Abstract
To investigate the effect of flour and starch of the Indonesian native tuber "taro" on the composition and activity of the gut microbiota in diabetic rats, streptozotocin (STZ)-induced diabetic rats were fed normal chow (AIN), or AIN in which corn starch was replaced by either taro flour or purified taro starch for 4 weeks. Fecal samples were collected at baseline and after 4 weeks, and the composition of microbial communities was measured using 16S rRNA sequencing, while SCFAs were measured using ion chromatography. Bodyweight declined upon DM induction with STZ. Feeding taro starch led to a lower reduction in bodyweight than feeding taro starch, but this was only significant for taro starch in weeks 2, 3, and 4 (p = 0.02, p = 0.01, and p < 0.01, respectively). Both taro starch and taro flour induced changes in the gut microbiota composition compared to AIN, which were different for taro flour and taro starch. Bifidobacterium, Sutterella, and Prevotella were markers for taro flour feeding, while Anaerostipes was a marker for taro starch feeding. Induction of diabetes also led to changes in the microbiota composition. Random Forest correctly predicted for 16 of 18 samples whether rats were diabetic or not and correctly predicted 6 of 12 microbiota samples belonging to either taro flour- or taro starch-fed groups, indicating also some significant overlap in the substrate, as expected. Taro starch and taro flour both led to a significant increase in the fecal concentrations of acetate, propionate, and butyrate.