Plastispheres as reservoirs of antimicrobial resistance: Insights from metagenomic analyses across aquatic environments

Abstract

Evidence suggests that plastic particles from various environments can accumulate harmful microorganisms and carry bacteria with antimicrobial resistance genes (ARGs). The so-called "plastisphere" might facilitate the spread of pathogens and antimicrobial resistance across environments, posing risks to human and animal health. This study aimed to analyze the diversity and abundance of ARGs found in plastispheres from various aquatic environments, identify clinically relevant pathogenic species, and ascertain bacterial hosts carrying ARGs. We present data from 36 metagenomes collected from plastispheres in different environments (freshwater, raw wastewater, and treated wastewater). The diversity and abundance of ARGs in the resistome of the plastispheres were analyzed through metagenomic methods. A total of 537 high-quality metagenomic-assembled genomes (MAGs) were constructed to identify clinically relevant pathogens and to link the detected ARGs to their bacterial hosts. The results show that the environment has the greatest influence on the abundance and diversity of ARGs in the plastispheres resistome, with the wastewater plastisphere containing a resistome with the highest diversity of ARGs. Resistance to beta-lactams, aminoglycosides, and tetracyclines were the most abundant resistance mechanisms detected in the different plastispheres. The construction of MAGs identified potential pathogens and environmental bacteria that confer resistance to one or several drug classes, with beta-lactams being the most pervasive form of AMR detected. This work enhances our understanding of the plastisphere's role in antimicrobial resistance dissemination and its ecological and public health risks.

Fig 1. The alpha diversity estimates of the resistance gene identified in the plastispheres from river water and wastewater.

Boxplots show the distribution of (A) Shannon diversity (richness and evenness) and (B) Simpson diversity (the abundance) values across different environments. The color indicates the different environments from which the plastispheres were collected, with individual samples represented as jittered points. The point shape represents the duration of incubation. The boxes show the interquartile range, with the horizontal line representing the median value.

Fig 2. Composition differences of the antimicrobial resistance genes in the plastispheres.

A non-metric multidimensional scaling (NMDS) plot illustrates the Bray-Curtis distances of the antimicrobial resistance genes composition in the plastispheres across the environments. The samples are shown with points and ellipses colored by the environment. A statistically significant PERMANOVA result (p < 0.001) indicates distinct ARG composition between the environments. The spread of points within each environment reflects variability in the composition of ARGs, supported by a statistically significant PERMDISP result (p = 7.354e-07).

Fig 3. Relative abundance of ARGs.

Fig 4. The number of ARGs found in the plastispheres.

(A) The box plot shows the total AMR load (RPKM) across the plastispheres from different environments. The number of ARGs in raw WW (n = 177) was statistically significantly higher compared to the number of ARGs from treated WW (n = 56, p. adj = 0.005) and river water plastispheres (Loc1: n = 22, p. adj = 9.9e-4. Loc 2: n = 9, p. adj = 9.9e-4). (B) The UpSet plot shows the number of unique and shared antimicrobial resistance genes detected in the plastispheres across different environments.

Fig 5. The relative abundance of ESKAPEE pathogens.

The boxplots illustrate the relative abundance (calculated from RPM = reads per million) of ESKAPEE pathogens for the different environments. The line within the boxes represents the median value. The upper and lower edges of the boxes represent the higher and lower quartiles, respectively. The single points on the diagram show all the data points included in the analysis.

Fig 6. Distribution of ARGs across MAGs in different environments.

The Sankey diagrams show the MAGs generated for each sampling location, their respective phyla across domains, and detected ARGs. No ARGs were detected in some MAGs, as indicated by the “No AMR Gene Detected” category. The number of MAGs corresponding to each node is shown in brackets for each node.

Fig 7. Antimicrobial resistance genes associated with bacterial hosts.

Network analysis represents identified MAGs in the plastispheres from the different environments: (A) Lier River Loc1, (B) Lier River Loc2, (C) Raw WW, and (D) Treated WW. The bacteria were grouped at the genus level, while species-level analysis was maintained for the ESKAPE group and E. coli. The bacterial genera are listed on the left side of the network. The identified genera are marked in blue, while the ESKAPEE pathogens are highlighted in orange. The associated antibiotic resistance drug classes are listed on the right side of the network and marked with different colors.