Analysis of global emergence and spread of antimicrobial resistance in 214k host and environmental samples

Research output: Book/ReportPh.D. thesis

130 Downloads (Pure)

Abstract

Antimicrobial resistance (AMR) has developed rapidly and now threatens to undermine our treatment of infectious diseases unless the spread of antimicrobial resistance genes (ARGs) is halted. AMR is a global challenge impacting human, animal, and planetary health, which calls for monitoring the abundance of ARGs across these three areas. Most of the current AMR surveillance systems only focus on clinical prevalences, failing to acknowledge that ARGs are also present in non-pathogenic microbes outside of hospital settings. In metagenomic sequencing, the goal is to recover genes for all organisms in a host or an environmental sample. That way, the abundance of ARGs can be quantified across all organisms in a sample, both unknown and known species.

Today, there is a vast amount of metagenomic sequencing datasets available in public repositories due to the good data-sharing practices during the academic publishing process. Most samples remain underutilized as few researchers have the computational and bioinformatic resources to analyze terabytes of data. However, the potential amount of information on microbial and AMR dynamics that can be extracted from these datasets using a standardized approach makes it worthwhile to explore. This has been the goal of this PhD project, where 214,095 metagenomic datasets were retrieved and analyzed to characterize the abundance of ARGs in host and environmental sources. With such a large pool of data, there are many questions on the distribution of ARGs that can be answered, and this project has focused on studying the differences in abundance for individual ARGs in local ecological settings and the overall co-occurrence of ARGs at a global scale. The three manuscripts enclosed in this PhD are presented below.

In Manuscript I, we carried out the download and processing of the 214,095 metagenomic datasets from the European Nucleotide Archive (ENA). Using the 442 · 1012 basepairs of sequencing reads, we aligned the reads against reference sequences from two databases, Silva and ResFinder, to determine the abundance of ARGs and bacterial genes. In this publication, we presented a brief characterization of overall trends in this collection and observed differences in resistome and microbiome compositions between different sample types. We also made the count data and the curated metadata available to promote the reuse of publicly available samples and further encourage sharing of raw sequencing data.

In Manuscript II, we studied the distribution of the family of mcr genes in the 214K collection of metagenomic samples. The mcr genes confer resistance to colistin, a lastresort antibiotic that is only used when all other treatment options fail. Our results confirmed that some of the mcr genes had spread around the world a while before being discovered. For example, we saw that the mcr-9 gene had been circulating for almost a decade before it was first reported. We also concluded that the differences in mcr abundances could largely be explained by the sampling source and location but that the genomic context of the mcr gene had not undergone significant changes. This manuscript confirmed the value of using publicly available metagenomic datasets for AMR surveillance and how the results can supplement existing surveillance programs.

In Manuscript III, following the characterization of only one group of genes in Manuscript II, we decided to investigate the abundance of all ARGs and how they cooccur. By inferring pairwise ARG correlations, we constructed correlation networks for different ecosystems that suggested that ARGs encoding resistance to different antimicrobials influences each other abundances. These observations suggest that using one antimicrobial in a specific environment induces the risk of resistance to multiple kinds of antimicrobials being indirectly selected, including resistance to the most critical antimicrobials for human medicine. We argued that the correlations could be used as risk profiles for guiding the safe use of antimicrobials in different settings.

Understanding how ARGs have spread through various ecological settings will be an effective tool in controlling and hopefully stopping the spread of AMR. The results presented in this thesis show how valuable it is to utilize sequencing datasets that are freely available in public repositories, coming one step closer to enabling global surveillance of AMR.
Original languageEnglish
Place of PublicationKgs. Lyngby
PublisherTechnical University of Denmark
Number of pages144
Publication statusPublished - 2022

Bibliographical note

The PhD was funded as part of the Global Surveillance of Antimicrobial Resistance project under the Novo Nordisk Foundation grant NNF16OC0021856 and as part of the VEO project that got funding from the European Union Horizon 2020 program under grant agreement 874735.

Fingerprint

Dive into the research topics of 'Analysis of global emergence and spread of antimicrobial resistance in 214k host and environmental samples'. Together they form a unique fingerprint.

Cite this