Culture-independent detection and characterisation of Mycobacterium tuberculosis and M. africanum in sputum samples using shotgun metagenomics on a benchtop sequencer

Abstract

Tuberculosis remains a major global health problem. Laboratory diagnostic methods that allow effective, early detection of cases are central to management of tuberculosis in the individual patient and in the community. Since the 1880s, laboratory diagnosis of tuberculosis has relied primarily on microscopy and culture. However, microscopy fails to provide species- or lineage-level identification and culture-based workflows for diagnosis of tuberculosis remain complex, expensive, slow, technically demanding and poorly able to handle mixed infections. We therefore explored the potential of shotgun metagenomics, sequencing of DNA from samples without culture or target-specific amplification or capture, to detect and characterise strains from the Mycobacterium tuberculosis complex in smear-positive sputum samples obtained from The Gambia in West Africa. Eight smear- and culture-positive sputum samples were investigated using a differential-lysis protocol followed by a kit-based DNA extraction method, with sequencing performed on a benchtop sequencing instrument, the Illumina MiSeq. The number of sequence reads in each sputum-derived metagenome ranged from 989,442 to 2,818,238. The proportion of reads in each metagenome mapping against the human genome ranged from 20% to 99%. We were able to detect sequences from the M. tuberculosis complex in all eight samples, with coverage of the H37Rv reference genome ranging from 0.002X to 0.7X. By analysing the distribution of large sequence polymorphisms (deletions and the locations of the insertion element IS6110) and single nucleotide polymorphisms (SNPs), we were able to assign seven of eight metagenome-derived genomes to a species and lineage within the M. tuberculosis complex. Two metagenome-derived mycobacterial genomes were assigned to M. africanum, a species largely confined to West Africa; the others that could be assigned belonged to lineages T, H or LAM within the clade of “modern” M. tuberculosis strains. We have provided proof of principle that shotgun metagenomics can be used to detect and characterise M. tuberculosis sequences from sputum samples without culture or target-specific amplification or capture, using an accessible benchtop-sequencing platform, the Illumina MiSeq, and relatively simple DNA extraction, sequencing and bioinformatics protocols. In our hands, sputum metagenomics does not yet deliver sufficient depth of coverage to allow sequence-based sensitivity testing; it remains to be determined whether improvements in DNA extraction protocols alone can deliver this or whether culture, capture or amplification steps will be required. Nonetheless, we can foresee a tipping point when a unified automated metagenomics-based workflow might start to compete with the plethora of methods currently in use in the diagnostic microbiology laboratory.

Whole genome sequencing Mycobacterium tuberculosis directly from sputum identifies more genetic diversity than sequencing from culture

“…M. tuberculosis WGS is usually performed on fresh or stored frozen cultured isolates to obtain sufficient purified mycobacterial DNA(23, 24). However, the culture process can change the population structure from that of the original sample due to genetic drift (random loss of lineages) and/or the selection of subpopulations more suited to growth in culture(25–27).…”

Autometa: Automated extraction of microbial genomes from individual shotgun metagenomes

“…This reference-free approach is moving the field of microbiology from phylogenetic profiling of communities and aggregate interpretation of metagenomic data to a higher resolution perspective of which organisms play what role in a given ecosystem. Such information can be invaluable in a diverse array of biotechnological applications, such as identifying the source of bioactive secondary metabolites in complex marine invertebrate communities (Piel et al , 2004;Miller et al , 2017;Lopera et al , 2017) , or antibiotic resistance mechanisms (Ashton et al , 2015;Shore et al , 2011;Moss et al , 2017;Loman et al , 2013;van der Helm et al , 2017) in uncultured clinical samples (Grumaz et al , 2016;Doughty et al , 2014) . However, despite the advances stemming from this paradigm shift in metagenomic analysis, a number of challenges remain.…”

Microbial bioinformatics 2020

“…Metagenomics as a diagnostic approach is likely to move closer to routine practice Doughty et al, 2014;Pallen, 2014;Wilson et al, 2014), but reliably disentangling pathogen genomes from metagenomesparticularly if short-read technologies still dominate the fieldis going to present a formidable challenge (Alneberg et al, 2014).…”