosonic acids are a highly diverse family of nine-carbon α-keto acids. The most naturally abundant nonulosonic acids are the sialic acids (N-acetylneuraminic acid [Neu5Ac]) and derivatives (1). Initially thought to be only a deuterostome feature, sialic acids have been found in virulence-associated bacterial cell surface glycoconjugates such as lipopolysaccharides, capsules, pili, and flagella (2–4). Furthermore, these sialylated structures have been shown to influence pathogenesis through immune evasion, adhesion, and invasion (5, 6). Sialyltransferases catalyze the transfer of sialic acid from cytidine-5′-monophospho-N-acetylneuraminic acid (CMP-Neu5Ac) to an acceptor and are key in the synthesis of sialoglycoconjugates. Known sialyltransferases have been classified into seven distinct Carbohydrate-Active enZYmes Database (CAZy) glycosyltransferase (GT) families: GT-29, GT-38, GT-42, GT-52, GT-80, GT-97, and GT-100 (7). In Campylobacter jejuni, the most common cause of bacterial gastroenteritis, CMP-Neu5Ac biosynthesis (neuA, neuB, and neuC) and GT-42 (cstII and cstIII) genes are present in the lipooligosaccharide (LOS) biosynthesis locus classes A, B, C, M, R, and V (8–10). C. jejuni strains carrying one of these genetic classes synthesize LOS structures generally resembling gangliosides (9, 11–13). In some cases, infection with a C. jejuni strain expressing ganglioside-like LOS induces production of cross-reactive anti-ganglioside antibodies. This leads to the development of Guillain-Barré syndrome (GBS), an acute autoimmune polyradiculoneuropathy disease with an ∼5% mortality rate (14). In addition to the LOS-associated GT-42 genes, some C. jejuni isolates may also carry cstI, a GT-42 gene located in the capsule region. However, despite being an active α-2,3-monosialyltransferase, CstI has no known role in C. jejuni LOS, capsular polysaccharide (CPS), or flagella (15).
Campylobacter coli, the second most common cause of campylobacteriosis, has also been isolated from GBS patients (16–19). Nevertheless, the role of C. coli in GBS has largely remained unclear due to the seeming absence of key elements for the synthesis of ganglioside-like LOS (i.e., GT-42 and neuABC genes). Recently, three newly identified C. coli LOS-associated GT-42 genes were reported to be in the LOS biosynthesis locus: cstIV, cstV, and cstVI (10, 20, 21). While cstVI is generally found as a pseudogene, cstIV and cstV may potentially be involved in LOS biosynthesis (20). In this study, we sought to explore the role of CstIV and CstV in the LOS biosynthesis of two C. coli strains: C. coli 73, which possesses the only LOS locus class known to carry a cstIV gene (1, 2), and C. coli 76339, which was the first identified strain to have a LOS locus class IX and a cstV gene (1, 3).
(This article was submitted to an online preprint archive [22].)
RESULTS
CstIV and CstV are involved in LOS biosynthesis.
The LOSs of ΔcstIV and ΔcstV strains showed an increased mobility on silver-stained SDS-PAGE gels relative to those of the wild-type (WT) genes (Fig. 1). Thus, deletion of cstV in C. coli 76339 and cstIV in C. coli 73 resulted in truncated LOSs. The complemented cstV mutant exhibited two LOS bands on SDS-PAGE gels; the upper one corresponded to the WT LOS, and the lower-molecular-weight band corresponded to the truncated LOS (see Fig. S1 in the supplemental material). This suggests that partial restoration of the phenotype was achieved upon complementation in cis of ΔcstV-SR4. Deletion of the gene for the second putative sialyltransferase of C. coli 76339, cstI located in the capsular locus, had no effect on LOS mobility on the SDS-PAGE gel (data not shown).
FIG 1.
Electrophoresis mobility comparison of C. coli 76339 LOS of WT and mutant strains (A) and C. coli 73 LOS of WT and mutant strains (B). C. jejuni 81-176 was used as a reference. Lanes marked with an asterisk show C. jejuni 81-176 LOS samples treated with neuraminidase to indicate the expected mobility shift when a Neu5Ac residue is removed.
Though complementation of the ΔcstIV strain was infeasible, owing to the absence of a suitable locus, it is unlikely that the mobility shift resulting from cstIV disruption was due to a polar effect, as cstIV is followed by genes that are transcribed in the opposite direction. While cstIV and cstV were found to be involved in LOS biosynthesis, neuraminidase treatment had no impact on LOS mobility (Fig. S2).
C. coli 76339 neuB1 is involved in the biosynthesis of a CstV substrate.
Since no clear shift in the electrophoretic mobility of C. coli 76339 LOS was detected after neuraminidase treatment (Fig. S2), the gene for the putative sialic acid synthase, neuB1, located downstream from cstV (Fig. 2), was knocked out to determine whether CMP-Neu5Ac was the donor molecule for CstV. The LOS of 76339 ΔneuB1-SR2 showed a profile similar to those of 76339 ΔcstV-SF1 and 76339 ΔcstV-SR4 (Fig. S3). Thus, inactivation of neuB1 results in an LOS truncation seemingly similar to the one observed in ΔcstV strains, suggesting the potential involvement of neuB1 in the synthesis of the CstV donor.
FIG 2.
Schematic representation of C. coli 76339 LOS biosynthesis locus class IX (A) and C. coli 73 LOS biosynthesis locus class II (B) wild types and corresponding mutants. Gray arrows represent conserved genes across LOS locus classes as described in reference 10. Green arrows represent the GT-42-encoding genes. Blue arrows represent the genes involved in CMP-Neu5Ac biosynthesis. White arrows show accessory genes with unknown functions. Striped arrows represent pseudogenes in LOS locus class II. Orange arrows correspond to the antibiotic resistance cassette used for producing the mutants. Numbers correspond to the gene clusters as described in reference 20.
