Evaluating effects of antibiotics across preclinical models of the human gastrointestinal microbiota

Abstract

While antibiotics play important roles in treating infections, disruption of the gastrointestinal microbiota during antibiotic treatment can lead to negative health consequences. However, for many antibiotics, the spectrum of activity has been determined for select isolates rather than for the range of microbes that populate the gastrointestinal tract. Here, we examined the response of communities of gastrointestinal microbes to antibiotics using two different model systems, human fecal minibioreactors and human microbiota associated mice. Communities established in minibioreactors using 12 different fecal donors were exposed to 12 different classes of antibiotics. Samples from three fecal donors were used to colonize germ-free mice from three different genetic backgrounds; progeny mice were then exposed to 6 of 12 antibiotics tested in minibioreactors. Initial bacterial community diversity was dependent on both the fecal donor and model system. Antibiotics affected a wide range of taxa across the phylogenetic spectrum, with many taxa similarly affected across treatments with different classes of antibiotics. Vancomycin, typically administered to treat Gram-positive bacterial infections, decreased the abundance of diverse taxa, including Gram-negative Bacteroidota species. Effects on some taxa were restricted by model system, indicating the importance of environmental context on antibiotic susceptibility. Altogether, these results indicate the complex interrelationships between microbiota composition and environmental context on antibiotic susceptibility and demonstrate strengths and weaknesses of each pre-clinical model system for evaluating effects of new antibiotics and other compounds with potential to disrupt the microbiota.