Molecular and in silico typing of the lipooligosaccharide biosynthesis gene cluster in Campylobacter jejuni and Campylobacter coli

Abstract

The extensive genetic variation in the lipooligosaccharide (LOS) core biosynthesis gene cluster has led to the development of a classification system; with 8 classes (I-VIII) for Campylobacter coli (C. coli) LOS region and with 23 classes (A-W) or four groups (1–4) for Campylobacter jejuni (C. jejuni) LOS region. PCR based LOS locus type identification for C. jejuni clinical isolates from a UK hospital as well as in silico LOS locus analysis for C. jejuni and C. coli genome sequences from GenBank was carried out to determine the frequencies of various LOS genotypes in C. jejuni and C. coli. Analysis of LOS gene content in 60 clinical C. jejuni isolates and 703 C. jejuni genome sequences revealed that class B (Group 1) was the most abundant LOS class in C. jejuni. The hierarchy of C. jejuni LOS group prevalence (group 1 > group 2 > group 3 > group 4) as well as the hierarchy of the frequency of C. jejuni LOS classes present within the group 1 (B > C > A > R > M > V), group 2 (H/P > O > E > W), group 3 (F > K > S) and group 4 (G > L) was identified. In silico analysis of LOS gene content in 564 C. coli genome sequences showed class III as the most abundant LOS locus type in C. coli. In silico analysis of LOS gene content also identified three novel LOS types of C. jejuni and previously unknown LOS biosynthesis genes in C. coli LOS locus types I, II, III, V and VIII. This study provides C. jejuni and C. coli LOS loci class frequencies in a smaller collection of C. jejuni clinical isolates as well as within the larger, worldwide database of C. jejuni and C. coli.

About the Authors:

Amber Hameed

Roles Data curation, Formal analysis, Investigation, Validation, Writing – original draft, Writing – review & editing

Affiliation: Centre for Physical Activity and Life Sciences, University of Northampton, Northampton, United Kingdom

Julian M. Ketley

Roles Conceptualization, Writing – review & editing

Affiliation: Department of Genetics and Genome Biology, University of Leicester, Leicester, United Kingdom

Alexandra Woodacre

Roles Formal analysis, Investigation, Methodology, Supervision, Validation, Writing – review & editing

Affiliations Centre for Physical Activity and Life Sciences, University of Northampton, Northampton, United Kingdom, Department of Genetics and Genome Biology, University of Leicester, Leicester, United Kingdom

ORCID logo https://orcid.org/0000-0002-5388-3180

Lee R. Machado

Roles Formal analysis, Investigation, Methodology, Project administration, Supervision, Validation, Visualization, Writing – review & editing

* E-mail: [email protected]

Affiliations Centre for Physical Activity and Life Sciences, University of Northampton, Northampton, United Kingdom, Department of Genetics and Genome Biology, University of Leicester, Leicester, United Kingdom

ORCID logo https://orcid.org/0000-0001-7974-2178

Gemma L. Marsden

Roles Conceptualization, Methodology, Writing – review & editing

Affiliation: Healthcare Infection Society, London, United Kingdom

ORCID logo https://orcid.org/0000-0001-8710-1071

Introduction

Campylobacter is a foodborne enteropathogen which transmits to humans mainly by consumption of Campylobacter contaminated dairy products and raw or partially cooked meat [1, 2]. Most Campylobacter species, including C. jejuni, C. coli, C. lari, C. fetus, C. upsaliensis, C. hypointestinalis, C. helveticus, C. lanienae, and C. mucosalis are found in warm-blooded animals [2–7]. However, some Campylobacter species, such as C. fetus and C. geochelonis can also occur in cold-blooded reptiles (lizard, tortoise, and snake) [8, 9]. Chicken is the main reservoir of Campylobacter through intestinal colonisation after hatching, usually at the age of 2–5 weeks [10–12]. Chickens contaminated with Campylobacter (with approximately 109 CFU/g caecal contents) are considered as a major source of Campylobacter transmission to humans [13, 14]. Campylobacter isolates can also be present as environmental contamination in non-livestock niches, which may be agricultural or non-agricultural [15–18].

The estimated number of cases of Campylobacter infection is approximately 96 million per year worldwide [19]. Campylobacter infection is characterised by an acute, self-limiting gastroenteritis in humans which causes various clinical symptoms including watery or bloody diarrhoea, abdominal pain, headache, fever, chills, and dysentery [20–22]. In some cases, neuronal disorders Guillain–Barré syndrome (GBS) and Miller Fisher syndrome (MFS), Reiter’s arthritis, and irritable bowel syndrome can also occur postinfection [23, 24].

Lipooligosaccharide (LOS) is an integral component of the outer cell membrane of Campylobacter and is synthesised by a cluster of lipooligosaccharide biosynthesis genes [25–27]. Each LOS biosynthesis gene present in this cluster produces an individual enzyme either for monosaccharide biosynthesis or addition of a particular monosaccharide to the LOS structure [28–31]. The LOS inner core biosynthesis genes are less variable and flank a highly variable central region of LOS outer core (OC) biosynthesis genes [27, 28, 32]. At one flank are waaC, waaM, and lgtF (also known as cj1135), and at the other, waaV, waaF, gmhA, waaE, waaD, gmhB, and cyf (also known as cj1153); these IC biosynthesis genes occur in the same order in almost all C. jejuni and C. coli strains. However, the OC biosynthesis gene cluster that extends from lgtF and waaV varies extensively among C. jejuni and C. coli strains and this region can acquire new genes by horizontal gene transfer during infection inside an animal host [33, 34]. Insertion or deletion events in this gene region give rise to a new LOS locus organisation or type in Campylobacter strains, which can be variable both in gene content and gene organisation [29, 35]. Disruption in resident LOS biosynthesis genes and allelic variation can also contribute to new LOS types [29]. A C. jejuni LOS class comprises a specific organisation of LOS genes named alphabetically; altogether 23 C. jejuni LOS classes (A through W) were previously described [27, 30, 32]. An updated and simplified classification system with categorisation of the previously described 23 C. jejuni LOS classes into four LOS groups (1–4) was presented in order to better understand the prevalence of C. jejuni LOS groups and group-related LOS classes [36]. Group 1 includes all those LOS locus types, A, B, C, R, M and V, which contain the sialic acid biosynthesis genes (neuA1, neuB1, neuC1 and cst-II/cst-III) whereas the other three groups had LOS loci with no sialic acid biosynthesis genes. Five classes (E, H, O, P and W) in LOS group 2, eight classes (D, F, K, Q, N, I, J, and S) in LOS group 3 and four classes (L, G, T, and U) in LOS group 4 were assimilated [36]. Variations in the LOS OC biosynthesis gene content cause modifications in the LOS OC structure [37, 38]. The LOS structures with variable OC epitopes help Campylobacter evade the host immune system as they mimic the human gangliosides [36, 38–42]. For this reason, antibodies produced against the LOS structural epitopes not only bind to LOS structures, but also to human gangliosides [33, 38, 39, 43]. The cross-reactivity of anti-LOS antibodies with human gangliosides is a critical contributory factor that leads to the development of GBS or MFS in humans [23, 24, 44]. Thus, the expression of variable cell surface LOS structures due to variable gene content in the LOS locus in C. jejuni is considered an important virulence factor and may have direct connection with the progression of different neural disorders [37, 45]. For example, C. jejuni strains with LOS locus class A and variable human ganglioside mimics (GM1a, GM1b, GD1a, and GD1b) may help C. jejuni to trigger GBS post-infection in Campylobacter infected patients [38, 40, 41, 46]. In contrast, C. jejuni strains with LOS class B and corresponding GQ1b-like LOS structures are suggested to be associated with MFS in Campylobacter infected patients [38, 47]. Based on the strong relationship between the variable LOS synthesis region and C. jejuni virulence, the current study aimed to determine the frequency of C. jejuni LOS genotypes from whole genome sequences present within GenBank (n = 703), as well as in C. jejuni clinical isolates (n = 60) from a UK hospital in order to further estimate the extent of gene variation in the C. jejuni LOS biosynthesis gene region and identify any novel C. jejuni LOS locus types. We previously analysed data from the literature relevant to the abundance of C. jejuni LOS classes in various geographical areas of the world to provide an overview of C. jejuni LOS genotype predominance [36]. The present study found that the prevalence of C. jejuni LOS group 1 classes (M, R, & V), group 2 class W, group 3 classes (K, Q, N, I, S & J) and all classes of group 4 (L, G, T & U) has never been identified before, which may be because of the relatively small size of collections involved in the previous studies. This study has for the first time presented the distribution of all, previously known LOS classes of C. jejuni using publicly accessible GenBank database.