C. coli 73 lacks Neu5Ac biosynthesis gene orthologues. However, like other C. coli strains, it synthetizes other nonulosonic acids through the activity of the neuB2 and neuB3 genes. Deletion of neuB2 had no impact on C. coli 73 LOS electrophoretic mobility (data not shown), and despite repeated attempts, no viable C. coli 73 ΔneuB3 mutants were obtained.
CstIV and CstV are associated with nonulosonate residues in C. coli LOS.
LOS compositions predicted by liquid chromatography-mass spectrometry (LC-MS) for the C. coli 76339 WT and mutants are shown in Table 1. C. coli 76339 contains a core oligosaccharide linked via two 3-deoxy-d-manno-oct-2-ulosonic acid (Kdo) molecules to a lipid A molecule. The core oligosaccharide of C. coli 76339 is composed of heptoses (Hep), hexoses (Hex), hexosamines (HexNAc), and NeuAc. The resulting MS/MS spectrum of m/z 1,064.0 obtained from the O-deacylated LOS of WT C. coli 76339 revealed a single ion at m/z 1,214.4 corresponding to Hex3·Hep2·PEtn1·Kdo1 (Fig. 3a). The fragment ions at m/z 1,052.4 and 890.3 correspond to the additional loss of two Hex residues. The spectra also revealed ions that derived from lipid A, m/z 693.5 and m/z 388.3, corresponding to HexN3N1·P1·(C14:0 3-OH)2 and HexN3N1·(C14:0 3-OH)1, respectively. The observation of fragment ions at m/z 292.1 and 274.1 provided evidence for the presence of sialic acid on core region LOS. The MS/MS spectrum of precursor ion m/z 1,064.0 from C. coli 76339 ΔcstI is similar to that from C. coli 76339 WT, in which the diagnostic ions for sialic acid were detected at m/z 292.1 and 274.1 (Fig. 3b). However, no sialic acid was detected in the MS/MS spectrum of C. coli 76339 ΔcstV (Fig. 3c). Thus, cstV is associated with the presence of NeuAc, while cstI plays no role in C. coli 76339 LOS biosynthesis.
TABLE 1.
Data for LC-MS in positive mode and proposed compositions for O-deacylated LOSs of C. coli 76339 and corresponding cstI and cstV knockout mutants
| Strain genotype | Observed ion (m/z) |
Molecular mass (Da) |
Proposed composition |
||||||
|---|---|---|---|---|---|---|---|---|---|
| [M + 3H]3+ | [M + 2H + NH4]3+ | [M + 2H]2+ | [M + H + NH4]2+ | Observed | Calculateda | Core oligosaccharide | Phosphorylation in lipid A | Acylation in lipid A | |
| WT | 1,063.43 | 1,069.11 | 3,187.30 | 3,187.35 | Kdo2·Hep2·Hex4·HexNAc1·NeuAc1 | PPEtn | 3 N-(C14:0 3-OH) | ||
| 1,104.43 | 1,110.11 | 3,310.30 | 3,310.36 | Kdo2·Hep2·Hex4·HexNAc1·NeuAc1·PEtn1 | PPEtn | 3 N-(C14:0 3-OH) | |||
| 1,138.48 | 1,144.17 | 3,412.46 | 3,412.56 | Kdo2·Hep2·Hex4·HexNAc1·NeuAc1 | PPEtn | 4 N-(C14:0 3-OH) | |||
| 1,179.51 | 1,185.17 | 3,535.51 | 3,535.57 | Kdo2·Hep2·Hex4·HexNAc1·NeuAc1·PEtn1 | PPEtn | 4 N-(C14:0 3-OH) | |||
| ΔcstI | 1,063.44 | 1,069.11 | 3,187.31 | 3,187.35 | Kdo2·Hep2·Hex4·HexNAc1·NeuAc1 | PPEtn | 3 N-(C14:0 3-OH) | ||
| 1,104.44 | 1,110.12 | 3,310.33 | 3,310.36 | Kdo2·Hep2·Hex4·HexNAc1·NeuAc1·PEtn1 | PPEtn | 3 N-(C14:0 3-OH) | |||
| 1,138.51 | 1,144.18 | 3,412.52 | 3,412.56 | Kdo2·Hep2·Hex4·HexNAc1·NeuAc1 | PPEtn | 4 N-(C14:0 3-OH) | |||
| 1,179.51 | 1,185.20 | 3,535.55 | 3,535.57 | Kdo2·Hep2·Hex4·HexNAc1·NeuAc1·PEtn1 | PPEtn | 4 N-(C14:0 3-OH) | |||
| ΔcstV | 1,225.56 | 1,234.09 | 2,449.14 | 2,449.14 | Kdo2·Hep2·Hex2·HexNAc1 | P | 3 N-(C14:0 3-OH) | ||
| 1,287.08 | 1,295.56 | 2,572.13 | 2,572.15 | Kdo2·Hep2·Hex2·HexNAc1 | PPEtn | 3 N-(C14:0 3-OH) | |||
| 1,348.58 | 2,695.14 | 2,695.16 | Kdo2·Hep2·Hex2·HexNAc1·PEtn1 | PPEtn | 3 N-(C14:0 3-OH) | ||||
Isotope-monoisotopic mass units were used for calculation of molecular mass values based on proposed compositions as follows: HexN, 161.0688; HexN3N, 160.0848; C14:0 3-OH, 226.1933; PEtn, 123.0085; P, 79.9663; Kdo, 220.0583; Hep, 192.0634; Hex, 162.0528; HexNAc, 203.0794; NeuAc, 291.0954; and H2O, 18.0106.
FIG 3.
MS/MS spectra for the precursor ions of O-deacylated LOS from the C. coli 76339 WT, m/z 1,064.0 (a); C. coli 76339 ΔcstI, m/z 1,064.0 (b); and C. coli 76339 ΔcstV, m/z 1,295.4 (c).
