The sequencing, assembly, and analysis of bacterial genomes is central to tracking and characterizing foodborne pathogens. The bulk of bacterial genome sequencing at the US Food and Drug Administration is performed using short-read Illumina MiSeq technology, resulting in highly accurate but fragmented genomic sequences. The MinION sequencer from Oxford Nanopore is an evolving technology that produces long-read sequencing data with low equipment cost. The goal of this study was to compare Campylobacter genome assemblies generated from MiSeq and MinION data independently, as well as hybrid genome assemblies combining both data types. Two reference strains and two field isolates of C. jejuni were sequenced using MiSeq and MinION, and the sequence data were assembled using the software programs SPAdes and Canu, respectively. Hybrid genome assembly was performed using the program Unicycler. Comparison of the C. jejuni 81-176 and RM1221 genome assemblies to the PacBio reference genomes revealed that the SPAdes assemblies had the most accurate nucleotide identity, while the hybrid assemblies were the most contiguous. Assemblies generated only from MinION data using Canu were the least accurate, containing many indels and substitutions that affected downstream analyses. The hybrid sequencing approach was the most useful for detecting plasmids, large genome rearrangements, and repetitive elements such as rRNA and tRNA genes. The full genomes of both C. jejuni field isolates were completed and circularized using hybrid sequencing, and a plasmid was detected in one isolate. Continued development of nanopore sequencing technologies will likely enhance the accuracy of hybrid genome assemblies and enable public health laboratories to routinely generate complete circularized bacterial genome sequences.

Campylobacter jejuni causes more than 2 million cases of gastroenteritis annually in the United States, and is also linked to the autoimmune sequelae Guillan–Barre syndrome (GBS). GBS often results in flaccid paralysis, as the myelin sheaths of nerve cells are degraded by the adaptive immune response. Certain strains of C. jejuni modify their lipooligosaccharide (LOS) with the addition of neuraminic acid, resulting in LOS moieties that are structurally similar to gangliosides present on nerve cells. This can trigger GBS in a susceptible host, as antibodies generated against C. jejuni can cross-react with gangliosides, leading to demyelination of nerves and a loss of signal transduction. The goal of this study was to develop a quantitative PCR (qPCR) method and use whole genome sequencing data to detect the Campylobacter sialyltransferase (cst) genes responsible for the addition of neuraminic acid to LOS. The qPCR method was used to screen a library of 89 C. jejuni field samples collected by the Food and Drug Administration Pacific Northwest Lab (PNL) as well as clinical isolates transferred to PNL. In silico analysis was used to screen 827 C. jejuni genomes in the FDA GenomeTrakr SRA database. The results indicate that a majority of C. jejuni strains could produce LOS with ganglioside mimicry, as 43.8% of PNL isolates and 46.9% of the GenomeTrakr isolates lacked the cst genes. The methods described in this study can be used by public health laboratories to rapidly determine whether a C. jejuni isolate has the potential to induce GBS. Based on these results, a majority of C. jejuni in the PNL collection and submitted to GenomeTrakr have the potential to produce LOS that mimics human gangliosides.

The three Campylobacter targets were simultaneously identified using artificially mixed genomic DNA and spiked raw milk. This SmartCycler-based multiplex qPCR is a rapid, specific, and sensitive method to identify C. jejuni, C. coli, and C. lari.

As the most prevalent bacterial cause of human gastroenteritis, food-borne Campylobacter infections pose a serious threat to public health. Whole Genome Sequencing (WGS) is a tool providing quick and inexpensive approaches for analysis of food-borne pathogen epidemics. Here we report the WGS and annotation of a Campylobacter coli strain, FNW20G12, which was isolated from milk in the United States in 1997 and carries multidrug resistance. The draft genome of FNW20G12 (DDBJ/ENA/GenBank accession number LWIH00000000) contains 1, 855,435 bp (GC content 31.4%) with 1902 annotated coding regions, 48 RNAs and resistance to aminoglycoside, beta-lactams, tetracycline, as well as fluoroquinolones. There are very few genome reports of C. coli from dairy products with multidrug resistance. Here the draft genome of FNW20G12, a C. coli strain isolated from raw milk, is presented to aid in the epidemiology study of C. coli antimicrobial resistance and role in foodborne outbreak.

Consumption of Campylobacter-contaminated food is one of the most common causes of bacterial diarrhea. A previously developed quantitative polymerase chain reaction (qPCR) utilizing the SmartCycler instrument platform for identification of Campylobacter jejuni, Campylobacter coli, and Campylobacter lari had to be modified to address the recent discontinuation of the SmartCycler system. In this study, a multiplex qPCR assay was optimized on the Applied Biosystems 7500 Fast (AB7500F) platform to continue using qPCR for the identification of three target Campylobacter spp. AB7500F qPCR efficiencies obtained by testing reference genomic DNA (gDNA) were 90.9%, 86.4%, and 94.6% for C. jejuni, C. coli, and C. lari, respectively, with all correlation coefficient values >0.99. The qPCR results exhibited 100% specificity by testing gDNA samples from 37 non-target reference strains and 86 target strains (50 C. jejuni, 27 C. coli, and 9 C. lari strains) in this study. The lowest detection level using gDNA was 4, 7, and 2 genome copies per reaction for C. jejuni, C. coli, and C. lari, respectively. With a 2-day enrichment procedure, the qPCR method correctly detected target species in a spiked food matrix (frog leg, an aquaculture product). The sensitivity in 25 g food matrix was 4 colony-forming units (CFUs) for C. jejuni, 3 CFUs for C. coli, and 2 CFUs for C. lari. The results suggest that this AB7500F-based qPCR has potential applications for the identification of C. jejuni, C. coli, and C. lari in contaminated food.