Eight previously established C. coli LOS classes have been described I-VIII [32]. The relationship between different C. coli LOS locus structures and virulence is not fully established [48]. This is due to the presence of a wide variety in C. coli LOS biosynthesis locus types and limited knowledge of how this results in C. coli LOS structures at molecular level [32, 49]. As a result, we were interested in identifying the LOS locus types which dominate and are frequently present in C. coli isolates. To achieve this, we determined the frequency of C. coli LOS types present in GenBank which represents a global database of C. coli genome sequence deposits (n = 564). A few studies have been carried out previously to estimate the distribution of C. coli LOS classes present within the collections of livestock associated C. coli strains (n = 33) and agriculture related C. coli strains (n = 261) [32, 50]. This study presents an up-to-date picture of LOS genotype predominance in C. coli and identifies previously unreported C. coli LOS biosynthesis genes. The overall aim of this work was to provide a hierarchy of the frequency of LOS classes and identify prevalence of any novel LOS genes in C. jejuni and C. coli. This will aid investigation of increasingly complex levels of LOS variation and the role such variation plays in Campylobacter infection.

Results

Identification of frequency of C. jejuni LOS locus classes in GenBank database

With the routine upload of whole genome sequences of Campylobacter spp. in Genbank we aimed to determine the frequencies of C. jejuni LOS locus classes and LOS groups in a collection of 703 C. jejuni GenBank sequences by in silico analysis (Fig 1 and S1 Table online).

[Figure omitted. See PDF.]

Fig 1. A Circos plot showing the distribution of C. jejuni LOS locus classes (A-W), subclasses (A1, A2, B1, B2) and LOS groups (1–4) in the online C. jejuni sequence database.

Each segment of the inner circle specifies the total number of C. jejuni strains sequences (703) extracted for the LOS classification. The frequency of C. jejuni isolates classified for each particular LOS class/group is mentioned in numbers (n out of 703) on the top of each inner circle segment and represented with ribbon width. The frequency of a C. jejuni LOS class/group in percentage (% of overall frequency-100) is shown with each outer circle segment and represented by the orange or coloured bars. Ribbons link each C. jejuni LOS class to its related LOS group.

https://doi.org/10.1371/journal.pone.0265585.g001

58% (n = 400 of 703) of C. jejuni sequences belonged to the LOS group 1. The LOS classes A, B, and C adopted a hierarchy with class B1 (n = 125; 18%) > C (n = 109; 16%) > A1 (n = 68; 10%) > A2 (n = 44; 6%) > B2 (n = 36; 5%), were the most common LOS group 1 related classes. Other group 1 related classes including R (n = 10; 1.4%), M (n = 7; 1%), and V (n = 1; <1%) were rare. 30% (n = 214) of sequences were positive for either class E (n = 16; 2%), class H (n = 72; 10%), class O (n = 63; 9%), class P (n = 44; 6%) or class W (n = 13; 2%) and therefore, belonged to LOS group 2. 10% (n = 73) of C. jejuni sequences were positive for LOS group 3 classes including D (n = 2; <1%), F (n = 36; 5%), K (n = 16; 2%), Q (n = 1; <1%), N (n = 1; <1%), I (n = 3; <1%), S (n = 6; 1%) and J (n = 4; <1%). Only 2% of strains (n = 10 and n = 4 positive for class G and L, respectively) belonged to group 4.

Identification of frequency of C. jejuni LOS locus classes among C. jejuni clinical isolates

We were interested in examining whether Genbank LOS sequences were representative of LOS sequences present more locally. To address this, gDNA from 60 clinical C. jejuni isolates was extracted and the origin of DNA only from C. jejuni strains was confirmed by performing PCR reactions with waaM and waaV LOS gene specific control primers. Each DNA sample, positive for waaM and waaV LOS genes, was assigned with a LOS class based on PCR and Sanger sequencing results (S5 Table online). 50 of 60 C. jejuni isolates were typeable while remaining 10 were non-typeable. 6 of 50 (12%) classified C. jejuni strains (54386, S2, 92691, 118973, 118715 and 93133Y) were PCR positive for more than one LOS class. The frequency of C. jejuni LOS locus classes (A through W), subclasses (A1, A2, B1, B2) and LOS groups (1–4) prevalent in a C. jejuni clinical strains collection was determined (Fig 2). 62% (n = 31) of C. jejuni strains belonged to group 1 LOS classes. This included A1 (n = 3; 6%), A2 (n = 4; 8%), B1 (n = 3; 6%), B2 (n = 8; 16%), C (n = 10; 20%), and a mixed ABC class (n = 3; 6%). No C. jejuni strain was positive for other group 1 related classes (M, R, V). 32% (n = 16) of C. jejuni strains were positive for group 2 classes. E (n = 1; 2%), H (n = 3; 6%), O (n = 1; 2%), P (n = 8; 16%) or a mix HP (n = 3; 6%) were identified. Only 6% (n = 3) C. jejuni strains were assigned to the LOS group 3 related class F. No C. jejuni strain was associated with other group 3 classes (D, K, Q, N, I, S, J). In addition, C. jejuni strains with the LOS group 4 classes, L, G, T, & U, were absent from the clinical C. jejuni isolates.

