microorganisms Article Differential Distribution of the wlaN and cgtB Genes, Associated with Guillain-Barré Syndrome, in Campylobacter jejuni Isolates from Humans, Broiler Chickens, and Wild Birds Pedro Guirado 1 , Sonia Paytubi 1,† , Elisenda Miró 2 , Yaidelis Iglesias-Torrens 2,3 , Ferran Navarro 2,3, * , Marta Cerdà-Cuéllar 4 , Camille Stephan-Otto Attolini 5 , Carlos Balsalobre 1 and Cristina Madrid 1, * 1 Departament de Genètica, Microbiologia i Estadística, Facultat de Biología, Universitat de Barcelona. Avda. Diagonal 643, 08028 Barcelona, Spain; pedro.guirado.frias@gmail.com (P.G.); s.paytubi@gmail.com (S.P.); cbalsalobre@ub.edu (C.B.) 2 Hospital de la Santa Creu i Sant Pau and Institut d’Investigació Biomèdica Sant Pau (IIB Sant Pau), Sant Quintí 89, 08041 Barcelona, Spain; EMiro@santpau.cat (E.M.); yaide87@yahoo.com (Y.I.-T.) 3 Departament de Genètica i Microbiologia. Universitat Autònoma de Barcelona, 08193 Cerdanyola del Vallès (Barcelona), Spain 4 IRTA, Centre de Recerca en Sanitat Animal (CReSA-IRTA-UAB). Campus de la Universitat Autònoma de Barcelona, Bellaterra, 08193 Barcelona, Spain; marta.cerda@irta.cat 5 Institute for Research in Biomedicine (IRB Barcelona), The Barcelona Institute of Science and Technology, Baldiri Reixac 10, 08028 Barcelona, Spain; camille.stephan@irbbarcelona.org * Correspondence: cmadrid@ub.edu (C.M.); Fnavarror@santpau.cat (F.N.); Tel.: +34-93-403-9382 (C.M.); +34-93-553-7380 (F.N.) † Current address: Infections and Cancer Laboratory. Cancer Epidemiology Research Program (CERP). Institut Català d’Oncologia (ICO)–IDIBELL. Avda. de la Gran Via 199, 08908 L’Hospitalet de Llobregat (Barcelona), Spain.  Received: 23 January 2020; Accepted: 24 February 2020; Published: 26 February 2020  Abstract: Campylobacter jejuni causes campylobacteriosis, a bacterial gastroenteritis with high incidence worldwide. Moreover, C. jejuni infection can trigger the polyneuropathic disorder denominated Guillain-Barré syndrome (GBS). The C. jejuni strains that can elicit GBS carry either wlaN or cgtB, coding both genes for a β-1,3-galactosyltransferase enzyme that is required for the production of sialylated lipooligosaccharide (LOSSIAL ). We described a differential prevalence of the genes wlaN and cgtB in C. jejuni isolates from three different ecological niches: humans, broiler chickens, and wild birds. The distribution of both genes, which is similar between broiler chicken and human isolates and distinct when compared to the wild bird isolates, suggests a host-dependent distribution. Moreover, the prevalence of the wlaN and cgtB genes seems to be restricted to some clonal complexes. Gene sequencing identified the presence of new variants of the G- homopolymeric tract within the wlaN gene. Furthermore, we detected two variants of a G rich region within the cgtB gene, suggesting that, similarly to wlaN, the G-tract in the cgtB gene mediates the phase variation control of cgtB expression. Caco-2 cell invasion assays indicate that there is no evident correlation between the production of LOSSIAL and the ability to invade eukaryotic cells. Keywords: Campylobacter; lipooligosaccharide; wlaN; cgtB; Guillain-Barré syndrome 1. Introduction Campylobacter jejuni causes the most common bacterial food-borne disease in Europe, a gastroenteritis named campylobacteriosis, which can range from mild, watery to bloody diarrhea [1,2]. Microorganisms 2020, 8, 325; doi:10.3390/microorganisms8030325 www.mdpi.com/journal/microorganisms Microorganisms 2020, 8, 325 2 of 11 Campylobacteriosis is mostly transmitted to humans by the consumption and improper handling of poultry meat [3]. Wild birds play a relevant role in the C. jejuni transmission route by being carriers that can infect both poultry and humans [4,5]. Moreover, campylobacteriosis is the most frequent infection preceding Guillain-Barré syndrome (GBS), an acute polyneuropathic disorder [6]. The strains of C. jejuni that can potentially elicit GBS are those that produce a sialylated LOS (LOSSIAL ), due to the well-documented molecular mimicry between the LOSSIAL and the saccharide component of the human GM1 ganglioside which is present in peripheral nerves [7]. Antibodies generated against Campylobacter LOSSIAL structures might cross-react with GM1 found in nerve tissue [1]. Consistently, during the acute-phase of C. jejuni-associated GBS infection, the sera contain high titers of antibodies against LOSSIAL that cross-react with GM1 gangliosides [8]. The operon