Search

Lack of reproducibility is a prominent problem in biomedical research. An important source of variation in animal experiments is the microbiome, but little is known about specific changes in the microbiota composition that cause phenotypic differences. Here, we show that genetically similar laboratory mice obtained from four different commercial vendors exhibited marked phenotypic variation in their susceptibility to Salmonella infection. Faecal microbiota transplant into germ-free mice replicated donor susceptibility, revealing that variability was due to changes in the gut microbiota composition. Co-housing of mice only partially transferred protection against Salmonella infection, suggesting that minority species within the gut microbiota might confer this trait. Consistent with this idea, we identified endogenous Enterobacteriaceae, a low-abundance taxon, as a keystone species responsible for variation in the susceptibility to Salmonella infection. Protection conferred by endogenous Enterobacteriaceae could be modelled by inoculating mice with probiotic Escherichia coli, which conferred resistance by using its aerobic metabolism to compete with Salmonella for resources. We conclude that a mechanistic understanding of phenotypic variation can accelerate development of strategies for enhancing the reproducibility of animal experiments. Variable susceptibility to Salmonella infection across genetically similar mice from commercial vendors is due to differential colonization of the gut microbiome by endogenous Enterobacteriaceae.

Sorbara et al. report a critical function of the healthy intestinal microbiota in preventing the expansion of antibiotic-resistant Enterobacteriaceae that depends on the production of high levels of short chain fatty acids and an acidic environment to trigger intracellular acidification of Enterobacteriaceae., Klebsiella pneumoniae, Escherichia coli, and other members of the Enterobacteriaceae family are common human pathogens that have acquired broad antibiotic resistance, rendering infection by some strains virtually untreatable. Enterobacteriaceae are intestinal residents, but generally represent, Graphical Abstract

Our skin and mucosal surfaces are colonized by diverse microbial communities, collectively known as the microbiota [1]. The microbiota provides benefits as microbial metabolites contribute to host nutrition and immune education, although the viability of germ-free animals conjectures that these two functions are not essential for life. However, environmental exposure makes germ-free animals prone to lethal infection, illustrating that the microbiota confers a third function that is often vital, namely, the ability to confer colonization resistance against pathogens [2]. Colonization resistance is an acquired trait, because the microbiota is assembled after birth by attaining maternal and environmental microbes [3]. To coexist, each species within the microbial community needs to be able to utilize a critical resource better than any other member of the microbiota, and the abundance of this growth-limiting resource determines the abundance of the species, a concept known as the nutrient-niche hypothesis [4]. The conceptual framework of the nutrient-niche hypothesis suggests that the neonate microbiota will mature until all discrete nutrient-niches have been filled with a suitable occupant, thereby reaching an equilibrium state [5]. Assuming the same anatomical location in different individuals exposes similar nutrient-niches, the nutrient-niche hypothesis further predicts that the metabolic pathways that enable each member within the microbial community to utilize its growth-limiting nutrient must be conserved between different individuals. Consistent with this prediction, metabolic pathways encoded by the microbiota are very similar between individuals [1]. However, carriage of microbial taxa varies greatly within a healthy population [1], an observation that is not explained by the nutrient-niche hypothesis and remains poorly understood. Priority effects generate variation in taxa carriage Host genetic variation explains only a small fraction of taxonomic microbiota variation between individuals, whereas environmental influences dominate this trait [6]. An important environmental influence in the gastrointestinal tract is the diet, which determines the availability of a subset of growth-limiting nutrients, thereby adding or subtracting nutrient-niches [7, 8]. For example, microbiota-accessible carbohydrates found in dietary fiber determine the abundance of fiber-consuming saccharolytic bacteria in the gut microbiota, and prolonged dietary fiber starvation can lead to an irreversible extinction of species specialized in devouring this critical resource by eliminating their nutrient-niche [8]. Although diet can generate statistically significant changes in the taxonomic composition of the gut microbiota, these changes are small compared to the variation observed between individuals. Furthermore, diet does not provide a plausible explanation for the taxonomic diversity observed in microbial communities outside the gastrointestinal tract [1]. Instead, a critical factor generating taxonomic microbiota diversity between individuals is the order of species arrival and timing by which host surfaces are colonized early in life [9]. The colonization order influences both the outcome of microbial community assembly and the ecological success of individual microbes [3, 9]. These priority effects are preserved in mice lacking adaptive immunity, suggesting that acquired host responses are not a major source of taxonomic diversity in the microbiota composition [9]. Priority effects are mediated through niche preemption or niche modification and can involve the genetic adaptation of microbes to a niche [9, 12], but the underlying mechanisms are incompletely understood. Mechanistic insights into this “first come, first serve” phenomenon suggest that the microbe that initially occupies a nutrient-niche in a neonate gains priority access to the growth-limiting nutrient that defines its nutrient-niche [10]. A growth-limiting resource that determines the abundance of facultative anaerobic Enterobacteriaceae (phylum Proteobacteria) within the microbiota of the large intestine is the availability of respiratory electron acceptors, such as oxygen [11]. Escherichia coli (family Enterobacteriaceae) has access to oxygen in the ceca of neonate chicks when it is inoculated one day prior to challenge with Salmonella enterica (family Enterobacteriaceae) but not when neonate chicks receive both species at the same time [10], suggesting that order and timing of gut colonization determine whether growth-limiting resources are accessible to a microbe. Henceforth we will refer to the concept that the founding occupant gains priority access to the growth-limiting resource that defines its nutrient-niche as the “founder hypothesis.” The founder hypothesis suggests that stochastic effects that govern the initial exposure of neonates to microbes that become founding occupants of each nutrient-niche are a prominent source of taxonomic variation in the microbiota composition between individuals (Fig 1) [3]. Open in a separate window Fig 1 The founder hypothesis. The principles of the founder hypothesis are shown schematically for a single nutrient-niche. Stochastic effects governing microbial exposure during infancy determine which microbial species (red or blue rods) establishes residency in the nutrient-niche, thereby generating diversity in taxa carriage between individuals. The founding occupant gains priority access to the growth-limiting resource that defines its nutrient-niche. These priority effects enable the occupant to confer colonization resistance against environmental exposure to microorganisms that are suitable contenders for the same nutrient-niche. The resulting resistance to stress imposed through environmental exposure to microorganisms produces microbiota resistance.