[Figure omitted. See PDF.]

Fig 2. A Circos plot showing the distribution of C. jejuni LOS locus classes (A-W), subclasses (A1, A2, B1, B2) and LOS groups (1–4) from clinical isolates.

Each segment of the inner circle specifies the total number of classified C. jejuni isolates (50) used for the PCR based typing assay. The frequency of C. jejuni isolates classified for each particular LOS class/group is mentioned in numbers (n out of 50) on the top of each inner circle segment and represented with ribbon width. The frequency of a C. jejuni LOS class/group in percentage (% of overall frequency-100) is mentioned with each outer circle segment and represented by the orange or coloured bars. Ribbon ends link each C. jejuni LOS class to its related LOS group.

https://doi.org/10.1371/journal.pone.0265585.g002

A comparison of C. jejuni LOS class and group frequencies, identified in both collections of clinical C. jejuni isolates and online C. jejuni sequences reveal that the frequencies of different classes were comparable between the two sources except class P (Fig 3).

[Figure omitted. See PDF.]

Fig 3. A comparison of C. jejuni LOS biosynthesis locus class and group frequencies found in the collections of GenBank C. jejuni sequences (n = 703) and C. jejuni clinical, typed strains (n = 50).

https://doi.org/10.1371/journal.pone.0265585.g003

Identification of novel LOS biosynthesis locus types in C. jejuni

C. jejuni 1336 (Accession no: CM000854.1), C. jejuni 414 (Accession no: ADGM01000014.1) and C. jejuni CFSAN054107 (Accession no: CP028185.1) were found with novel LOS gene organisations or LOS types which were designated as class X, class Y and class Z respectively. The LOS biosynthesis region present between previously known LOS genes (cj1135, ORF17 and waaV) contained 13 LOS biosynthesis genes in C. jejuni 1336, 5 in C. jejuni 414 and 5 in C. jejuni CFSAN054107 (Fig 4).

[Figure omitted. See PDF.]

Fig 4. The genetic organisation of C. jejuni 1336, C. jejuni 414, and C. jejuni CFSAN05410 that contain novel LOS gene content.

13 LOS biosynthesis genes in C. jejuni 1336, 5 LOS biosynthesis genes in C. jejuni 414 and 5 LOS biosynthesis genes in C. jejuni CFSAN054107 occurred between previously known LOS genes (cj1135, ORF17, and waaV). Green arrows: previously reported LOS genes in C. jejuni strains; Pink arrows: previously known, variable LOS genes that had similarity to the LOS biosynthesis genes of other C. jejuni strains; Blue arrows: LOS genes that had similarity to the CPS biosynthesis genes of other C. jejuni strains; Purple arrows: LOS genes that had similarity to the LOS biosynthesis genes of C. coli strains. The direction of arrow represents the direction of gene transcription. Black star: Gene with an unknown function.

https://doi.org/10.1371/journal.pone.0265585.g004

Functions of 12 of 13 C. jejuni 1336 LOS genes are not known while one of them encodes an aminotransferase (WbdK). Functions of 4 of 5 C. jejuni 414 LOS genes are unknown while one of them encodes FkbM family methyltransferase. In C. jejuni CFSAN054107, 4 of 5 LOS genes encode glycosyltranferases and the remaining one encodes a methlytransferase (data from GenBank database). Further, the origin of these LOS novel genes was predicted by blast searching each gene against all sequences available in GenBank. 4 LOS genes in C. jejuni 1336 locus had >99% similarity with the capsular polysaccharide biosynthesis (CPS) genes of other C. jejuni strains, suggesting a possible gene transfer of CPS loci from other C. jejuni strains to the C. jejuni 1336 LOS locus. Six LOS genes of C. jejuni 1336 and one gene of C. jejuni 414 had no identity with the previously known C. jejuni LOS genes. Instead, they had >99% similarity with various LOS biosynthesis genes of C. coli, suggesting interspecies gene recombination events.

Identification of frequency of C. coli LOS locus types in GenBank database

The frequency of different C. coli LOS locus types (Fig 5) present in the online, publicly available GenBank database was identified by in silico analysis of 564 C. coli sequences (S3 Table online). C. coli LOS class III (41%; n = 229) was the most abundant in the GenBank database followed by class VIII (18%; n = 103), class I (13%; n = 70), class II (8%; n = 47), class VII (7%; n = 42), class IV (5% (n = 27), class VI (4%; n = 25) and class V (4%; n = 21).

[Figure omitted. See PDF.]

Fig 5. Distribution of C. coli LOS locus classes within the online C. coli sequences GenBank database.

The number of C. coli strains associated with each C. coli LOS class in 2D column chart and corresponding percentages of C. coli strains in Pie chart, represent the frequency of C. coli LOS locus classes within the global collection of C. coli isolates.

https://doi.org/10.1371/journal.pone.0265585.g005

Identification of previously unknown LOS biosynthesis genes in C. coli

Genes at the LOS biosynthesis cluster’s one flank (waaC, waaM, lgtF) and at the other (waaV, waaF, gmhA, waaE, waaD, gmhB) are present with the same order or organisation in almost all Campylobacter strains and are involved in the biosynthesis of the LOS inner core [30, 49]. LOS class W (C. jejuni M1) [32] and class E (C. jejuni 81116) contain two genes between waaF and gmhA while class B (C. jejuni 81–176) contains a single gene between waaF and gmhA. Similarly, previously unreported LOS biosynthesis genes, localised between waaF and gmhA in the distal end of five C. coli LOS locus types (I, II, III, V, VIII) were found in this study. Each C. coli LOS type had insertion of one LOS biosynthesis gene between waaF and gmhA (Fig 6, also given with C. coli reference strains in S4 Table online). These LOS genes are present at the same position in all five types of C. coli LOS locus but vary in size and nucleotide composition. The class I gene has a maximum size of ~1.4 kb and is distantly related to those genes which occur at the same position in other C. coli LOS locus types (II, III, V & VIII). The class II gene has similarity (~91%) to genes present in other classes (III, V and VIII). Type III and VIII genes are found to be identical (100%) and the class V gene is partially (~51%) similar to these identical genes. The presence of these previously unreported LOS biosynthesis genes was confirmed in 436 C. coli GenBank sequences (grey coloured columns in S3 Table online) within the LOS type I (n = 63), II (n = 43), III (n = 229), V (n = 3), and VIII (n = 98).

[Figure omitted. See PDF.]

Fig 6. Genomic organisation of previously unreported C. coli LOS core genes and their relative similarity.