responsible for the biosynthesis of LOS is highly diverse among C. jejuni isolates, with up to 19 different locus classes, named from A to S [9,10]. Only classes A, B, and C carry genes coding for a β-1,3-galactosyltransferase catalyzing the addition of a galactose molecule, which is required for the production of a GM1-like LOSSIAL structure [11]. Two different genes have been identified to encode for this enzymatic activity: cgtB (in class A and B) and wlaN (in class C) [10,12]. The amino acid identity between CgtB and WlaN is about 58% [12]. Genomic analyses revealed the presence of hypervariable sequences in C. jejuni genes coding for proteins involved in the biosynthesis of surface structures [13]. Most of these hypervariable sequences consist of short homopolymeric nucleotide tracts. The wlaN gene contains a G-tract located within the coding sequence, rendering the expression of the wlaN gene under phase variation control, presumably by a slipped-strand mechanism [12]. Three different G-tract variants have been described, carrying either six, eight, or nine residues [12,14]. The presence of an 8G-tract results in a full-length product (ON-phase), whereas either a 6G- or a 9G-tract results in truncated proteins (OFF-phase). In this work, we describe the prevalence of the genes responsible for LOSSIAL production, cgtB and wlaN, in C. jejuni isolates from different ecological origin: humans, broiler chickens, and wild birds. The presence and diversity of homopolymeric G-tract regions within both LOSSIAL -related genes are studied. The correlation between the production of LOSSIAL and the ability to invade eukaryotic cells is evaluated. 2. Materials and Methods 2.1. Bacterial Strains and Culture Conditions The C. jejuni strain collection, previously described [15], is composed of 150 isolates obtained from faeces of three different sources: human patients suffering from symptomatic gastroenteritis (50 isolates), broiler chickens (50 isolates), and wild birds (50 isolates). Human isolates were obtained from the Santa Creu i Sant Pau Hospital (Barcelona) strain collection. Broiler chicken isolates belong to a Campylobacter strain collection at IRTA-CReSA, collected at different slaughterhouses located in Barcelona, Lleida, and Tarragona (Catalonia). The wild bird samples, collected in Catalonia and Alboran Island, were obtained from six different species (Spatula clypeata, Ciconia ciconia, Corvus corax, Columba livia, Larus michahellis, and Larus audouinii). The strain 81-176 (ATCC BAA-2151) was used as the reference strain [13]. Complete genomes of Campylobacter jejuni strains were recovered from an NCBI database. Isolates were cultured onto Columbia blood agar (CBA) plates (Scharlau) and incubated at 42 ◦ C for 48 h in microaerophilic conditions (CampyGen, Oxoid), unless otherwise stated. 2.2. PCR Amplification Genomic DNA was extracted from cultures grown onto CBA plates using the InstaGene Matrix Kit (Bio-Rad Laboratoires). PCR reactions (PCR Master Mix x2, Thermo Scientific) were performed using 35 ng of DNA as a template and the presence of wlaN and cgtB genes were tested using the specific primers indicated in Table 1. Primers used to amplify the housekeeping gene gltA were included in the Microorganisms 2020, 8, 325 3 of 11 PCR mixtures as an internal control of the PCR reaction. Amplified fragments of wlaN and cgtB were purified and sequenced with the same primers that were used in the PCR reaction. Table 1. Primers used in this work. Primer Sequence 5’–3’ PCR product (bp) Reference cgtB-F TTAAGAGCAAGATATGAAGGT 740 [12] cgtB-R GCACATAGAGAACGCTACAA wlaN-F TGCTGGGTATACAAAGGTTGTG 561 [16] wlaN-R AGGTCCATTACCGCATACCA gltA-F GCCCAAAGCCCATCAAGCGGA 141 [17] gltA-R GCGCTTTGGGGTCATGCACA 2.3. Analysis of LOS Cell biomass from one CBA plate was resuspended in 1 ml of PBS and the OD600 of the cell suspension determined. After centrifugation, the cells were resuspended in Laemmli buffer, adjusting the final volume accordingly to the OD600 of the cell suspension (OD600 × 0.5) (whole cell extracts). LOS was isolated as described previously [18]. Briefly, samples were incubated at 100 ◦ C for 10 min and centrifuged at 10,000× g for 10 min. A 50 µl aliquot of the supernatant was treated with proteinase K (20 mg/ml; 2 h at 65 ◦ C). LOS samples were fractionated by 15% SDS-PAGE and visualized by a carbohydrate specific silver staining method [16]. LOSSIAL was detected by Western blot using a peroxidase-conjugated labelled cholera toxin B (HRP-CT), as previously described [12]. Coomasie staining of whole cell extracts was used as the loading control. 