Clinically relevant antibiotic resistance genes are linked to a limited set of taxa within gut microbiome worldwide

Abstract

The acquisition of antimicrobial resistance (AR) genes has rendered important pathogens nearly or fully unresponsive to antibiotics. It has been suggested that pathogens acquire AR traits from the gut microbiota, which collectively serve as a global reservoir for AR genes conferring resistance to all classes of antibiotics. However, only a subset of AR genes confers resistance to clinically relevant antibiotics, and, although these AR gene profiles are well-characterized for common pathogens, less is known about their taxonomic associations and transfer potential within diverse members of the gut microbiota. We examined a collection of 14,850 human metagenomes and 1666 environmental metagenomes from 33 countries, in addition to nearly 600,000 isolate genomes, to gain insight into the global prevalence and taxonomic range of clinically relevant AR genes. We find that several of the most concerning AR genes, such as those encoding the cephalosporinase CTX-M and carbapenemases KPC, IMP, NDM, and VIM, remain taxonomically restricted to Proteobacteria. Even cfiA, the most common carbapenemase gene within the human gut microbiome, remains tightly restricted to Bacteroides, despite being found on a mobilizable plasmid. We confirmed these findings in gut microbiome samples from India, Honduras, Pakistan, and Vietnam, using a high-sensitivity single-cell fusion PCR approach. Focusing on a set of genes encoding carbapenemases and cephalosporinases, thus far restricted to Bacteroides species, we find that few mutations are required for efficacy in a different phylum, raising the question of why these genes have not spread more widely. Overall, these data suggest that globally prevalent, clinically relevant AR genes have not yet established themselves across diverse commensal gut microbiota.

Fig. 1. Distribution of clinically relevant AR genes across global human-associated microbiomes globally and cultured isolates.

The top heatmap shows the prevalence of AR gene families in human gut-associated microbiomes sequenced from each country. Immediately below is a heatmap showing the total prevalence in gut, skin and oral microbiomes. The bottom heatmap shows the prevalence of AR gene families in sequenced isolates by taxonomic class. Color scales represent the percent of metagenomes (top) or isolates (bottom) for which the gene family was detected. At right of the top heatmap are two barplots, one showing the proportion of samples from heathy (blue) and not-healthy (red) individuals, as classified by curatedMetagenomicDataCuration, and a second showing the total number of samples from each country. At right of the bottom heatmap are two barplots, the first showing the proportion of isolates from pathogenic (yellow), non-pathogenic (red), or unannotated species (gray), as determined by CARD, and the second showing the total number of isolates surveyed per class. At the bottom, a barplot shows the total number of AR gene varients in each gene family (light orange) and the number detected in this study (dark orange) in the metagenomic datasets. Nearly all genes in each family shown were identified in the isolate genomes.

Fig. 2. Time allows for genes to become more prevalent and spread more broadly.

A, B For each gene family, the time since initially reported in the scientific literature, a proxy for AR gene emergence, is plotted against the prevalence in human gut microbiomes (left) and number of genera in which the gene is found (right). Marker color/shape indicate the AR gene class. Two-sided Pearson’s ρ, p-value and R² are reported. C Statistics (two-sided Pearson’s ρ, p-value and R²) are reported for the years since AR gene emergence and taxonomic spread computed at different taxonomic levels.

Fig. 3. OIL-PCR confirms that select AR genes are taxonomically restricted.

A A heatmap showing the number and percentage of gut microbiome samples that were positively screened for each AR gene, by quantitative PCR (for Vietnam, India and Pakistan) or metagenomic read alignment (Honduras). Bolded AR genes were further screened for taxonomic associations by OIL-PCR. Dark gray boxes indicate qPCR reactions that were not performed. B OIL-PCR was performed to detect taxonomic associations for CTX-M, AAC-(6’)-IB, and qnrS in positive gut microbiome samples from the international cohorts. Each row represents an individual’s gut microbiome sample. Colors represent AR gene variants. Phylogenies show the relationships between detected ASVs. The percent similarity is listed. To the right of each sample is a rank-ordered distribution of abundances of the individual ASVs in the individual’s gut microbiome, as determined by 16 S rRNA sequencing. Lines, colored according to the species designated in the phylogenies, are placed at the rank of the ASVs detected by OIL-PCR in each sample. C OIL-PCR results performed in cfiA, cblA, and cepA, displayed in the same manner as in (B). 31 individuals’ samples are shown in (B) and (C), with two individuals having been tested for the genes in both (B) and (C).

Fig. 4. Bacteroides-specific beta-lactamases CfiA and CblA are functional in E. coli.

A AR genes TEM, cfiA, cepA, cblA, and cfxA were cloned onto plasmids either with no promoter (TEM only), their native Bacteroides promoter, or a synthetic E. coli promoter, transformed into E. coli and tested on ampicillin to determine their minimum inhibitory concentrations (MIC). B MICs to various antibiotics were determined using antibiotic strips on agar for E. coli carrying cfiA and cblA, placed under its native Bacteroides promoter (yellow) and an E. coli-specific promoter (purple). E. coli without plasmid were used as a control. C E. coli tested for growth on agar plates treated with cefotaxime strips. E. coli harboring no plasmid, or a plasmid harboring cfiA under control of its native B. fragilis promoter, the synthetic E. coli promoter, or a mutant library of B. fragilis cfiA promoters (Mutated). D Histogram of mutations within the B. fragilis native promoter identified in the mutated cfiA promoter library after 18 h of growth in 0.5 ug/ml of cefotaxime. The native B. fragilis promoter sequence is shown, as are the predicted B. fragilis and E. coli promoters in the native sequence, the mutated consensus sequence and a depiction of the canonical E. coli promoter.