The extracted MS spectra from O-deacylated LOSs of C. coli strain 73 WT and C. coli 73 ΔcstIV-SF3 are shown in Fig. S4a and Fig. S5a. A similar lipid A moiety was indicated by the MS/MS spectrum obtained from the O-deacylated LOS of C. coli strain 73 WT (Fig. 4a and Fig. S4b and c). The spectra also revealed ions that derived from lipid A, m/z 693.5 and m/z 388.3, corresponding to HexN3N1·P1·(C14:0 3-OH)2 and HexN3N1·(C14:0 3-OH)1, respectively. The observation of fragment ions at m/z 317.2 and 299.1 provided evidence for the presence of a residue with a molecular weight of 334.2 Da or 316.2 Da for its anhydrous form on core region LOS. These masses are consistent with free diacetamido-nonulosonate (diNAc-nonulosonate) and its conjugated form, respectively. However, these characteristic ions were not detected in the MS/MS spectrum of C. coli 73 ΔcstIV-SF3 (Fig. 4b and Fig. S5b and c), thus suggesting a role for cstIV in the biosynthesis of diNAc-nonulosonate LOS in C. coli 73.
FIG 4.
MS/MS spectra for the precursor ions of O-deacylated LOS from the C. coli 73 WT, m/z 1,072.4 (a), and C. coli 73 ΔcstIV-SF3, m/z 1,566.0 (b).
No sialyltransferase activity was detected for CstIV and CstV using in vitro assays.
To determine whether CstIV and CstV are capable of transferring Neu5Ac, C. coli crude protein extracts were tested for sialyltransferase activity using sugar acceptors labeled with either boron-dipyrromethene (BDP) or fluorescein (FCHASE). No sialyltransferase activity was detected in the crude protein extracts of C. coli 73 and C. coli 73 ΔcstIV-SF3. Monospecific α-2,3-sialyltransferase activity was detected in C. coli 76339 WT, C. coli 76339 ΔcstV-SF1, and C. coli 76339 ΔcstV-SR4 crude protein extracts using BDP-lactose (BDP-Lac) and BDP-N-acetyllactosamine (BDP-LacNAc) (Fig. S6). Since the genome of C. coli 76339 is known to carry the gene encoding the monospecific CstI α-2,3-sialyltransferase, the assays were also performed using C. coli 76339 ΔcstI-XR3 and C. coli 76339 ΔcstV-SRΔcstI-XR1 protein extracts (data not shown). No measurable enzymatic activity was detected with any of the tested acceptors in these ΔcstI strains, which demonstrated that the activity detected in C. coli 76339 WT was due to CstI. Furthermore, all tested recombinant CstIV and CstV showed no activity with any of the tested acceptors despite being expressed and tested under the same conditions as a C. jejuni CstII recombinant protein used as a positive control. Thus, these results suggest either that none of the tested glycans was a suitable acceptor or that another nonulosonate is the actual donor for these enzymes (data not shown).
DISCUSSION
C. jejuni GT-42 was the first glycosyltransferase from this CAZy family to be enzymatically and structurally characterized; CstII variants can be either monofunctional α2,3-sialyltransferases or bifunctional α2,3-/α2,8-sialyltransferases, while CstI and CstIII are monofunctional α2,3-sialyltransferases (15, 23–25). CstII and CstIII activity has been shown to be essential for the biosynthesis of ganglioside-like LOS structures, which are linked to GBS onset (12, 25). Despite the importance of GT-42 enzymes in virulence and pathogenesis (26–29), the activity of these glycosyltransferases has not been explored in other Campylobacter species. Approximately 29% of C. coli genomes have been found to contain a GT-42-encoding gene within the LOS biosynthesis locus (20). While cstVI was the most common LOS-associated GT-42-encoding gene in C. coli, in 99% of the analyzed genomes it was observed to be present as a pseudogene (20). Thus, we focused our attention on the role of cstIV and cstV in LOS biosynthesis. Until recently, cstV had been identified solely in the genome of C. coli 76339 (21). However, in a systematic screen of publicly available C. coli genomes, several cstV positive strains were identified (20). Since in vitro assays have been previously used to determine the activity of C. jejuni GT-42 enzymes (9, 15), a similar approach was attempted to define CstIV and CstV activity. C. coli 76339 crude protein extracts were tested for sialyltransferase activity, as Neu5Ac had been previously detected in the strain’s LOS (21). Monofunctional sialyltransferase activity was initially observed but was found to be due to CstI activity. As in C. jejuni, C. coli 76339 cstI is located outside the LOS biosynthesis locus and encodes an α2,3-sialyltransferase which has no role in LOS biosynthesis (15, 21). Although transcriptomic analysis showed polycistronic expression of LOS biosynthesis genes, indicating the active expression of cstV (data not shown), no sialyltransferase activity was detected on the protein extracts of the cstI mutant strain. Inactivation of neuB1 or cstV resulted in identical LOS electrophoretic profiles. Additionally, LC-MS analysis showed that the inactivation of cstV resulted in the loss of 2 Hex residues and 1 NeuAc residue. Nevertheless, recombinant CstV exhibited no detectable activity with any of the tested acceptors. Thus, it is very likely that cstV is associated with C. coli 76339 LOS sialylation. Yet further studies are required to identify CstV natural acceptor and corroborate its activity in vitro.
