Disentangling microbiota-dependent and -independent impacts of antibiotics on host physiology

Antibiotics are believed to affect host physiology by altering the gut microbiota. However, it remains unclear which effects of antibiotics are microbiota dependent and which are microbiota independent. Here we present a resource encompassing host and microbiota responses to antibiotic treatment and a computational tool with which to analyse them. Metagenomic and transcriptomic analyses of conventional and germ-free mice revealed that antibiotic administration routes and their combinations affect host and microbiota responses. We developed CompoRes—a computational tool that distinguishes the microbiota-independent impact of antibiotic-induced perturbations, revealing that antibiotics directly impact host physiology. We found that diurnal expression of major histocompatibility complex class II in the mouse colon is controlled by glucose via feeding. This pattern was lost during antibiotic treatment as vancomycin altered feeding behaviour in a microbiota-independent and evolutionarily conserved manner. Together, our study establishes a direct, non-microbial pathway by which antibiotics reshape host immunity and circadian physiology. Using a computational tool called CompoRes, the authors show that antibiotics can alter feeding behaviour, immunity and circadian regulation, revealing a non-microbial pathway by which antibiotic treatment reshapes host physiology.