The position of novel gene content between waaF and gmhA in C. coli LOS types (I, II, III, V and VIII). The sizes of these genes, obtained from the GenBank database, are given in kb. Each dotted line links two genes to represent the similarity between them in terms of query cover score (non-bold) and Megablast identity score (bold).

https://doi.org/10.1371/journal.pone.0265585.g006

The Class I gene encodes a β-Kdo transferase, class II gene encodes a phosphoheptose isomerase and genes in other classes produce glycosyltranferases for LOS synthesis (Data extracted from GenBank; S4 Table online).

Association of C. jejuni and C. coli LOS loci distribution to Campylobacter sources

The source of microbe isolation is usually specified with each sequence recorded in GenBank. The prevalence of C. jejuni (Fig 7) and C. coli (Fig 8) LOS genotypes in different Campylobacter niches was estimated by examining the online published sources (also given in S1 and S3 Tables online) of these C. jejuni and C. coli strains. C. jejuni and C. coli strains, isolated from faecal samples, were not included in this analysis as the actual source was unknown. The frequency of LOS genotypes within the pool of human C. jejuni isolates with online sequence data was comparable to the frequency of LOS genotypes identified within the collection of C. jejuni clinical isolates. For example, C. jejuni strains with LOS class F were common in our clinical isolates in comparison to other group 3 LOS classes (K, Q, N, I, J, S) and similarly, most abundantly isolated from humans prior to submit their genome sequences in GenBank. C. jejuni LOS group 1 associated C. jejuni strains were found mostly in humans [B (n = 33; 8%) > C (n = 29; 7%) > A (n = 24; 6%)] and chickens [B (n = 23; 5%) > A (n = 14; 3%) > C (n = 9; 2%)]. Humans, chickens, and the animal farm environment were the common isolation sources for C. jejuni and almost every C. jejuni LOS class was associated with at least one of these sources. Moreover, the most predominant LOS class III related C. coli strains were largely isolated from animal farm environments (n = 165; 37%), as well as from humans (n = 20, 5%) and chickens (n = 16, 4%), indicating that these are the common niches for type III LOS locus containing C. coli strains. The second most prevalent LOS class VIII was also frequent in the environment (n = 31, 7%) and at similar levels to isolates from humans (n = 26, 6%) and chickens (n = 25, 6%). These results indicated that chicken and humans are the most common hosts for C. jejuni while animal farm environments (particularly farm water and soil) is the most common niche for C. coli.

[Figure omitted. See PDF.]

Fig 7. Frequency of C. jejuni LOS locus classes in different Campylobacter sources.

Environment involves animal farm soil and animal farm water as Campylobacter sources.

https://doi.org/10.1371/journal.pone.0265585.g007

[Figure omitted. See PDF.]

Fig 8. Frequency of C. coli LOS locus classes in different Campylobacter sources.

Environment involves animal farm soil and animal farm water as Campylobacter sources.

https://doi.org/10.1371/journal.pone.0265585.g008

Discussion

By analysing the distribution of C. jejuni LOS locus types, we determined that the frequency of LOS classes within the LOS group 1 was similar in both collections of C. jejuni online sequences [class B (23%) > class C (16%) ≥ class A (16%)] and C. jejuni clinical isolates [class B (22%) > class C (20%) > class A (14%)]. The LOS class B was the most common class in C. jejuni isolates from a clinical cohort as well as in the C. jejuni GenBank database. It has been reported previously in other studies [29, 47, 51–53] as the most common LOS class in humans as well as poultry C. jejuni isolates. The current study highlights that class C is the second most abundant LOS locus class in C. jejuni. Many other studies have described the LOS class C as the major class of C. jejuni LOS biosynthesis locus [54, 55]. However, in comparison to the high prevalence of LOS class C (42%) in clinical isolates in Sweden [55], a very small number of clinical strains (2%) in Bangladesh had association with LOS locus C [47]. Most of the GBS-related C. jejuni strains in Bangladesh and China possessed the LOS locus class A rather than class C [47, 56, 57], suggesting that C. jejuni LOS class distribution is likely to vary geographically. High frequencies of classes B and C may be present due to their ability to encode heterogenous ganglioside mimics, which is advantageous for pathogenesis [58, 59]. The frequency of sialic acid biosynthesis LOS loci (A, B & C) in the online C. jejuni GenBank database as well as among the clinical C. jejuni isolates was high, although the former collection contained other C. jejuni sources (e.g., animals, birds and farm soil) in addition to human. It demonstrates that LOS group 1 containing LOS locus classes A, B, and C are commonly present in every type of C. jejuni source population. We have described this earlier in our literature review [36]. The low rate of GBS and MFS in Campylobacter infected patients [41, 46] despite high predominance of GBS/MFS associated LOS classes (A, B, and C) in the clinical cohort supports the notion that some other factors in addition to LOS structures significantly contribute to the development of these neural diseases [35, 36, 47]. Other LOS classes from group 1 (R, V and M), also contain genes for the biosynthesis of sialylated LOS structures [30, 32, 42], but they were absent from the clinical isolates, and this may reflect the fact that, even in a large repository, sequences belonging to these classes represented only 2.5% of the online database. The reasons for poor distribution of these classes remain unclear. 16% of LOS typed C. jejuni clinical strains had the LOS group 2 related class P and 10% of analysed C. jejuni sequences belonged to the LOS group 2 related class H, marking the class P in our local clinical C. jejuni collection and class H in the online C. jejuni sequence database as the most predominant LOS group 2 classes. These contrasting results might be explained because both LOS class P and H share almost similar gene content except for two LOS biosynthesis genes (Orf 26’ and Orf28) [57] and therefore only a few studies [47, 52] have ever considered these classes as two separate classes. LOS group 2 appeared as the second most abundant group among a small population of C. jejuni from the enteritis cases, although these types of loci lack the sialic acid biosynthesis genes and likely produce non-sialylated LOS structures [30, 60]. Therefore, LOS sialylation appears to be advantageous when present, but may not be absolutely critical for Campylobacter survival in animal hosts. Ganglioside-like structures other than GM1 or GQ1b were infrequent in LOS locus E associated C. jejuni strains as evidenced using serological assays [40]. However, the prevalence was low and structural assays were not employed. Therefore, it currently remains unclear whether ganglioside mimicry is produced in any non-group 1 strains. Within the LOS group 3, class F was the most prevalent class among C. jejuni clinical isolates (6%) and in the online sequence database (5%). Class K (2%) was the second most common class of group 3, whereas other LOS classes (I, S, J, D, Q, and N) were less frequent (≤1%) in GenBank database. C. jejuni sequences in a very small number (<3%) belonged to group 4. In contrast, no group 3 related classes (except for class F) and group 4 related classes were identified from C. jejuni clinical isolates, which may be because of the relatively small size of the collection. The hierarchy of LOS group prevalence was group 1 > group 2 > group 3 > group 4 in both types of collections of C. jejuni isolates.