After cstVI, cstIV is the most common orthologue, being present in ∼38% of the genomes positive for a LOS-associated GT-42. Previously, no evidence of Neu5Ac had been found in the LOSs of strains containing a cstIV orthologue (30). This was to be expected, as Neu5Ac biosynthesis genes are rarely present in strains carrying cstIV (20). Furthermore, no sialyltransferase activity was detected either in C. coli 73 protein extracts or in recombinant CstIV. Nevertheless, deletion of cstIV in C. coli 73 resulted in a truncated LOS, thus suggesting a link between cstIV and LOS biosynthesis. Sequence alignment of CstIV with previously characterized GT-42 sialyltransferases revealed numerous amino acid substitutions at conserved positions (Fig. S7) (31). Additionally, superimposition of CstIV on the C. jejuni CstII structure identified various substitutions at amino acids involved in substrate interactions (24, 32–34). Interestingly, most substitutions predicted to impact CstIV activity were in the amino acids associated with CMP-Neu5Ac, particularly with the Neu5Ac moiety. Moreover, these substitutions were conserved in multiple CstIV orthologues (24, 33, 34). Together, the results pointed at the possibility of an alternative sugar donor for CstIV. Detection of a diNAc-nonulosonate residue in C. coli 73 WT LOS and its absence in C. coli 73 ΔcstIV-SF3 prompted an investigation on genes potentially linked to the synthesis of this residue. In C. coli, neuB2 (ptmC, legI) and neuB3 (pseI) are conserved flagellum glycosylation genes involved in the synthesis of legionaminic and pseudaminic acid derivatives, respectively (35–41). Deletion of neuB2 had no impact on C. coli 73 LOS electrophoretic mobility, implying that neuB2 is not involved in the synthesis of CstIV donor. Despite repeated attempts, no viable C. coli 73 ΔneuB3 mutants were obtained. Although neuB3 deletion has been successful in C. coli VC167, disruption of flagellin glycosylation and the potential truncation of the LOS might have resulted in a lethal phenotype for C. coli 73 (41). In sum, it is tempting to speculate that the diNAc-nonulosonate residue in the C. coli 73 WT corresponds to pseudaminic acid. However, the nature of this residue cannot be inferred from MS/MS spectra alone since many diNAc-nonulosonate variants have been identified (42).
In conclusion, although we could not determine the complete structures of the LOS outer cores of C. coli 73 and C. coli 76339, we have established that they both contain nonulosonic acid. We have also unequivocally demonstrated that CstIV and CstV are involved in the synthesis of LOS in their respective strains and, more specifically, that they are responsible of the transfer of a nonulosonic acid residue to the outer core.
MATERIALS AND METHODS
Bacterial strains, plasmids, and growth conditions.
Bacterial strains used in this study are listed in Table 2. Two C. coli strains, expressing either cstIV or cstV, were selected. Figure 2 shows a schematic representation of the LOS locus and the position of the insertion of the antibiotic resistance cassette for the mutational studies. C. coli 73 possesses LOS locus class II, which contains a copy of cstIV and lacks sialic acid biosynthesis genes neuABC or other copies of putative sialyltransferases. However, strain 73 possesses the conserved biosynthesis pathways for both legionaminic acid (including legionaminic acid synthetase gene neuB2) and pseudaminic acid (including pseudaminic acid synthetase gene neuB3) in the flagellum glycosylation region (30). C. coli 76339 harbors two copies of putative sialyltransferase genes: cstV as part of LOS locus class IX (Fig. 2) and a C. jejuni cstI orthologue located in the capsule region (20, 21). In addition to the conserved legionaminic and pseudaminic acid biosynthesis pathways, as in C. coli 73, C. coli 76339 possesses all the genes for the biosynthesis of sialic acid (neuAB1C) within the LOS locus class IX (Fig. 2). C. coli cultivation and DNA isolation were carried out as previously described, unless specified otherwise (21).
TABLE 2.
Bacterial strains used in this study
| Strain | Genotype and/or phenotype | Reference or source |
|---|---|---|
| C. coli | ||
| 76339 | cstI cstV neuB | 21 |
| 76339 ΔcstV-SF1 | cstI ΔcstV Ery neuB | This study |
| 76339 ΔcstV-SR4 | cstI ΔcstV Ery neuB | This study |
| 76339 ΔneuB-SR1 | cstI ΔneuB Ery cstV | This study |
| 76339 ΔcstI-XR3 | ΔcstI CAT cstV neuB | This study |
| 76339 ΔcstV-SR4 ΔcstI-XR1 | ΔcstI CAT ΔcstV Ery neuB | This study |
| 76339 ΔneuB-SR2 | cstI ΔneuB Ery cstV | This study |
| 76339 ΔcstV-SR4 Δggt:cstV-2 | cstI ΔcstV Ery neuB Δggt:cstV CAT | This study |
| 73 | cstIV | 30 |
| 73 ΔcstIV-SF3 | ΔcstIV Ery | This study |
| 73 ΔneuB2 | ΔneuB2 Ery | This study |
| C. jejuni 81-176 | ||
| E. coli AD202 | 47 |
Construction of ΔcstIV, ΔcstV, ΔcstI, and ΔneuB mutants.
Chromosomal mutant strains of C. coli 76339 (20, 21) and C. coli 73 (20, 30) were generated by homologous recombination with suicide vectors containing genes inactivated by the insertion of an antibiotic resistance cassette. All recombinant plasmids and primers are shown in the supplemental material (Fig. S8 to S13). The genes cstIV, cstV, neuB1, neuB2, and neuB3 were inactivated by the insertion of an erythromycin resistance cassette (EryC) (43), while cstI was disrupted with a chloramphenicol acetyltransferase cassette (CAT) (44). The inactivation of cstV in C. coli 76339 was performed by inserting the eryC cassette either in the direction of the gene (SR) or in the opposite direction (SF). Preparation of electrocompetent cells and transformation were done as previously described (44). Selection of the mutants was done on nutrient blood agar (NBA) supplemented with either 10 μg ml−1 of erythromycin or 12.5 μg ml−1 of chloramphenicol. Homologous recombination of all mutants was verified by PCR. Figure 2 shows a summary of the mutations performed in the LOS locus of C. coli strains 76339 and 73.
Complementation studies.