2.4. Invasion Assay Adherence and invasion assays were performed using human colonic carcinoma (Caco-2) cells, as previously described [19]. Caco-2 cells were seeded in 24-well plates at 2 × 104 cells per well and incubated for 8 days at 37 ◦ C. Bacteria, grown on CBA plates under microaerophilic conditions for 24 h at 37 ◦ C, were resuspended in PBS plus 1% inactivated-FBS (PBS-F) and the bacterial concentration was adjusted at approximately 2 × 108 cfu/ml (OD600 of 0.04). Confluent monolayers of Caco-2 cells were washed once with PBS and infected with 0.5 ml of the bacterial suspension. To allow bacterial adherence and internalization, monolayers and bacteria were incubated for 3 h at 37 ◦ C and 5% CO2 in a humified atmosphere. For total cell-associated bacteria (intracellular and adhered) quantification, the unbound bacteria were removed from cell monolayers by washing with PBS, the cells were lysed with 0.5 ml Triton X-100 (1%) for 10 min, and the total cell-associated bacteria was determined by serial dilutions on CBA plates. For intracellular bacteria quantification, infected monolayers were washed with PBS and incubated in 0.5 ml of fresh PBS-F with 150 ml/ml gentamycin (Sigma) to kill extracellular bacteria. After 2 h at 37 ◦ C, cells were washed and lysed following the same procedure as for the total cell-associated bacteria. The amount of both intracellular and total cell-associated bacteria (intracellular and adhered) was determined in triplicate assays. The invasion index was calculated as the percentage of intracellular bacteria relative to the total cell-associated bacteria. 3. Results 3.1. Differential Prevalence of wlaN and cgtB Genes in C. jejuni Isolates from Human Patients, Broiler Chickens, and Wild Birds The presence of the wlaN and cgtB genes in a collection of 150 C. jejuni strains from human patients (50), broiler chickens (50), and wild birds (50) was determined by PCR. A similar percentage of strains were positive for LOSSIAL related genes (wlaN+ and cgtB+ ) among human (28%) and broiler chicken (22%) strains. The percentage increased to 40% among wild bird strains. Interestingly, more striking 3.1. Differential Prevalence of wlaN and cgtB Genes in C. jejuni Isolates from Human Patients, Broiler Chickens, and Wild Birds The presence of the wlaN and cgtB genes in a collection of 150 C. jejuni strains from human patients (50),2020, Microorganisms broiler 8, 325chickens (50), and wild birds (50) was determined by PCR. A similar percentage 4 of 11 of strains were positive for LOSSIAL related genes (wlaN+ and cgtB+) among human (28%) and broiler chicken (22%) strains. The percentage increased to 40% among wild bird strains. Interestingly, more differences striking exist in the differences prevalence exist of each specific in the prevalence of eachgene depending specific on the origin gene depending of the on the strains origin (Figure of the strains1). The wlaN gene was more frequently detected among human and broiler chicken (Figure 1). The wlaN gene was more frequently detected among human and broiler chicken strains strains (20% and 16%, respectively) (20% than the cgtB and 16%, respectively) thangene the(8% cgtBand 6%,(8% gene respectively). In contrast, an and 6%, respectively). inverse distribution In contrast, an inverse was found among wild bird strains, with 34% of the strains cgtB + and only 6% wlaN + . Therefore, distribution was found among wild bird strains, with 34% of the strains cgtB and only 6% wlaN+. + among the LOS SIAL proficient strains, the wlaN gene was responsible Therefore, among the LOSSIAL proficient strains, the wlaN gene for wastheresponsible LOS modification for theinLOS72% of the humaninand modification 72% broiler of thechicken human andstrains, whereas broiler chicken cgtB gene thestrains, was responsible whereas the cgtB gene in was 85%responsible of the wild bird strains. in 85% of the wild bird strains. Figure 1. 1. Prevalence PrevalenceofofwlaN wlaN and and cgtB cgtB genes genes in C.injejuni C. jejuni strainsstrains from different from different origins.origins. H, human H,patients; human patients; B, broiler B, broiler chickens; chickens; W, wild W, wild birds. birds. analyses Statistical Statisticalwere analyses were performed performed usingchi-squared using Pearson’s Pearson’s chi‐ test squared (R Studiotest software). p