Towards a general model for predicting minimal metal concentrations co-selecting for antibiotic resistance plasmids

Abstract

Many antibiotic resistance genes co-occur with resistance genes for transition metals, such as copper, zinc, or mercury. In some environments, a positive correlation between high metal concentration and high abundance of antibiotic resistance genes has been observed, suggesting co-selection due to metal presence. Of particular concern is the use of copper and zinc in animal husbandry, leading to potential co-selection for antibiotic resistance in animal gut microbiomes, slurry, manure, or amended soils. For antibiotics, predicted no effect concentrations have been derived from laboratory measured minimum inhibitory concentrations and some minimal selective concentrations have been investigated in environmental settings. However, minimal co-selection concentrations for metals are difficult to identify. Here, we use mathematical modelling to provide a general mechanistic framework to predict minimal co-selective concentrations for metals, given knowledge of their toxicity at different concentrations.We apply the method to copper (Cu), Zinc (Zn), mercury (Hg), lead (Pb) and silver (Ag), predicting their minimum co-selective concentrations in mg/L (Cu: 5.5, Zn: 1.6, Hg: 0.0156, Pb: 21.5, Ag: 0.152). To exemplify use of these thresholds, we consider metal concentrations from slurry and slurry-amended soil from a UK dairy farm that uses copper and zinc as additives for feed and antimicrobial footbath: the slurry is predicted to be co-selective, but not the slurry-amended soil. This modelling framework could be used as the basis for defining standards to mitigate risks of antimicrobial resistance applicable to a wide range of environments, including manure, slurry and other waste streams..

Antimicrobial resistance in dairy slurry tanks: a critical point for measurement and control

“…ISEcp1CTX-M-15 elements). The model predicts that zinc resistance is highly prevalent, rising to fluctuate around 80%, with co-occurrence of tetracycline and zinc resistance, typically fluctuating between 10 and 15%, consistent with predictions that the metal concentrations in the tank are co-selective (69).…”

“…A mechanistic, multi-strain model of AMR in the slurry tank was constructed to simulate a range of relevant farm management scenarios that would have been impractical or unethical to carry out empirically. In brief, it is a coupled ordinary differential equation model of bacterial populations including logistic growth, death (baseline and antimicrobial induced), horizontal transfer and fitness cost of resistance, inflow and outflow (68,69). The model considered mobile resistance to penicillin, tetracycline, cephalexin, cefquinome, copper, and zinc, and was simulated for a full year in order to capture the recorded input of cephalexin and other antibiotics.…”

Antimicrobial Resistance Patterns of Escherichia coli Isolated from Sheep and Beef Farms in England and Wales: A Comparison of Disk Diffusion Interpretation Methods

“…Previous research indicates that the geochemical conditions of soils, particularly copper concentrations, are correlated with the abundance of antimicrobial resistance genes [40,41]. Furthermore, copper has been shown to co-select for tetracycline resistance in experimental conditions [42], and mathematical models suggest that this co-selection may occur at copper concentrations as low as 5.5 mg/mL [43]. Our results suggest that tetracycline non-susceptibility may be more prevalent on some farms due to the environmental exposure to copper in soils.…”

Influence of soil characteristics and metal(loid)s on antibiotic resistance genes in green stormwater infrastructure in Southern California

“…To assess the risk of metal(loid) co-selection on AR, the concept of minimum co-selective concentrations (MCC) was previously adopted and evaluated for many environmental compartments by Seiler and Berendonk (2012) and Arya et al (2021). More specifically, minimum metal(loid)s concentrations associated with increased ARG abundances were defined as MCC.…”

“…Since metal(loid) concentrations varied greatly in the environments, metal(loid)s exceeding their MCCs in their respective environments may be co-selective for ARGs. Currently, there are only a few studies reporting MCC datasets in soil environments (Arya et al, 2021;Seiler and Berendonk, 2012), with no MCC datasets considering GSI settings studied here. Therefore, the assessment of metal(loid) co-selective effects on AR in GSI soils is limited.…”