Non-antibiotic drugs break colonization resistance against pathogenicGammaproteobacteria

Abstract

SummaryNon-antibiotic drugs can alter the composition of the gut microbiome, with largely undefined implications for human health. Here we compared the susceptibility of commensal and pathogenic bacteria to non-antibiotic drugs and found that pathogens show higher drug resistance, which could favor their expansion after treatment. We then developed a model system to screen for drug-microbiome interactions that increase the risk of enteropathogenic infections. Approximately 35% of the >50 drugs we tested increased the abundance ofSalmonellaTyphimurium in synthetic and human stool-derived microbial communities. This was due to direct effects of non-antibiotics on individual commensals, altered microbial interactions within communities and the potential ofSalmonellato exploit different metabolic niches. Non-antibiotic drugs that favoredSalmonellaexpansionin vitroalso promoted other enteric pathogens and increasedSalmonellaloads in gnotobiotic and conventional mice. These findings may inform future strategies to control pathogen proliferation and to assess individual microbiota-drug-pathogen risks for infection.

Proton-pump inhibitors increaseC. difficileinfection risk by altering pH rather than by affecting the gut microbiome based on a bioreactor model

“…Table 1). These 21 bacterial species represent 7 bacterial phyla, 11 families, and 18 genera, covering 68.6% of the pathways detected in the human microbiome 29 . To assess omeprazole sensitivity, we revisited previous data from our lab 29 , where bacterial growth in mGAM was measured over 20 hours in the presence of varying omeprazole concentrations.…”

“…These 21 bacterial species represent 7 bacterial phyla, 11 families, and 18 genera, covering 68.6% of the pathways detected in the human microbiome 29 . To assess omeprazole sensitivity, we revisited previous data from our lab 29 , where bacterial growth in mGAM was measured over 20 hours in the presence of varying omeprazole concentrations. From the same dataset, we also examined the sensitivity of our strains to clindamycin, an antibiotic that is associated with a high risk for CDI [2][3][4][5] .…”

“…Subsequently, we exposed omeprazole-treated communities to C. difficile and quantified its growth in these communities. We first revisited previous data on the omeprazole sensitivity of human fecal samples 29 . Consistent with single-species sensitivities, the growth of all stool-derived communities was unaffected by omeprazole at all tested concentrations (Figure 2A).…”

“…Our multiple-bioreactor system (MBS) can be operated under an aerobic or anaerobic atmosphere and can be used for batch cultivation or in chemostat mode, where fresh medium is continuously supplied and spent medium is removed at the same rate. To demonstrate the functionality and reproducibility of our MBS with a gut microbial community, we conducted a pilot experiment in three bioreactors using a reduced version of Com21 29 , referred to here as Com18. Com18 lacks E. coli, Veillonella parvula, and Eggerthella lenta due to concerns about E. coli dominance and initial difficulties in growing Veillonella and Eggerthella species.…”

“…Thus, we employed various in vitro systems, ranging from batch cultivation to bioreactor systems, to precisely quantify the consequences of pH changes and physiological omeprazole concentrations on defined and human-stool-derived gut microbial communities. Subsequently, we challenged pH-and drug-perturbed communities with C. difficile and monitored its growth within these communities 29 . Our findings provide strong evidence that the increase in C. difficile growth associated with omeprazole is primarily a result of pH changes rather than direct interference of the drug with gut microbes.…”