Research Article

Genetic determinants of in vivo fitness and diet responsiveness in multiple human gut Bacteroides

See allHide authors and affiliations

Science  02 Oct 2015:
Vol. 350, Issue 6256, aac5992
DOI: 10.1126/science.aac5992

You are currently viewing the abstract.

View Full Text

Log in to view the full text

Log in through your institution

Log in through your institution

Diet shapes host and gut microbe fitness

The human gut microbiota is hugely diverse, with many strain variants having a multiplicity of effects on host metabolism and immunity. To define some of these functions, Wu et al. made libraries of mutants of Bacteroides species known for their capacity to process otherwise intractable dietary fiber. Germ-free mice colonized with defined gut microbiota communities containing the mutants were fed specific diets containing different ratios of fat and fiber. Genes, strains, and species were identified that were associated with specific metabolic pathways. The community responses to dietary shifts were manipulated in an attempt to characterize species for their probiotic or therapeutic potential.

Science, this issue 10.1126/science.aac5992>

Structured Abstract


Relatively little is known about the genetic factors that allow members of the human gut microbiota to occupy their niches. Identification of these factors is important for understanding mechanisms that determine microbiota assembly and perturbation through diet, disease, and clinical treatments. Discovery of these factors should enable new approaches for intervening therapeutically in the functional properties of the human gut microbiota. We present a generalizable approach by which to identify fitness determinants for multiple bacterial strains simultaneously in a model human gut microbiota, obtain gene-level characterization of responses to diet change, and design prebiotics for precision microbiota manipulation.


We developed a method—multi-taxon INsertion Sequencing (INSeq)—for monitoring the behavior of tens of thousands of transposon (Tn) mutants of multiple bacterial species and strains simultaneously in the guts of gnotobiotic mice. We focused on four prominent human gut Bacteroides: one strain of B. cellulosilyticus, one strain of B. ovatus, and two strains of B. thetaiotaomicron. INSeq libraries, each composed of 87,000 to 167,000 isogenic Tn mutant strains, were produced (single site of Tn insertion per mutant strain; a total of 11 to 26 Tn insertions represented in the library per gene; and 82 to 92% genes covered per genome). The four mutant libraries were introduced into germ-free mice together with 11 wild-type species commonly present in the human gut microbiota. Animals were given a diet rich in fat and simple sugars but devoid of complex polysaccharides [diet 1 (D1)] or one rich in plant polysaccharides and low in fat (D2), either monotonously or in the sequence D1-D2-D1 or D2-D1-D2. Wecalculated a “fitness index” for each gene on the basis of the relative abundance of its INSeq reads in the fecal or cecal microbiota compared with the input library. In vivo INSeq data were correlated with INSeq data generated from organisms cultured under defined in vitro conditions; microbial RNA-seq profiling of the community’s metatranscriptome; and reconstructions of metabolic pathways, regulons, and polysaccharide utilization loci. On the basis of the results, we designed a prebiotic intervention.


Multi-taxon INSeq (i) provided a digital readout of the remarkably consistent pattern of community assembly; (ii) identified shared as well as species-, strain-, and diet-specific fitness determinants associated with a variety of metabolic or nutrient processing pathways, including those involving amino acids, carbohydrates, and vitamins/cofactors; (iii) enabled quantitative gene-level measurement of the resilience of the responses to diet perturbations; (iv) revealed that arabinoxylan, the most common hemicellulose in cereals, could be used to deliberately manipulate the representation of Bacteroides cellulosilyticus; and (v) defined the niche adjustments of this and the other Bacteroides to arabinoxylan supplementation of the high-fat diet.


In principle, the approach described can be used to obtain a more comprehensive understanding of how host genotype, diet, physiologic, metabolic, and immune factors, as well as pathologic states, affect niches in gut and nongut habitats, as well as to facilitate development of therapeutic interventions for modifying community structure/function.

Identification of a prebiotic that increases the abundance of B. cellulosilyticus.

(Left) The four mutant libraries were pooled together with 11 other phylogenetically diverse wild-type strains, and this consortium, representing an artificial human gut microbiota, was introduced into germ-free mice. Community assembly, the effects of diet, and recovery from diet oscillations were characterized at a community, strain, and gene level in these gnotobiotic animals by use of multi-taxon INSeq. (Middle) Multi-taxon INSeq revealed an arabinoxylan utilization locus in B. cellulosilyticus that is critical for the organism’s fitness in the high-fat/simple-sugar diet (D1) context but not in the D2 context. A homologous arabinoxylan utilization locus in B. ovatus was not a fitness determinant with either diet. (Right) Consistent with this finding, supplementation of drinking water with arabinoxylan in mice consuming D1 selectively increased the abundance of B. cellulosilyticus but not B. ovatus.


Libraries of tens of thousands of transposon mutants generated from each of four human gut Bacteroides strains, two representing the same species, were introduced simultaneously into gnotobiotic mice together with 11 other wild-type strains to generate a 15-member artificial human gut microbiota. Mice received one of two distinct diets monotonously, or both in different ordered sequences. Quantifying the abundance of mutants in different diet contexts allowed gene-level characterization of fitness determinants, niche, stability, and resilience and yielded a prebiotic (arabinoxylan) that allowed targeted manipulation of the community. The approach described is generalizable and should be useful for defining mechanisms critical for sustaining and/or approaches for deliberately reconfiguring the highly adaptive and durable relationship between the human gut microbiota and host in ways that promote wellness.

View Full Text