Needle in a haystack Using a direct sequencing approach to characterise Campylobacter in human stool

Abstract T6

Presenter: Bilal Djeghout (Quadram Institute Bioscience)

Authors: Bilal Djeghout, Steven Rudder, Thanh Le-Viet, Gemma Kay, Ngozi Elumogo, Andrew Page, John Wain, Nicol Janecko

Direct sequencing provides comprehensive functional profiles of mixed microbial communities enabling the study of organisms traditionally difficult to culture within a sample. Analytical sample processing and bioinformatic outputs for high resolution genome assembly are still restricted by sample characteristics, extraction techniques and limits of detection. The aim of this study was to optimise a method for characterising Campylobacter spp in stool metagenomes from Campylobacter-infected patients, in Norfolk, United Kingdom. An experimental method was adapted using criteria for depleting cell-free DNA, enriching microbial DNA and enhancing high molecular weight extraction to improve down-stream whole genome-based sequence analysis. Campylobacter jejuni, C. coli and C. lari presence was confirmed by PCR while real-time qPCR quantified the relative abundance of Campylobacter. Short- and long-read sequencing was conducted using Illumina Novaseq and Oxford Nanopore Technologies MinION platforms, respectively. A bespoke bioinformatics pipeline used tailored tools for metagenomes and Campylobacter. Nine samples were sequenced (Six Campylobacter PCR-positive; one negative using short-read sequencing; two PCR-positive samples using long-read sequencing). A total of seven samples (7/9) yielded Campylobacter abundance estimates of 5 to 500,000 cells/uL by real-time qPCR. The number of Campylobacter-specific sequenced reads ranged from 1,188 to 372,290. Samples with 50,000 or more cells in a metagenome sample resulted in Campylobacter assembled to >90% of a complete genome and represented 0.5% to 2% proportion of the total reads within a sample. From short-read sequences, Campylobacter was partially-assembled in four of the seven samples and, three (3/7) contained sufficient assemblies for characterisation. From the long-read sequences, data was not sufficient for characterisation. Of the three near-complete sequenced assemblies, two Campylobacter jejuni and one C. coli were characterised. C. jejuni genomes derived from direct sequencing were classified as ST61 and ST21, respectively with resistance gene blaOXA-61 in both genomes and 76 and 67 virulence indicators, respectively. The sample containing C. coli was classified as ST827 with blaOXA-61 and 73 virulence indicators. This study demonstrates ~50,000 of Campylobacter within a metagenome coupled with tailored direct sequencing methods yields sufficient information for near-complete Campylobacter genome assembly to characterise key attributes of this foodborne pathogen. Refinement to processing and post processing methods are ongoing and will further enhance the quality and quantity of direct sequencing information produced by short- and long-read sequencing. This study contributes to the understanding of Campylobacter using tailored metagenomic and informatic method approaches.

About the presenter

I am microbiologist with an interest in both biology and molecular biology of zoonotic pathogens within different ecological spheres. I joined the Quadram Institute Bioscience as a Post-doctoral research scientist in June 2019 to work on a Gates funded project studying the transmission of Campylobacter in rural Bangladesh. Previous to this, I was awarded a BSc in General Microbiology and a MSc in Molecular and Cellular Biology from the University of 8 Mai 45, Guelma, Algeria. My PhD project at the PhD School of Life Sciences and Biotechnologies at the University of Sassari, Italy focused on genetic characterization of Salmonella serovars isolated from human and poultry sources in four Algerian cities, including the determination of antimicrobial resistance using phenotypic and genotypic approaches (Illumina whole genome sequencing). In my first post-doctoral research scientist role at QIB in Prof. John Wains research group, I worked on designing methods for detection and isolation of Campylobacter utilising sequence- and culture-based approaches. I also worked on knowledge sharing with our collaborators in Child Health Research Foundation (CHRF) in Dhaka, Bangladesh, where I had been involved in setting-up the laboratory for the Oxford Nanopore Technology sequencing. In my new role as a post-doctoral research scientist in Dr. Janeckos research group, I continue to be involved with the transmission of Campylobacter in rural Bangladesh project, as well as optimising a targeted metagenomic culture-independent method for studying Campylobacter populations in different ecological spheres.

Presenting in Speaking session 3 - Epidemiology and public health