Six C. jejuni strains were found positive for more than two LOS classes using PCR, which occur due to co-infection in patients with multiple C. jejuni strains. The co-infection occurrence with multiple C. jejuni strains has been previously observed in GBS patients [61]. Another reason could be the occurrence of LOS gene recombination during infection as it has been observed previously [33, 34]. The in-silico prediction for the presence of six C. coli LOS genes in C. jejuni 1336 and one C. coli LOS gene in C. jejuni 414 highlight the occurrence of interspecies genes recombination events. C. jejuni does not only harbour the genes from C. coli, but C. coli can also uptake and acquire C. jejuni DNA, especially when they are present in the same niche [18]. C. jejuni and C. coli share 71% of LOS biosynthesis genes and 65% of CPS biosynthesis genes because of recombination events [32].

LOS locus type III was abundant (41%; n = 229 of 564) in the online GenBank database of C. coli sequences. The high frequency of class III (28%; n = 72 of 261) within the agriculture-associated C. coli strains has been reported in a previous study [50]. The reasons behind the high prevalence of LOS class III in C. coli are yet to be investigated. C. coli strain 76339 contains the sialic acid biosynthesis genes (cst-V, neuA, neuB, & neuC) and sialic acids in its LOS structure [49, 62]. In the current study, all C. coli LOS sequences extracted from GenBank were scrutinised for the presence of these sialic acid biosynthesis genes and only C. coli RM4661 (Accession no: CP007181.1) was found to contain sialic acid synthesis genes (cst, neuB & neuC) in the LOS locus. However, C. coli LOS locus classes with cst alleles were identified in another study where XV-XXIV had cst-II; XIII and XIV had cst-III and IX, XXV, XXVI had cst-V. All these classes were positive for neuABC [49, 50]. Other LOS classes of C. coli including II, III, XXVII—XXXV had different alleles of cst (cst-IV and cst-VI) and only a small fraction of these classes (6.44% and 4.51% of strains positive with cst-IV and cst-VI respectively) had neuABC positioned outside of LOS locus [50]. Despite having genes (sialyltransferases) associated with the LOS sialylation, C. coli strains with ganglioside mimicry have not been reported so far [48–50, 63]. The most common LOS class (III) and the second most common class (VIII) linked C. coli strains were largely isolated from humans, which is concordant with a previous study, where half (57%) of the clinical isolates belonged to class III, VIII and II [64]. All C. coli classes tended to come from humans, chickens and the farm environment, suggesting that animal farm water and soil, in addition to chickens, are the primary sources of C. coli transmission to humans. This agrees with a previous study that reported agriculture associated C. coli as an emerging human pathogen [17, 18]. In this study, previously unreported LOS genes were identified in the C. coli LOS biosynthesis locus types I, II, III, V and VIII. The identified LOS biosynthesis genes in C. coli vary at the sequence and functional level among C. coli LOS locus classes (based on data available in GenBank). It highlights that other region of LOS biosynthesis locus also vary amongst C. coli strains in addition to the central region of LOS locus that can impact the LOS structure in C. coli. The sequence level variation in these genes and their possible putative functions have been determined in silico but require further functional characterisation.

The genome sequences deposited in GenBank may not be absolutely correct due to occurrence of errors at the experimental or sequence data analysis stages [65], and therefore can produce false-positive results. In this study, variation in LOS biosynthesis locus was analysed at the base sequence level to determine the presence of clustered whole LOS genes rather than finding the modifications between sequence bases. The identification of several LOS genes within a single genomic sequence reduces the possibility of false-positive results but does not eliminate it.

In conclusion, this work compares the frequency of various C. jejuni LOS locus classes in local versus global collections of C. jejuni isolates and provides an overview of predominance of various LOS biosynthesis gene clusters in the populations of C. jejuni and C. coli.

Materials and methods

Collection of bacterial strains and their growth

In a 12-month period from November 2015–2016, C. jejuni isolates (n = 60; S5 Table online) from anonymised clinical samples from Northampton General Hospital, UK were collected by swabbing cultured Charcoal-Cefoperazone-Deoxycholate Agar (CCDA) plates. Amines and charcoal swabs (Thermo Fisher Scientific) were used for collection and transportation of Campylobacter isolates. Bacterial isolates were cultured again within 24 hours of collection and grown on MHA plates at 37°C for 24–48 hours under a microaerobic atmosphere of 5% O2, 10% CO2 and 85% N2. The microaerobic environment was provided by either using CampyGen sachets (Oxoid Limited) in 2.5 L air-tight jars or BOC gas mixture (2% H2, 5% O2, 10% CO2 and 83% N2) in a Whitley G2 workstation (Don Whitley Scientific).

DNA extraction from C. jejuni clinical strains

For the extraction of genomic DNA (gDNA) from C. jejuni isolates, the protocol provided by the DNeasy Blood and Tissue kit manufacturer (Qiagen) was followed.

Designing of LOS class specific primers

25 C. jejuni LOS class specific PCR primer pairs including waaM and waaV LOS gene specific control primers (S2 Table online) were designed using Clone Manager Professional Suite (Version 8; Scientific & Educational Software, Morrisville, USA) and purchased from the Eurofins Genomics (Ebersberg, Germany). Each LOS class specific primer pair spanned the junction of two adjacent LOS genes and were specific to two LOS biosynthesis genes rather than a single gene.

PCR for LOS typing

PCR master mix (20 μL) was prepared after mixing the template DNA (~50 ng) with forward and reverse primers (10μM each), MyTaq™ red DNA polymerase (0.25 μL containing 1.25 units; Bioline Reagents Ltd) and 5X MyTaq™ red Reaction Buffer (5 μL containing 5 mM dNTPs and 15 mM MgCl2; Bioline Regents Ltd). Each reaction was carried out in a thermocyler (TECHNE) using appropriate cycling conditions: 1 cycle of initial template denaturation (95°C; 5 min) followed by 35 cycles of template amplification (template denaturation at 95°C for 30 sec, primer annealing at optimised temperature for 35 sec, primer extension at 72°C for 30 sec per 1 kb of expected PCR product size) and one cycle of final extension (72°C for 5 min). PCR products were resolved by electrophoresis on 1% (w/v; dissolved in 1X TAE buffer) agarose gel, which were stained with SYBR® safe stain (10,000X; Invitrogen) and visualised under UV light in G: Box (Syngene). PCRs with gDNA of reference C. jejuni strains (as indicated in S5 Table online) were used as positive controls and PCR with gDNA of C. jejuni 11168Δ32–52 (a mutant strain of C. jejuni NCTC11168 lacking LOS biosynthesis region from gene cj1132-cj1152) was used as a negative control in all PCR reactions.

Sanger sequencing of PCR products

The protocol provided with the Eurofins Mix2Seq kit was followed for Sanger sequencing PCR products. According to the protocol, 15 μL purified PCR product (1–15 ng/μL) was mixed with 2 μL of either forward or reverse primer stock solution (10 pmol/μL). 17 μL DNA/primer mix was pipetted into a Mix2Seq tube and sent to Eurofins, Wolverhampton, U.K. for sequencing. Sequence data, obtained online in fasta format, was analysed using Clone Manager Professional Suite.

