Behavioral response to fiber feeding is cohort-dependent and associated with gut microbiota composition in mice

Lucy J Mailing, Jacob M Allen, Brandt D Pence, Jennifer Rytych, Yi Sun, Tushar K Bhattacharya, Pul Park, Tzu-Wen L Cross, Robert H McCusker, Kelly S Swanson, George C Fahey, Justin S Rhodes, Keith W Kelley, Rodney W Johnson, Jeffrey A Woods

Recent data has supported a role for the gut microbiota in improving cognition and shaping behavior. Here, we assessed whether pectin, a soluble, fermentable fiber, could enhance learning and memory in mice. Two cohorts of young male C57Bl/6 J mice, C1 (n = 20) and C2 (n = 20), were obtained from Jackson Laboratory and randomized to semi-purified AIN-93 M diets containing 5% pectin (n = 10) or cellulose (n = 10). After 16 weeks, learning and memory was assessed by Morris Water Maze (MWM) and microbiota composition was analyzed by 16S rRNA sequencing. Despite identical treatment, we observed differences in learning and memory abilities between cohorts, along with distinct microbiotas. In C1, pectin-fed mice spent a higher percentage of time in the target quadrant at the 24-hour probe trial of the MWM versus cellulose-fed mice; in C2, no effect of pectin was observed. In both cohorts, UniFrac distance revealed significant differences in gut microbial communities between cellulose-fed and pectin-fed mice. UniFrac analysis also revealed significantly different bacterial communities between cohorts. Further analysis demonstrated that the microbial genera Oscillospira, Bilophila, and Peptostreptococcoceae were more abundant in C1 versus C2, and positively associated with distance from the platform during the 24-hour probe test. These data support previous findings that differences in the gut microbiota may play a role in host response to a dietary intervention and could partly explain irreproducibility in psychological and behavioral experiments. Further research is needed to determine if a causal relationship exists.