What Is the Role of the Environment in the Emergence of Novel Antibiotic Resistance Genes? A Modeling Approach

Abstract

It is generally accepted that intervention strategies to curb antibiotic resistance cannot solely focus on human and veterinary medicine but must also consider environmental settings. While the environment clearly has a role in transmission of resistant bacteria, its role in the emergence of novel antibiotic resistance genes (ARGs) is less clear. It has been suggested that the environment constitutes an enormous recruitment ground for ARGs to pathogens, but its extent is practically unknown. We have constructed a model framework for resistance emergence and used available quantitative data on relevant processes to identify limiting steps in the appearance of ARGs in human pathogens. We found that in a majority of possible scenarios, the environment would only play a minor role in the emergence of novel ARGs. However, the uncertainty is enormous, highlighting an urgent need for more quantitative data. Specifically, more data is most needed on the fitness costs of ARG carriage, the degree of dispersal of resistant bacteria from the environment to humans, and the rates of mobilization and horizontal transfer of ARGs. This type of data is instrumental to determine which processes should be targeted for interventions to curb development and transmission of ARGs in the environment.

Keywords: human and animal health; mobile genetic elements; mobilization; origin of antibiotic resistance genes; pathogenic bacteria.

Figure 1

Overview of the model framework. Processes are framed in bold. Arrows display the influence of different parameters on the processes. Colored arrows represent the pathways included in the model (E₁ to E₆).

Figure 2

Valid parameter ranges (A) and process rates (B) for the main model after 70 years of simulated time. The range of S extends from around 0.99 to 1.002. Since every parameter has its own definition, the parameter values in (A) have slightly different meanings (see Table 2).

Figure 3

Correlations between the variables in the main model after 70 years of simulated time. Blue colors represent negative correlation values, red colors represent positive associations, and white indicates unrelated parameters. totalEmg represents the total number of ARGs that have emerged on MGEs in human pathogens.

Figure 4

Time dependence of model process rates (A) and key model parameters E (B) and S (C) over the simulated time span from the start of antibiotic use to the emergence of ARGs in human pathogens. ARG contribution is expressed as the number of ARGs originating from each process (maximum of 2200 at 10,000 days) in (A). Dotted lines in (A) and dashed lines in (B) and (C) represent the range in which 95% of the values fall in the simulations. Lines with dots and dashes in (B) and (C) represent the range in which 50% of the values fall in the simulations.