Bioinformatic analysis

The genome sequences of 703 C. jejuni strains (125 complete; 578 draft) and 564 C. coli strains (22 complete; 542 draft) available on the 1st March 2018 were obtained from GenBank and scrutinised by alignment using Megablast (https://blast.ncbi.nlm.nih.gov). The information related to the reference strains with previously defined LOS types (subject sequences) has been given in S2 and S4 Tables while for query sequences, used in this analysis, has been given in S1 and S3 Tables. A LOS gene was considered present if ≥80% of the query sequence was effectively mapped to the reference LOS gene sequence and had ≥80% nucleotide identity with the reference LOS gene sequence. Subsequently, based on the presence or absence of distinct LOS genes, combination of LOS genes or a LOS class was identified, and assigned to a particular C. jejuni or C. coli query sequence. Megablast detects a query by a single accession number for a complete genome sequence but incorporates multiple sub-accession numbers for a draft genome sequence. Therefore, to reduce the burden of query sequences, those draft sequence contigs were identified which contained the LOS biosynthesis gene sequence. For this purpose, all sequenced contigs associated with each draft sequence were aligned against the commonly present two LOS gene (waaC and waaF) sequences using Megablast. Subsequently, contigs with LOS sequence were used further for LOS classification analysis.

Ethics statement

Campylobacter isolates from anonymised clinical samples were collected by the Swab method under the sterile conditions from already cultured plates. No experimentation was carried out on humans during this research and this research did not involve the use of human tissues, fluids or DNA samples.

Supporting information

S1 Table. LOS types of C. jejuni complete (n = 125) and draft sequences (n = 578).

https://doi.org/10.1371/journal.pone.0265585.s001

(PDF)

S2 Table. Primers for the identification of C. jejuni LOS locus classes.

https://doi.org/10.1371/journal.pone.0265585.s002

(PDF)

S3 Table. LOS types of C. coli complete (n = 22) and draft sequences (n = 542).

https://doi.org/10.1371/journal.pone.0265585.s003

(PDF)

S4 Table. Summary of C. coli reference strains used to define LOS classes (Richards et al., 2013) in this study.

https://doi.org/10.1371/journal.pone.0265585.s004

(PDF)

S5 Table. Summary of C. jejuni LOS locus typing and PCR products’ sequencing results.

https://doi.org/10.1371/journal.pone.0265585.s005

(PDF)

Acknowledgments

We would like to thank Andrea O’Connell at Northampton General Hospital for providing the clinical isolates.

Citation: Hameed A, Ketley JM, Woodacre A, Machado LR, Marsden GL (2022) Molecular and in silico typing of the lipooligosaccharide biosynthesis gene cluster in Campylobacter jejuni and Campylobacter coli. PLoS ONE 17(3): e0265585. https://doi.org/10.1371/journal.pone.0265585

References

1. Robinson DA. Infective dose of Campylobacter jejuni in milk. Br Med J. (Clinical research ed). 1981;282(6276):1584. pmid:6786571

2. Rahimi E, Kazemeini HR, Safaei S, Allahbakhshi K, Momeni M, Riahi M. Detection and identification of Campylobacter spp. from retail raw chicken, turkey, sheep and goat meat in Ahvaz, Iran. African J Microbiol Res. 2010;4(15):1620–3.

3. Roberts L, Lawson GH, Rowland AC. Experimental infection of neonatal pigs with Campylobacter sputorum subspecies mucosalis with special reference to the oral cavity. Vet Microbiol, 1980;5(3):249–255.

4. Gebhart CJ, Edmonds P, Ward GE, Kurtz HJ, Brenner DONJ. “Campylobacter hyointestinalis” sp. nov.: a New Species of Campylobacter Found in the Intestines of Pigs and Other Animals. J Clin Microbiol. 1985;21(5):715–720. pmid:3998100

5. Inglis GD, Kalischuk LD, Busz HW, Kastelic JP. Colonization of cattle intestines by Campylobacter jejuni and Campylobacter lanienae. Appl Environ Microbiol. 2005;71(9):5145–53. pmid:16151098

6. Workman SN, Mathison GE, Marc C, Lavoie MC. Pet Dogs and Chicken Meat as Reservoirs of Campylobacter spp. in Barbados Pet Dogs and Chicken Meat as Reservoirs of Campylobacter spp. in Barbados. J Clin Microbiol. 2005;43(6):2642–50. pmid:15956378

7. Weis AM, Miller WA, Byrne BA, Chouicha N, Boyce WM, Townsend AK. Prevalence and pathogenic potential of Campylobacter isolates from free-living, human-commensal American crows. Appl Environ Microbiol. 2014;80(5):1639–44. pmid:24375131

8. Wang CM, Wu ZY, Shia WY, Jhou YJ, Tung KC, Shyu CL. Complete genome sequence of Campylobacter fetus subsp. testudinum strain Pet-3, isolated from a lizard (Hydrosaurus pustulatus). Genome Announc. 2015;3(1):1–14. pmid:25700400

9. Piccirillo A, Niero G, Calleros L, Pérez R, Naya H, Iraola G. Campylobacter geochelonis sp. nov. isolated from the western Hermann’s tortoise (Testudo hermanni hermanni). Int J Syst Evol Microbiol. 2016;66(9):3468–76. pmid:27266587

10. Neill SD, Campbell JN, Greene JA. Campylobacter species in broiler chickens. Avian Pathol. 1984;13(4):777–85. pmid:18766887

11. Humphrey TJ, Henley A, Lanning DG. The colonization of broiler chickens with Campylobacter jejuni: some epidemiological investigations. Epidemiol Infect. 1993;110(3): 601–607. pmid:8519325

12. Berndtson E, Danielsson-Tham ML, Engvall A. Campylobacter incidence on a chicken farm and the spread of Campylobacter during the slaughter process. Int J Food Microbiol. 1996;32(1–2):35–47. pmid:8880326

13. Atanassova V, Ring C. Prevalence of Campylobacter spp. in poultry and poultry meat in Germany. Int J Food Microbiol. 1999;51(2–3):187–90. pmid:10574094

14. Newell DG, Fearnley C. Sources of Campylobacter colonization in broiler chickens. Appl Environ Microbiol. 2003;69(8):4343–51. pmid:12902214

15. Champion OL, Gaunt MW, Gundogdu O, Elmi A, Witney AA, Hinds J, et al. Comparative phylogenomics of the food-borne pathogen Campylobacter jejuni reveals genetic markers predictive of infection source. PNAS. 2005;102(44):16043–16048. pmid:16230626

