Urban green roofs are increasingly installed to support biodiversity and deliver ecosystem services in cities, but their role in shaping environmental antimicrobial resistance is not well understood. This study applied long-read metagenomic sequencing to an experimental extensive green roof system to characterize the resident antimicrobial resistance genes (ARGs). The experimental design included four management regimes intended to test how vegetation and the addition of organic amendments (green waste) influence the roof substrate resistome. Green waste is commonly used to improve substrate quality but has been suggested as a potential source of ARGs; the study therefore aimed to resolve whether management choices promote clinically relevant resistance or shape an intrinsic, environmentally structured resistome.
The authors sampled plots from an experimental extensive green roof system with four distinct management regimes that varied in vegetation and organic amendment (green waste) application. They used long-read metagenomic sequencing to detect and characterize ARGs in the substrate microbiome. The approach enabled assignment of ARGs to genomic contexts (chromosomal or plasmid) and to associated taxa. The experimental details and exact sequencing workflows are reported in the source article.
Across the four management regimes the researchers detected a total of 62 ARGs. The resistome was dominated by target-modification and mixed mechanisms of resistance rather than by efflux-based multidrug resistance. Specific gene examples highlighted in the results included bacA (conferring resistance to bacitracin) and multiple genes conferring resistance to rifamycin (arr, rox, rph). The pattern indicates prevalence of resistance to naturally occurring antibiotics as opposed to the multidrug efflux mechanisms more commonly associated with anthropogenic contamination.
Most ARGs identified were chromosomally encoded. Only two ARGs were located on plasmids, indicating limited plasmid-borne mobility within the sampled resistomes. Taxonomic associations linked the ARGs primarily to non-pathogenic environmental taxa, rather than to established clinical pathogens. This genomic and taxonomic context supports the interpretation that the detected resistome reflects intrinsic environmental resistance pools.
Management regime had a statistically significant effect on several resistome metrics. Specifically, the regime influenced ARG richness (number of different ARGs detected), ARG composition (which ARGs were present), and plasmid abundance. However, the average ARG abundance normalized to genome size did not differ across management treatments, indicating that while composition and diversity shifted, per-genome ARG load remained similar between regimes.
Two ARGs, aph3-II and tlmA, were found to be enriched in the unamended samples (those without green-waste amendments). These genes were carried by oligotrophic bacteria, suggesting that microbial competition in nutrient-limited substrates may select for particular resistance determinants. Overall, the resistome observed on green roof substrates was driven by ecological constraints—nutrient availability, microbial competition, and habitat conditions—rather than by selection pressures typically associated with anthropogenic pollutants and clinical antibiotic use.
The authors interpret their findings to indicate that green roof management shapes a substrate resistome dominated by environmental, chromosomally encoded ARGs linked to natural antibiotic production and ecological interactions. Because the detected ARGs were primarily associated with non-pathogenic taxa and not with multidrug efflux systems commonly co-selected by pollution, the resistome on these urban green roofs does not appear to be dominated by clinical pathogenic resistance. Nonetheless, the paper emphasizes that additional research is required to fully assess potential risks and to guide safe incorporation of green roofs into urban planning within a One Health framework.
The source article notes the need for further research to evaluate potential risks associated with green roof resistomes. Specific experimental details, sample sizes, geographic scope, and temporal dynamics are reported in the full article and may constrain generalizability; where methodological specifics were not included in the abstract, the source should be consulted for full details. The authors recommend follow-up studies to better quantify transfer potential, temporal stability of ARGs, and any conditions that might promote movement toward clinically relevant hosts.
In this experimental assessment, urban green roofs hosted an intrinsic resistome of 62 ARGs shaped by management regime and ecological context. The resistome was characterized by chromosomal, target-modification and mixed resistance mechanisms affecting naturally occurring antibiotics (examples: bacA, arr, rox, rph), limited plasmid localization, and associations with non-pathogenic environmental taxa. Management altered ARG richness, composition, and plasmid abundance but did not change genome size–normalized ARG abundance. The collective findings suggest that green roof resistomes are governed by ecological constraints rather than by clinical selection pressures; however, further work is needed to determine risk pathways and to support the safe expansion of green roofs under a One Health perspective.
Conflict of interest and ethics: the authors declared no competing interests and noted ethics approval and consent were not applicable.