Selective colonization of microplastics, wood and glass by antimicrobial-resistant and pathogenic bacteria

Abstract

The Plastisphere is a novel niche whereby microbial communities attach to plastic debris, including microplastics. These communities can be distinct from those found in the surrounding environment or those attached to natural substrates and may serve as a reservoir of both pathogenic and antimicrobial-resistant (AMR) bacteria. Owing to the frequent omission of appropriate comparator particles (e.g. natural substrates) in previous studies, there is a lack of empirical evidence supporting the unique risks posed by microplastics in terms of enrichment and spread of AMR pathogens. This study investigated selective colonization by a sewage community on environmentally sampled microplastics with three different polymers, sources and morphologies, alongside natural substrate (wood), inert substrate (glass) and free-living/planktonic community controls. Culture and molecular methods (quantitative polymerase chain reaction (qPCR)) were used to ascertain phenotypic and genotypic AMR prevalence, respectively, and multiplex colony PCR was used to identify extra-intestinal pathogenic Escherichia coli (ExPECs). From this, polystyrene and wood particles were found to significantly enrich AMR bacteria, whereas sewage-sourced bio-beads significantly enriched ExPECs. Polystyrene and wood were the least smooth particles, and so the importance of particle roughness on AMR prevalence was then directly investigated by comparing the colonization of virgin vs artificially weathered polyethylene particles. Surface weathering did not have a significant effect on the AMR prevalence of colonized particles. Our results suggest that the colonization of plastic and non-plastic particles by AMR and pathogenic bacteria may be enhanced by substrate-specific traits.

Keywords: Escherichia coli; Plastisphere; antimicrobial resistance; biofilm; extra-intestinal pathogen; microplastic; wastewater.

Fig. 1.. Schematic overview of particle inoculation protocol. S, particle that sunk during inoculation and F, particle that floated during inoculation. Created in BioRender. Stevenson, E. (2024) BioRender.com/q54k724

Fig. 2.. Left: total c.f.u. ml⁻¹ (i.e. all colony phenotypes counted on non-antibiotic-supplemented agar plates, square root transformed) (top) and 16S rRNA gene copy number (square root transformed) (bottom) for biofilm communities (biological replicate=6). *P<0.05, **P<0.01 and ****P<0.0001 pairwise test (Kruskal–Wallis followed by Dunn’s post hoc test and FDR adjusted for multiple comparisons). Right: relative abundance of E. coli, coliforms and non-coliforms in cultured communities extracted from inoculated particles and liquid phase.

Fig. 3.. Phenotypic resistance prevalence for particle biofilm communities and liquid phase (biological replicate=6). *P<0.05 and **P<0.01 pairwise test (Kruskal–Wallis followed by Dunn’s post hoc test and FDR adjusted for multiple comparisons). Top: ampicillin, middle: ciprofloxacin and bottom: trimethoprim.

Fig. 4.. intI1 standardized to 16S rRNA prevalence (biological replicate=6) for particle biofilm communities and liquid phase. *P<0.05, **P<0.01 and ***P<0.001 pairwise test (Kruskal–Wallis followed by Dunn’s post hoc test and FDR adjusted for multiple comparisons).

Fig. 5.. ExPEC prevalence (biological replicate=6) for biofilm communities and liquid phase. *P<0.05, **P<0.01 and *** P<0.001 pairwise test (Kruskal–Wallis followed by Dunn’s post hoc test and FDR adjusted for multiple comparisons).

Fig. 6.. Left: total c.f.u. mL⁻¹ (top) and 16S rRNA gene copy number (square root transformed) (bottom) for virgin and weathered biofilm communities (biological replicate=6). Right: relative abundance of E. coli, coliforms and non-coliforms in cultured communities extracted from inoculated particles and liquid phase. vLP, virgin liquid phase and wLP, weathered liquid phase.

Fig. 7.. Phenotypic resistance prevalence for particle biofilm communities and liquid phase (biological replicate=6). *P<0.05 and **P<0.01 pairwise test (Kruskal–Wallis followed by Dunn’s post hoc test and FDR adjusted for multiple comparisons). Top: ampicillin, middle: ciprofloxacin and bottom: trimethoprim. vLP, virgin liquid phase and wLP, weathered liquid phase.

Fig. 8.. intI1 standardized to 16S rRNA prevalence (biological replicate=6) for Plastisphere communities and liquid phase. *P<0.05 and **P<0.01 pairwise test (Kruskal–Wallis followed by Dunn’s post hoc test and FDR adjusted for multiple comparisons). vLP, virgin liquid phase and wLP: weathered liquid phase.