16. Wilson DJ, Gabriel E, Leatherbarrow AJH, Cheesbrough J, Gee S, Bolton E, et al. Tracing the source of campylobacteriosis. PLoS Genet. 2008;4(9). pmid:18818764

17. Sheppard SK, Dallas JF, Wilson DJ, Strachan NJC, McCarthy ND, Jolley KA, et al. Evolution of an agriculture-associated disease causing Campylobacter coli clade: Evidence from national surveillance data in Scotland. PLoS One. 2010;5(12):1–9. pmid:21179537

18. Sheppard SK, Didelot X, Jolley KA, Darling AE, Pascoe B, Meric G, et al. Progressive genome-wide introgression in agricultural Campylobacter coli. Mol Ecol. 2013;22(4):1051–64. pmid:23279096

19. Havelaar AH, Kirk MD, Torgerson PR, Gibb HJ, Hald T, Lake RJ, et al. World Health Organization Global Estimates and Regional Comparisons of the Burden of Foodborne Disease in 2010. PLoS Med. 2015;12(12):e1001923. pmid:26633896

20. van Spreeuwel JP, Duursma GC, Meijer CJ, Bax R, Rosekrans PC, Lindeman J. Campylobacter colitis: histological immunohistochemical and ultrastructural findings. Gut. 1985;26(9):945–51. pmid:4029720

21. Black RE, Levine MM, Clements ML, Hughes TP, Blaser MJ. Experimental Campylobacter jejuni infection in humans. J Infect Dis. 1988;157(3):472–9. pmid:3343522

22. Perkins DJ, Newstead GL. (1994) Campylobacter jejuni enterocolitis causing peritonitis, ileitis and intestinal obstruction. The ANZ J Surg. 1994;64(1):55–58.

23. Endtz HP, Ang CW, Van Den Braak N, Duim B, Rigter A, Price LJ, et al. Molecular characterization of Campylobacter jejuni from patients with Guillain-Barré and Miller Fisher syndromes. J Clin Microbiol. 2000;38(6):2297–301. pmid:10834992

24. McCarthy N, Giesecke J. Incidence of Guillain-Barré syndrome following infection with Campylobacter jejuni. Am J Epidemiol. 2001;153(6):610–4. pmid:11257070

25. Mandrell RE, McLaughlin R, Kwaik YA, Lesse A, Yamasaki R, Gibson B, et al. Lipooligosaccharides (LOS) of some Haemophilus species mimic human glycosphingolipids, and some LOS are sialylated. Infect Immun. 1992;60(4):1322–1328. pmid:1372291

26. Moran AP. Structure and conserved characteristics of Campylobacter jejuni lipopolysaccharides. J Infect Dis. 1997;176:S115–21. pmid:9396693

27. Gilbert M, Karwaski MF, Bernatchez S, Young NM, Taboada E, Michniewicz J, et al. The genetic bases for the variation in the lipo-oligosaccharide of the mucosal pathogen, Campylobacter jejuni. Biosynthesis of sialylated ganglioside mimics in the core oligosaccharide. J Biol Chem. 2002;277(1):327–37. pmid:11689567

28. Karlyshev A, Ketley J, Wren B. The glycome. FEMS Microbiol Rev. 2005;29(2):377–90. pmid:15808749

29. Parker CT, Horn ST, Gilbert M, Miller WG, Woodward DL, Mandrell RE. Comparison of Campylobacter jejuni lipooligosaccharide biosynthesis loci from a variety of sources. J Clin Microbiol. 2005;43(6):2771–81. pmid:15956396

30. Parker CT, Gilbert M, Yuki N, Endtz HP, Mandrell RE. Characterization of lipooligosaccharide-biosynthetic loci of Campylobacter jejuni reveals new lipooligosaccharide classes: Evidence of mosaic organizations. J Bacteriol. 2008;190(16):5681–9. pmid:18556784

31. Iwata T, Chiku K, ichi Amano K, Kusumoto M, Ohnishi-Kameyama M, Ono H, et al. Effects of Lipooligosaccharide Inner Core Truncation on Bile Resistance and Chick Colonization by Campylobacter jejuni. PLoS One. 2013;8(2). pmid:23437265

32. Richards VP, Lefébure T, Pavinski Bitar PD, Stanhope MJ. 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. 2013;14(1):200–13.

33. Gilbert M, Godschalk PCR, Karwaski MF, Ang CW, Van Belkum A, Li J, et al. Evidence for Acquisition of the Lipooligosaccharide Biosynthesis Locus in Campylobacter jejuni GB11, a Strain Isolated from a Patient with Guillain-Barré Syndrome, by Horizontal Exchange. Infect Immun. 2004;72(2):1162–5. pmid:14742567

34. Phongsisay V, Perera VN, Fry BN. Exchange of lipooligosaccharide synthesis genes creates potential Guillain-Barré syndrome-inducible strains of Campylobacter jejuni. Infect Immun. 2006;74:1368–1372. pmid:16428786

35. Revez J, Hänninen ML. Lipooligosaccharide locus classes are associated with certain Campylobacter jejuni multilocus sequence types. Eur J Clin Microbiol Infect Dis. 2012;31:2203–9. pmid:22298242

36. Hameed A, Woodacre A, Machado LR, Marsden GL. An Updated Classification System and Review of the Lipooligosaccharide Biosynthesis Gene Locus in Campylobacter jejuni. Front Microbiol. 2020;11:677. pmid:32508756

37. Müller J, Meyer B, Hänel I, Hotzel H. Comparison of lipooligosaccharide biosynthesis genes of Campylobacter jejuni strains with varying abilities to colonize the chicken gut and to invade Caco-2 cells. J Med Microbiol. 2007;56(12):1589–1594.

38. Godschalk PCR, Kuijf ML, Li J, St. Michael F, Ang CW, Jacobs BC, et al. Structural characterization of Campylobacter jejuni lipooligosaccharide outer cores associated with Guillain- Barré Syndrome and Miller Fisher syndromes. Infect Immun. 2007;75(3):1245–54. pmid:17261613

39. Linton D, Gilbert M, Hitchen PG, Dell A, Morris HR, Wakarchuk WW, et al. Phase variation of a β-1,3 galactosyltransferase involved in generation of the ganglioside GM1-like lipo-oligosaccharide of Campylobacter jejuni. Mol Microbiol. 2000;37(3):501–14. pmid:10931344

40. Godschalk PCR, Heikema AP, Gilbert M, Komagamine T, Wim Ang C, Glerum J, et al. The crucial role of Campylobacter jejuni genes in anti-ganglioside antibody induction in Guillain- Barré Syndrome. J Clin Invest. 2004;114(11):1659–65. pmid:15578098

41. Mortensen NP, Kuijf ML, Ang CW, Schiellerup P, Krogfelt KA, Jacobs BC, et al. Sialylation of Campylobacter jejuni lipo-oligosaccharides is associated with severe gastro-enteritis and reactive arthritis. Microbes Infect. 2009;11(12):988–94. pmid:19631279

42. Houliston RS, Vinogradov E, Dzieciatkowska M, Li J, St Michael F, Karwaski MF, et al. Lipooligosaccharide of Campylobacter jejuni: Similarity with multiple types of mammalian glycans beyond gangliosides. J Biol Chem. 2011;286(14):12361–70. pmid:21257763

43. Hameed A. Human Immunity against Campylobacter Infection. 2019;19(6):1–17.

44. Nachamkin I, Allos BM, Ho T. Campylobacter species and Guillain-Barré syndrome. Clin Microbiol Rev. 1998;11(3):555–67. pmid:9665983

45. Wachira VK, Peixoto HM, and de Oliveira MRF. Systematic review of factors associated with the development of Guillian-Barré syndrome 2007–2017: what has changed? Trop. Med. Int. Health 24. 2019;24(2):132–142.