Complementation of C. coli 76339 ΔcstV-SR4 was done in cis by integration of cstV under the control of the active promoter of the gamma glutamyltranspeptidase gene (ggt). ggt is an accessory gene in C. coli and has no role in LOS biosynthesis. Additionally, the ggt locus is located far from the LOS locus and its deletion does not induce a loss in bacterial viability. The suicide vector containing inactivated ggt by the insertion of a cstV and CAT (pGEM-ggt-cstV-CAT) (Fig. S14) was used to transform C. coli 76339 ΔcstV-SR4 electrocompetent cells as described above. Transformants were selected on NBA supplemented with 12.5 μg ml−1 of chloramphenicol. Homologous recombination of mutants was verified by testing for GGT activity as before (45). Complementation of cstIV was not possible due to the absence of a suitable locus.
LOS silver staining.
LOS profiles were assessed by silver staining as described earlier (30). Additionally, LOS sensitivity to neuraminidase was assessed by treating crude LOS with 2 IU/ml of Clostridium perfringens neuraminidase (Sigma-Aldrich) overnight at 37°C.
Mass spectrometry analysis of C. coli LOS composition.
Following treatment with 1% formaldehyde in phosphate-buffered saline (PBS; pH 7.4), C. coli cell pellets were washed three times in PBS and lyophilized. Then cells were dehydrated by a sequence of 2 washes in each of 70% ethanol (in PBS), 100% ethanol, and 100% acetone. The dehydrated cells were treated with proteinase K, RNase A, and DNase I as previously described (46). Digested cells were then treated with hydrazine to cleave O-linked fatty acids (46). The O-deacylated LOS samples were analyzed by LC-MS by coupling a Waters Premier Q-TOF with an Agilent 1260 capillary LC system. Mass spectrometry was operated in positive-ion detection mode. Liquid chromatography separation was done on an Agilent Eclipse XDB C8 column (5 μm; 50 by 1 mm). The flow rate was 20 μl/min. Solvent A was aqueous 0.2% formic acid–0.028% ammonia; solvent B was isopropanol with 0.2% formic acid–0.028% ammonia. The following gradient was used: 0 to 2 min with 10% solvent B, 2 to 16 min of a linear gradient to 85% solvent B, 16 to 25 min of 85% solvent B, 25 to 30 min, and equilibration at 10% solvent B.
Sialyltransferase activity test in C. coli protein extracts.
To test for sialyltransferase activity, C. coli 76339 and 73 were grown for 16 h in nutrient broth 2 (Oxoid) (100 rpm, microaerobic atmosphere, and 37°C). Cells were harvested by centrifugation (10,000 × g for 15 min at 4°C) and resuspended in 50 mM HEPES (pH 7.5) containing a protease inhibitor cocktail (Sigma). Cells were then lysed by mechanical disruption and debris was removed by centrifugation (10,000 × g for 15 min at 4°C). Sialyltransferase activity of protein extracts was tested on boron-dipyrromethene (BDP)-labeled Lac, LacNAc, and 3′ sialyllactose or fluorescein (FCHASE)-labeled α-GalNAc, β-GalNAc, GM3, α-Gal, β-GlcNAc, α-Glc, β-Glc, or Hep-Hep-Glc as the acceptor. Reactions were performed at 37°C in 10-μl volumes containing 50 mM HEPES (pH 7.5), 10 mM MgCl2, 1 mM CMP-NeuAc, 0.5 mM labeled acceptor, and 6 μl of extract. To stop enzymatic reactions, an equal volume of 80% acetonitrile was added. Enzymatic activity was assessed by thin-layer chromatography on silica using a solvent system of ethyl acetate-methanol-water-acetic acid (4:2:1:0.1).
Expression and activity of recombinant C. coli GT-42 enzymes.
cstIV from C. coli 73 and cstV from C. coli 76339 were amplified and ligated to pCW and pCW-MalET plasmids (47). Ligation products were then electroporated in E. coli 10β for plasmid amplification. After sequence confirmation, plasmids were electroporated into E. coli AD202 or BL21 for protein expression. Cells containing the protein expression vectors were grown in 200 ml of 2YT medium supplemented with 150 μg/ml of ampicillin and 0.2% glucose at 25°C with shaking at 250 rpm. After an A600 of ∼0.6 was reached, protein overexpression was induced with 0.5 mM isopropyl-β-d-thiogalactopyranoside (IPTG) and cultures were further incubated for 16 h. Cells were harvested by centrifugation (10,000 × g for 15 min at 4°C), and crude protein extracts were run in 12% SDS-PAGE gels and stained with Coomassie blue to verify overexpression. In addition, cstIV and cstV genes were synthesized with a T7 promoter and a ribosome binding site upstream of the coding sequence and a T7 terminator downstream from the stop codon (Thermo Scientific). Synthesized products were inserted into pMA-T vector backbone and proteins were synthesized using the cell-free PURExpress in vitro protein synthesis kit (New England Biolabs Inc.). Recombinant proteins were screened for sialyltransferase activity as described above.
Supplementary Material
ACKNOWLEDGMENTS
This research project was supported by University of Helsinki research grant no. 313/51/2013 and a travel grant from the Walter Ehrström Foundation. A.K. was supported by the Microbiology and Biotechnology graduate program from the University of Helsinki.
We thank Marja-Liisa Hänninen for providing the strains and Arnoud H. M. van Vliet for providing the erythromycin resistance cassette. We thank Denis Brochu for help with the preparation of the samples for mass spectrometry analysis and data presentation.
A.K. and M.R. designed and coordinated the study. A.K. generated all C. coli mutants. A.K., M.G., and W.W. participated in enzymatic assays. J.S. and J.L. performed LC-MS analysis. J.S., J.L., and M.G. interpreted LC-MS data. A.K. drafted the manuscript. All authors contributed to data interpretation, critically reviewed the manuscript, and approved the final version as submitted.
We declare that we have no competing interests.
