Date of Award

Winter 12-15-2018

Author's School

Graduate School of Arts and Sciences

Author's Department

Biology & Biomedical Sciences (Molecular Genetics & Genomics)

Degree Name

Doctor of Philosophy (PhD)

Degree Type



Our lab previously reported that a subpopulation of human gut microbes (Bacteroides) can prevent development of obesity and its associated metabolic abnormalities in gnotobiotic mice colonized with the gut microbiota of obese (Ob) humans (Ridaura et al., Science 2013). This effect is diet-dependent and occurs when mice are consuming a representative (‘healthy’) USA diet low in saturated fats and high in fruits and vegetables (LoSF-HiFV) but not a diet high in saturated fats and low in fruits and vegetables (HiSF-LoFV). These human gut Bacteroides contain large repertoires of carbohydrate active enzymes (CAZymes) required to break down dietary plant fibers which are not specified by our H. sapiens genome.

Studies have indicated that the diversity of microbes in the gut communities of humans living in Westernized societies is significantly reduced compared to the microbiota of humans whose dietary practices include greater consumption of plant-derived components, including fibers. My thesis focused on using gnotobiotic mice to identify plant fibers that increase the representation of consumer microbial communities, notably these Bacteroides, and characterizing the effects of these fibers on microbial community and host metabolism.

A model human gut microbial community composed of cultured and sequenced bacterial strains recovered from a single human donor and including the targeted Bacteroides was introduced into adult germ-free C57BL/6J mice. These gnotobiotic mice were used to screen 34 commercially available fiber ingredients incorporated into 144 diets formulated from base HiSF-Lo/FV diet. Experiments were designed to identify ingredients capable of enhancing the proportional representation of one or more of the four Bacteroides. We found that the four targeted Bacteroides had distinct and in some cases overlapping responses to the 34 ingredients. Two fibers were selected as leads based on the magnitude of the increases in targeted Bacteroides that they produced, their affordability, their sustainable sourcing (each came from the waste streams of food manufacturing) and their ability to be incorporated into food prototypes with satisfactory organoleptic properties.

Using mice colonized with the defined culture collection monotonously fed HiSF-LoFV with or without one or the other lead fibers, shotgun sequencing of their microbial community DNA (to identify community composition), and targeted GC-MS, LC-QQQ-S mass spectrometry plus other biochemical methods, I identified fiber-specific and colonization-dependent increases in (i) microbial production of succinate, a key bacterial product of fiber fermentation, (ii) intestinal gluconeogenesis and (iii) liver glycogen levels (a feature associated with improved insulin sensitivity). Similar microbial community and host effects were observed in mice fed the same fiber-supplemented diets but colonized with an intact uncultured microbiota from an obese human donor.

Using untargeted LC-QTOF-MS, I identified a serotonin metabolite in the cecal contents of mice fed one of the two fibers. Analyses of input diet components and gut contents recovered from germ-free and colonized mice established that this metabolite was generated in a microbiota-dependent fashion. In vitro studies of the targeted Bacteroides cultured in absence or presence of the fiber preparation established that B. ovatus was responsible for generation of this serotonin metabolite. Further characterization of the effects of this metabolite is on-going. One function uncovered to date is an apparent effect on gluconeogenesis manifest by an increase in glucose levels in intestinal epithelial organoid cultures incubated with this metabolite. These results link a metabolite generated by a member of the gut microbiota from a plant fiber source to host metabolic state. Further work along these lines – to narrow down precisely which microbes in the human microbiota participate in biotransformations of fiber components to produce novel products that alter host metabolism – will be critical in the design of next-generation, microbiota-directed foods with health benefits.


English (en)

Chair and Committee

Jeffrey I. Gordon

Committee Members

Daniel E. Goldberg, Scott J. Hultgren, Samuel Klein, Clay F. Semenkovich,


Permanent URL:

Available for download on Thursday, December 26, 2120

Included in

Biology Commons