46. Nachamkin I, Liu J, Li M, Ung H, Moran AP, Prendergast MM, et al. Campylobacter jejuni from patients with Guillain-Barré syndrome preferentially expresses a GD1a-like epitope. Infect Immun. 2002;70(9):5299–303. pmid:12183587

47. Islam Z, van Belkum A, Wagenaar JA, Cody AJ, de Boer AG, Sarker SK, et al. Comparative population structure analysis of Campylobacter jejuni from human and poultry origin in Bangladesh. Eur J Clin Microbiol Infect Dis. 2014;33(12):2173–81. pmid:24962195

48. Van Belkum A, Jacobs B, Van Beek E, Louwen R, Van Rijs W, Debruyne L, et al. Can Campylobacter coli induce Guillain-Barré syndrome? Eur J Clin Microbiol Infect Dis. 2009;28(5):557–60. pmid:19002726

49. Kolehmainen A, Rossi M, Stupak J, Li J, Gilbert M, Wakarchuk W. Genetics behind the biosynthesis of nonulosonic acid containing lipooligosaccharides in Campylobacter coli. J Bacteriol, 2019;201(8):e00759–18. pmid:30692173

50. Culebro A, Machado MP, Carriço JA, Rossi M. Origin, evolution, and distribution of the molecular machinery for biosynthesis of sialylated lipooligosaccharide structures in Campylobacter coli. Sci Rep, 2018;8:3028. pmid:29445215

51. Quiñones B, Parker CT, Janda JM, Miller WG, Mandrell RE. Detection and genotyping of Arcobacter and Campylobacter isolates from retail chicken samples by use of DNA oligonucleotide arrays. Appl Environ Microbiol. 2007;73(11):3645–3655. pmid:17416693

52. Habib I, Louwen R, Uyttendaele M, Houf K, Vandenberg O, Nieuwenhuis EE, et al. Correlation between genotypic diversity, lipooligosaccharide gene locus class variation, and caco-2 cell invasion potential of Campylobacter jejuni isolates from chicken meat and humans: Contribution to virulotyping. Appl Environ Microbiol. 2009;75(13):4277–88. pmid:19411422

53. Islam Z, Sarker SK, Jahan I, Farzana KS, Ahmed D, Faruque ASG, et al. Capsular genotype and lipooligosaccharide locus class distribution in Campylobacter jejuni from young children with diarrhea and asymptomatic carriers in Bangladesh. Eur J Clin Microbiol Infect Dis. European Journal of Clinical Microbiology & Infectious Diseases; 2018;37(4):723–8.

54. Ellström P, Feodoroff B, Hänninen ML, Rautelin H. Characterization of clinical Campylobacter jejuni isolates with special emphasis on lipooligosaccharide locus class, putative virulence factors and host response. Int J Med Microbiol. Elsevier GmbH.; 2013;303(3):134–9. pmid:23528202

55. Ellström P, Hansson I, Nilsson A, Rautelin H, Olsson Engvall E. Lipooligosaccharide locus classes and putative virulence genes among chicken and human Campylobacter jejuni isolates. BMC Microbiol. BMC Microbiology; 2016;16(1):1–6. pmid:27871232

56. Islam Z, van Belkum A, Wagenaar JA, Cody AJ, de Boer AG, Tabor H, et al. Comparative genotyping of Campylobacter jejuni strains from patients with Guillain-Barré syndrome. PLoS One. 2009;4(9):1–8. pmid:19789649

57. Jiang H, Zhang MJ, Liu RC, Tian XY, Gu YX, Zhang JZ. Characteristics of lipo-oligosaccharide loci of Campylobacter jejuni isolates associated with Guillain-Barré syndrome from Hebei, China. Int J Mol Sci. 2010;11(3):1155–61. pmid:20480006

58. Guerry P, Szymanski CM, Prendergast MM, Hickey TE, Ewing CP, Pattarini DL, et al. Phase Variation of Campylobacter jejuni 81–176 Lipooligosaccharide Affects Ganglioside Mimicry and Invasiveness In Vitro. Infect Immun. 2002;70:787–93. pmid:11796612

59. St Michael F, Szymanski CM, Li J, Chan KH, Khieu NH, Larocque S, et al. The structures of the lipooligosaccharide and capsule polysaccharide of Campylobacter jejuni genome sequenced strain NCTC 11168. Eur J Biochem. 2002;269(21):5119–5136. pmid:12392544

60. Poly F, Read TD, Chen YH, Monteiro MA, Serichantalergs O, Pootong P, et al. Characterization of two Campylobacter jejuni strains for use in volunteer experimental-infection studies. Infect Immun. 2008;76(12):5655–67. pmid:18809665

61. Godschalk PCR, Gilbert M, Jacobs BC, Kramers T, Tio-Gillen AP, Ang CW, et al. Co-infection with two different Campylobacter jejuni strains in a patient with the Guillain-Barré syndrome. Microbes Infect. 2006;8(1):248–53. pmid:16213180

62. Skarp-de Haan CPA, Culebro A, Schott T, Revez J, Schweda EKH, Hänninen ML, et al. Comparative genomics of unintrogressed Campylobacter coli clades 2 and 3. BMC Genomics. 2014;15(1). pmid:24524824

63. Funakoshi K, Koga M, Takahashi M, Hirata K, Yuki N. Campylobacter coli enteritis and Guillain–Barré syndrome: No evidence of molecular mimicry and serological relationship. J Neurol Sci. 2006;1(2):163–168. pmid:16600306

64. Culebro A, Revez J, Pascoe B, Friedmann Y, Hitchings MD, Stupak J, et al. Large sequence diversity within the biosynthesis locus and common biochemical features of Campylobacter coli lipooligosaccharides. J Bacteriol. 2016;198(20):2829–40. pmid:27481928

65. Pertsemlidis A, Fondon JW. Having a BLAST with bioinformatics (and avoiding BLASTphemy). Genome Biol.2001;2(10):1–10. pmid:11597340