The positions and opinions presented in this article are those of the authors alone and are not intended to represent the views or scientific works of EFSA.
Footnotes
Supplemental material for this article may be found at https://doi.org/10.1128/JB.00759-18.
REFERENCES
- 1.Angata T, Varki A. 2002. Chemical diversity in the sialic acids and related α-keto acids: an evolutionary perspective. Chem Rev 102:439–470. doi: 10.1021/cr000407m. [DOI] [PubMed] [Google Scholar]
- 2.Gamian A, Jones C, Lipiński T, Korzeniowska-Kowal A, Ravenscroft N. 2000. Structure of the sialic acid-containing O-specific polysaccharide from Salmonella enterica serovar Toucra O48 lipopolysaccharide. Eur J Biochem 267:3160–3167. doi: 10.1046/j.1432-1327.2000.01335.x. [DOI] [PubMed] [Google Scholar]
- 3.Hood DW, Makepeace K, Deadman ME, Rest RF, Thibault P, Martin A, Richards JC, Moxon ER. 1999. Sialic acid in the lipopolysaccharide of Haemophilus influenzae: strain distribution, influence on serum resistance and structural characterization. Mol Microbiol 33:679–692. doi: 10.1046/j.1365-2958.1999.01509.x. [DOI] [PubMed] [Google Scholar]
- 4.Kahler CM, Martin LE, Shih GC, Rahman MM, Carlson RW, Stephens DS. 1998. The (α2→8)-linked polysialic acid capsule and lipooligosaccharide structure both contribute to the ability of serogroup B Neisseria meningitidis to resist the bactericidal activity of normal human serum. Infect Immun 66:5939–5947. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Lo H, Tang CM, Exley RM. 2009. Mechanisms of avoidance of host immunity by Neisseria meningitidis and its effect on vaccine development. Lancet Infect Dis 9:418–427. doi: 10.1016/S1473-3099(09)70132-X. [DOI] [PubMed] [Google Scholar]
- 6.Louwen R, Heikema A, van Belkum A, Ott A, Gilbert M, Ang W, Endtz HP, Bergman MP, Nieuwenhuis EE. 2008. The sialylated lipooligosaccharide outer core in Campylobacter jejuni is an important determinant for epithelial cell invasion. Infect Immun 76:4431–4438. doi: 10.1128/IAI.00321-08. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Lombard V, Golaconda Ramulu H, Drula E, Coutinho PM, Henrissat B. 2014. The carbohydrate-active enzymes database (CAZy) in 2013. Nucleic Acids Res 42:D490–D495. doi: 10.1093/nar/gkt1178. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Parker CT, Gilbert M, Yuki N, Endtz HP, Mandrell RE. 2008. Characterization of lipooligosaccharide-biosynthetic loci of Campylobacter jejuni reveals new lipooligosaccharide classes: evidence of mosaic organizations. J Bacteriol 190:5681–5689. doi: 10.1128/JB.00254-08. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Gilbert M, Karwaski M, Bernatchez S, Young NM, Taboada E, Michniewicz J, Cunningham A, Wakarchuk WW. 2002. The genetic bases for the variation in the lipo-oligosaccharide of the mucosal pathogen, Campylobacter jejuni. J Biol Chem 277:327–337. doi: 10.1074/jbc.M108452200. [DOI] [PubMed] [Google Scholar]
- 10.Richards VP, Lefébure T, Pavinski Bitar PD, Stanhope MJ. 2013. Comparative characterization of the virulence gene clusters (lipooligosaccharide [LOS] and capsular polysaccharide [CPS]) for Campylobacter coli, Campylobacter jejuni subsp. jejuni and related Campylobacter species. Infect Genet Evol 14:200–213. doi: 10.1016/j.meegid.2012.12.010. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Godschalk PCR, Kuijf ML, Li J, St Michael F, Ang CW, Jacobs BC, Karwaski M, Brochu D, Moterassed A, Endtz HP, van Belkum A, Gilbert M. 2007. Structural characterization of Campylobacter jejuni lipooligosaccharide outer cores associated with Guillain-Barré and Miller Fisher syndromes. Infect Immun 75:1245–1254. doi: 10.1128/IAI.00872-06. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Godschalk PCR, Heikema AP, Gilbert M, Komagamine T, Ang CW, Glerum J, Brochu D, Li J, Yuki N, Jacobs BC, van Belkum A, Endtz HP. 2004. The crucial role of Campylobacter jejuni genes in anti-ganglioside antibody induction in Guillain-Barré syndrome. J Clin Invest 114:1659–1665. doi: 10.1172/JCI200415707. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Houliston RS, Vinogradov E, Dzieciatkowska M, Li J, St Michael F, Karwaski M, Brochu D, Jarrell HC, Parker CT, Yuki N, Mandrell RE, Gilbert M. 2011. Lipooligosaccharide of Campylobacter jejuni: similarity with multiple types of mammalian glycans beyond gangliosides. J Biol Chem 286:12361–12370. doi: 10.1074/jbc.M110.181750. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Yuki N. 2005. Carbohydrate mimicry: a new paradigm of autoimmune diseases. Curr Opin Immunol 17:577–582. doi: 10.1016/j.coi.2005.09.004. [DOI] [PubMed] [Google Scholar]
- 15.Gilbert M, Brisson J, Karwaski M, Michniewicz J, Cunningham A, Wu Y, Young NM, Wakarchuk WW. 2000. Biosynthesis of ganglioside mimics in Campylobacter jejuni OH4384: identification of the glycosyltransferase genes, enzymatic synthesis of model compounds, and characterization of nanomole amounts by 600-MHz 1H and 13C NMR analysis. J Biol Chem 275:3896–3906. doi: 10.1074/jbc.275.6.3896. [DOI] [PubMed] [Google Scholar]
