Whole-genome sequencing and gene sharing network analysis powered by machine learning identifies antibiotic resistance sharing between animals, humans and environment in livestock farming

Abstract

Anthropogenic environments such as those created by intensive farming of livestock, have been proposed to provide ideal selection pressure for the emergence of antimicrobial-resistant Escherichia coli bacteria and antimicrobial resistance genes (ARGs) and spread to humans. Here, we performed a longitudinal study in a large-scale commercial poultry farm in China, collecting E. coli isolates from both farm and slaughterhouse; targeting animals, carcasses, workers and their households and environment. By using whole-genome phylogenetic analysis and network analysis based on single nucleotide polymorphisms (SNPs), we found highly interrelated non-pathogenic and pathogenic E. coli strains with phylogenetic intermixing, and a high prevalence of shared multidrug resistance profiles amongst livestock, human and environment. Through an original data processing pipeline which combines omics, machine learning, gene sharing network and mobile genetic elements analysis, we investigated the resistance to 26 different antimicrobials and identified 361 genes associated to AR phenotypes; 58 of these were known AMR-associated genes and 35 were associated to multidrug resistance. We uncovered an extensive network of genes, correlated to AMR phenotypes, shared among livestock, humans, farm and slaughterhouse environments. We also found several human, livestock and environmental isolates sharing closely related mobile genetic elements carrying AR genes across host species and environments. In a scenario where no consensus exists on how antibiotic use in the livestock may affect antibiotic resistance in the human population, our findings provide novel insights into the broader epidemiology of antimicrobial resistance in livestock farming. Moreover, our original data analysis method has the potential to uncover AR transmission pathways when applied to the study of other pathogens active in other anthropogenic environments characterised by complex interconnections between host species.

Dissecting microbial communities and resistomes for interconnected humans, soil, and livestock

“…A subset of these samples, 46 chicken faeces samples and 30 human faeces samples, were in common between the present study and Peng et al [24]. Compared with the metagenome data we study here, the whole genome study [24] found stronger evidence of clustering between isolates from different hosts, however, hotspots of similarities were found between samples from livestock and the nose and hands of humans. However, nose and hands were excluded from this study as there were no corresponding metagenome samples.…”

“…Whilst the gut microbiota is largely commensal and important for health, several studies have found that the microbiota can act as a reservoir for ARGs [16,17], which in other studies have been found to be transferred between livestock and humans [18]. Several studies have found both direct (in food production) [19][20][21][22][23][24][25] and indirect (through food consumption) [26][27][28] evidence of similar antibiotic-resistant bacteria (ARBs) and ARGs between humans and animals/meat. However, the role of farm animals in the emergence and dissemination of ARBs and their resistance determinants to humans is poorly understood and controversial, with some studies comparing samples related on a regional level showing distinct human and animal lineages [29,30].…”

“…Culture-based methods using whole genome sequencing combined with antibiotic susceptibility testing (AST) and machine learning (ML) have proved to be a powerful prediction tool for genomic features associated with AMR [24,[33][34][35][36][37]. However, these approaches are limited to only those bacteria that are easily culturable, and in practice are focused on a few selected species [38].…”

“…Although food production settings are emphasized as high-risk transmission points [ 1 ], only a few studies to date have considered the AMR present in livestock, the humans in direct contact with them (i.e. farm/slaughterhouse workers), and their environment [ 19 – 22 , 24 , 25 ]. In addition, most of these investigations have been conducted in HICs [ 19 – 22 ] and there are few studies in LMICs, where risks are likely to be higher [ 1 ].…”

PhyGraFT: a network-based method for phylogenetic trait analysis

“…Moreover, antigenic shift with reassortment further complicates the evolutionary history. Additionally, there are other cases in which we can quantify similarities but cannot convert these into evolutionary distances, such as gene-sharing networks (14, 15) or protein structure similarity networks (16). Such networks appear not only for studying typical evolution but also for various research into bacterial evolution, including horizontal gene transfer (17) and antibody evolution (18).…”