Impact of fluoroquinolone and heavy metal pollution on antibiotic resistance maintenance in aquatic ecosystems

Abstract

Background: Freshwater pollution with compounds used during anthropogenic activities could be a major driver of antibiotic resistance emergence and dissemination in environmental settings. Fluoroquinolones and heavy metals are two widely used aquatic pollutants that show a high stability in the environment and have well-known effects on antibiotic resistance selection. However, the impact of these compounds on antibiotic resistance maintenance in aquatic ecosystems remains unknown. In this study, we used a microcosm approach to determine the persistence of two fluoroquinolones (ciprofloxacin, ofloxacin) and two heavy metals (copper and zinc) in the Rhône river over 27 days. In addition, we established links between antibiotic and metal pollution, alone and in combination, and the composition of freshwater bacterial communities, the selection of specific members and the selection and maintenance of antibiotic and metal resistance genes (ARGs and MRGs) using a metagenomics approach.

Results: Whereas ofloxacin was detected at higher levels in freshwater after 27 days, copper had the strongest influence on bacterial communities and antibiotic and metal resistance gene selection. In addition, heavy metal exposure selected for some ARG-harboring bacteria that contained MRGs. Our research shows a heavy metal-driven transient co-selection for fluoroquinolone resistance in an aquatic ecosystem that could be largely explained by the short-term selection of Pseudomonas subpopulations harboring both fluoroquinolone efflux pumps and copper resistance genes.

Conclusion: This research highlights the complexity and compound-specificity of dose-response relationships in freshwater ecosystems and provides new insights into the medium-term community structure modifications induced by overall sub-inhibitory levels of antibiotic and heavy metal pollution that may lead to the selection and maintenance of antibiotic resistance in low-impacted ecosystems exposed to multiple pollutants.

Keywords: Antibiotic resistance; Aquatic ecosystems; Fluoroquinolone; Freshwater; Heavy metals; Maintenance; Metagenomics.

Fig. 1

Concentrations of ciprofloxacin, ofloxacin, copper and zinc measured in Rhône river water after 27 days of exposure to antibiotics, metals and both. n = 3. One non-polluted sample at day 0 (control day 0) was measured to determine background pollution levels (no statistics were performed on this sample). Only significant differences are shown. *p-value < 0.05. **p-value < 0.01. ***p-value < 0.001

Fig. 2

Non-metric multidimensional scaling of the Bray-Curtis dissimilarity calculated from the relative abundance of ASVs annotated to the class level. Dot color represents the five conditions used in the study, whereas dot size represents exposure time. NMDS stress = 0.166. n = 3

Fig. 3

Population dynamics over time and under different pollution scenarios of the 15 most abundant ASVs associated with the most abundant genus found in river water samples, Pseudomonas. A, B, C: total inferred abundance (ASV abundance divided by sequencing depth and multiplied by the number of copies of the 16 S rRNA gene) of the three most abundant ASVs (ASV 1, ASV 25, ASV 28). D: total inferred abundance of the remaining 12 ASVs. AB1: antibiotics one dose. ABmult: antibiotics multiple doses. AB + M: antibiotics + metals. D0: day 0; D3: day 3. D6: day 6. D27: day 27. n = 3

Fig. 4

Relative abundance (normalized by sequencing depth) of antibiotic and metal resistance genes in river water metagenomes. A: relative abundance of total antibiotic resistance genes (ARGs). B: relative abundance of fluoroquinolone resistance genes. C: relative abundance of copper resistance genes. D: relative abundance of zinc resistance genes. Only significant differences are shown. *p-value < 0.05. **p-value < 0.01. ***p-value < 0.001. **** p-value < 0.0001. n = 3

Fig. 5

Relative abundance of metagenome assembled genomes (MAGs) selected by metals in the presence and/or in the absence of antibiotics. A: MAG associated with Pseudomonas.B: MAG associated with Prosthecobacter.C: MAG associated to Aquisediminimonas. Relative abundance was calculated using the percentage of recruitment of each MAG across samples, which represents the percentage of sequences from a sample recruited into a MAG. Only significant differences are shown. *p-value < 0.05. **p-value < 0.01. ***p-value < 0.001. **** p-value < 0.0001. n = 3