- 16.Wulffen H. v, Hartard C, Scharein E. 1994. Seroreactivity to Campylobacter jejuni and gangliosides in patients with Guillain-Barré syndrome. J Infect Dis 170:828–833. doi: 10.1093/infdis/170.4.828. [DOI] [PubMed] [Google Scholar]
- 17.Funakoshi K, Koga M, Takahashi M, Hirata K, Yuki N. 2006. Campylobacter coli enteritis and Guillain–Barré syndrome: no evidence of molecular mimicry and serological relationship. J Neurol Sci 246:163–168. doi: 10.1016/j.jns.2006.02.010. [DOI] [PubMed] [Google Scholar]
- 18.Bersudsky M, Rosenberg P, Rudensky B, Wirguin I. 2000. Lipopolysaccharides of a Campylobacter coli isolate from a patient with Guillain-Barré syndrome display ganglioside mimicry. Neuromuscul Disord 10:182–186. doi: 10.1016/S0960-8966(99)00106-6. [DOI] [PubMed] [Google Scholar]
- 19.van Belkum A, Jacobs B, van Beek E, Louwen R, van Rijs W, Debruyne L, Gilbert M, Li J, Jansz A, Megraud F, Endtz H. 2009. Can Campylobacter coli induce Guillain-Barré syndrome? Eur J Clin Microbiol Infect Dis 28:557–560. doi: 10.1007/s10096-008-0661-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Culebro A, Machado MP, Carriço JA, Rossi M. 2017. Origin, evolution, and distribution of the molecular machinery for biosynthesis of sialylated lipooligosaccharide structures in Campylobacter coli. bioRxiv doi: 10.1101/225771. [DOI] [PMC free article] [PubMed]
- 21.Skarp-de Haan CPA, Culebro A, Schott T, Revez J, Schweda EKH, Hänninen M-L, Rossi M. 2014. Comparative genomics of unintrogressed Campylobacter coli clades 2 and 3. BMC Genomics 15:129. doi: 10.1186/1471-2164-15-129. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Kolehmainen A, Rossi M, Stupak J, Li J, Gilbert M, Wakarchuk W. 2018. Genetics behind the biosynthesis of nonulosonic acid containing lipooligosaccharides in Campylobacter coli. bioRxiv 10.1101/254235. [DOI] [PMC free article] [PubMed]
- 23.Chiu CPC, Lairson LL, Gilbert M, Wakarchuk WW, Withers SG, Strynadka NCJ. 2007. Structural analysis of the α-2,3-sialyltransferase Cst-I from Campylobacter jejuni in apo and substrate-analogue bound forms. Biochemistry 46:7196–7204. doi: 10.1021/bi602543d. [DOI] [PubMed] [Google Scholar]
- 24.Chiu CPC, Watts AG, Lairson LL, Gilbert M, Lim D, Wakarchuk WW, Withers SG, Strynadka NCJ. 2004. Structural analysis of the sialyltransferase CstII from Campylobacter jejuni in complex with a substrate analog. Nat Struct Mol Biol 11:163–170. doi: 10.1038/nsmb720. [DOI] [PubMed] [Google Scholar]
- 25.Guerry P, Ewing CP, Hickey TE, Prendergast MM, Moran AP. 2000. Sialylation of lipooligosaccharide cores affects immunogenicity and serum resistance of Campylobacter jejuni. Infect Immun 68:6656–6662. doi: 10.1128/IAI.68.12.6656-6662.2000. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Mortensen NP, Kuijf ML, Ang CW, Schiellerup P, Krogfelt KA, Jacobs BC, van Belkum A, Endtz HP, Bergman MP. 2009. Sialylation of Campylobacter jejuni lipo-oligosaccharides is associated with severe gastro-enteritis and reactive arthritis. Microb Infect 11:988–994. doi: 10.1016/j.micinf.2009.07.004. [DOI] [PubMed] [Google Scholar]
- 27.Kuijf ML, Samsom JN, van Rijs W, Bax M, Huizinga R, Heikema AP, van Doorn PA, van Belkum A, van Kooyk Y, Burgers PC, Luider TM, Endtz HP, Nieuwenhuis EES, Jacobs BC. 2010. TLR4-mediated sensing of Campylobacter jejuni by dendritic cells is determined by sialylation. J Immunol 185:748–755. doi: 10.4049/jimmunol.0903014. [DOI] [PubMed] [Google Scholar]
- 28.Huizinga R, van Rijs W, Bajramovic JJ, Kuijf ML, Laman JD, Samsom JN, Jacobs BC. 2013. Sialylation of Campylobacter jejuni endotoxin promotes dendritic cell-mediated B cell responses through CD14-dependent production of IFN-β and TNF-α. J Immunol 191:5636–5645. doi: 10.4049/jimmunol.1301536. [DOI] [PubMed] [Google Scholar]
- 29.Huizinga R, Easton AS, Donachie AM, Guthrie J, van Rijs W, Heikema A, Boon L, Samsom JN, Jacobs BC, Willison HJ, Goodyear CS. 2012. Sialylation of Campylobacter jejuni lipo-oligosaccharides: impact on phagocytosis and cytokine production in mice. PLoS One 7:e34416. doi: 10.1371/journal.pone.0034416. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Culebro A, Revez J, Pascoe B, Friedmann Y, Hitchings MD, Stupak J, Sheppard SK, Li J, Rossi M. 2016. Large sequence diversity within the biosynthesis locus and common biochemical features of Campylobacter coli lipooligosaccharides. J Bacteriol 198:2829–2840. doi: 10.1128/JB.00347-16. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Schur MJ, Lameignere E, Strynadka NC, Wakarchuk WW. 2012. Characterization of α2,3- and α2,6-sialyltransferases from Helicobacter acinonychis. Glycobiology 22:997–1006. doi: 10.1093/glycob/cws071. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Chan PHW, Lairson LL, Lee HJ, Wakarchuk WW, Strynadka NCJ, Withers SG, McIntosh LP. 2009. NMR spectroscopic characterization of the sialyltransferase CstII from Campylobacter jejuni: histidine 188 is the general base. Biochemistry 48:11220–11230. doi: 10.1021/bi901606n. [DOI] [PubMed] [Google Scholar]
- 33.Lee HJ, Lairson LL, Rich JR, Lameignere E, Wakarchuk WW, Withers SG, Strynadka NCJ. 2011. Structural and kinetic analysis of substrate binding to the sialyltransferase Cst-II from Campylobacter jejuni. J Biol Chem 286:35922–35932. doi: 10.1074/jbc.M111.261172. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Aharoni A, Thieme K, Chiu CPC, Buchini S, Lairson LL, Chen H, Strynadka NCJ, Wakarchuk WW, Withers SG. 2006. High-throughput screening methodology for the directed evolution of glycosyltransferases. Nat Methods 3:609–614. doi: 10.1038/nmeth899. [DOI] [PubMed] [Google Scholar]
- 35.Schoenhofen IC, McNally DJ, Brisson J, Logan SM. 2006. Elucidation of the CMP-pseudaminic acid pathway in Helicobacter pylori: synthesis from UDP-N-acetylglucosamine by a single enzymatic reaction. Glycobiology 16:8C–14C. doi: 10.1093/glycob/cwl010. [DOI] [PubMed] [Google Scholar]
- 36.Schoenhofen IC, Vinogradov E, Whitfield DM, Brisson J, Logan SM. 2009. The CMP-legionaminic acid pathway in Campylobacter: biosynthesis involving novel GDP-linked precursors. Glycobiology 19:715–725. doi: 10.1093/glycob/cwp039. [DOI] [PubMed] [Google Scholar]
- 37.Sundaram A, Pitts L, Muhammad K, Wu J, Betenbaugh M, Woodard R, Vann W. 2004. Characterization of N-acetylneuraminic acid synthase isoenzyme 1 from Campylobacter jejuni. Biochem J 383:83–89. http://www.biochemj.org/content/383/1/83.abstract. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Linton D, Karlyshev AV, Hitchen PG, Morris HR, Dell A, Gregson NA, Wren BW. 2000. Multiple N‐acetyl neuraminic acid synthetase (neuB) genes in Campylobacter jejuni: identification and characterization of the gene involved in sialylation of lipo‐oligosaccharide. Mol Microbiol 35:1120–1134. doi: 10.1046/j.1365-2958.2000.01780.x. [DOI] [PubMed] [Google Scholar]
- 39.Lewis AL, Desa N, Hansen EE, Knirel YA, Gordon JI, Gagneux P, Nizet V, Varki A. 2009. Innovations in host and microbial sialic acid biosynthesis revealed by phylogenomic prediction of nonulosonic acid structure. Proc Natl Acad Sci U S A 106:13552–13557. doi: 10.1073/pnas.0902431106. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.McNally DJ, Aubry AJ, Hui JP, Khieu NH, Whitfield D, Ewing CP, Guerry P, Brisson JR, Logan SM, Soo EC. 2007. Targeted metabolomics analysis of Campylobacter coli VC167 reveals legionaminic acid derivatives as novel flagellar glycans. J Biol Chem 282:14463–14475. doi: 10.1074/jbc.M611027200. [DOI] [PubMed] [Google Scholar]
- 41.Logan SM, Kelly JF, Thibault P, Ewing CP, Guerry P. 2002. Structural heterogeneity of carbohydrate modifications affects serospecificity of Campylobacter flagellins. Mol Microbiol 46:587–597. doi: 10.1046/j.1365-2958.2002.03185.x. [DOI] [PubMed] [Google Scholar]
- 42.Varki A, Schnaar RL, Schauer R. 2017. Sialic acids and other nonulosonic acids, p 179–195. In Varki A, Cummings RD, Esko JD, Stanley P, Hart GW, Aebi M, Darvill AG, Kinoshita T, Packer NH, Prestegard JH, Schnaar RL, Seeberger PH (ed), Essentials of glycobiology, 3rd ed The Consortium of Glycobiology Editors, La Jolla, CA. [Google Scholar]
- 43.Thomas MT, Shepherd M, Poole RK, van Vliet AHM, Kelly DJ, Pearson BM. 2011. Two respiratory enzyme systems in Campylobacter jejuni NCTC 11168 contribute to growth on L-lactate. Environ Microbiol 13:48–61. doi: 10.1111/j.1462-2920.2010.02307.x. [DOI] [PubMed] [Google Scholar]
- 44.Olkkola S, Culebro A, Juntunen P, Hänninen M, Rossi M. 2016. Functional genomics in Campylobacter coli identified a novel streptomycin resistance gene located in a hypervariable genomic region. Microbiology 162:1157–1166. doi: 10.1099/mic.0.000304. [DOI] [PubMed] [Google Scholar]
- 45.de Haan CPA, Llarena A, Revez J, Hänninen M. 2012. Association of Campylobacter jejuni metabolic traits with multilocus sequence types. Appl Environ Microbiol 78:5550–5554. doi: 10.1128/AEM.01023-12. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46.Li J, Koga M, Brochu D, Yuki N, Chan K, Gilbert M. 2005. Electrophoresis-assisted open-tubular liquid chromatography/mass spectrometry for the analysis of lipooligosaccharide expressed by Campylobacter jejuni. Electrophoresis 26:3360–3368. doi: 10.1002/elps.200500145. [DOI] [PubMed] [Google Scholar]
- 47.Bernatchez S, Gilbert M, Blanchard M, Karwaski M, Li J, DeFrees S, Wakarchuk WW. 2007. Variants of the β1,3-galactosyltransferase CgtB from the bacterium Campylobacter jejuni have distinct acceptor specificities. Glycobiology 17:1333–1343. doi: 10.1093/glycob/cwm090. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.




