Frontiers in Nutrition (Mar 2022)

Classification of the Occurrence of Dyslipidemia Based on Gut Bacteria Related to Barley Intake

  • Satoko Maruyama,
  • Satoko Maruyama,
  • Tsubasa Matsuoka,
  • Tsubasa Matsuoka,
  • Tsubasa Matsuoka,
  • Koji Hosomi,
  • Jonguk Park,
  • Mao Nishimura,
  • Mao Nishimura,
  • Haruka Murakami,
  • Kana Konishi,
  • Motohiko Miyachi,
  • Hitoshi Kawashima,
  • Kenji Mizuguchi,
  • Kenji Mizuguchi,
  • Toshiki Kobayashi,
  • Tadao Ooka,
  • Zentaro Yamagata,
  • Jun Kunisawa,
  • Jun Kunisawa,
  • Jun Kunisawa,
  • Jun Kunisawa,
  • Jun Kunisawa

DOI
https://doi.org/10.3389/fnut.2022.812469
Journal volume & issue
Vol. 9

Abstract

Read online

Barley is a grain rich in β-glucan, a soluble dietary fiber, and its consumption can help maintain good health and reduce the risk of metabolic disorders, such as dyslipidemia. However, the effect of barley intake on the risk of dyslipidemia has been found to vary among individuals. Differences in gut bacteria among individuals may be a determining factor since dietary fiber is metabolized by gut bacteria and then converted into short-chain fatty acids with physiological functions that reduce the risk of dyslipidemia. This study examined whether gut bacteria explained individual differences in the effects of barley intake on dyslipidemia using data from a cross-sectional study. In this study, participants with high barley intake and no dyslipidemia were labeled as “responders” to the reduced risk of dyslipidemia based on their barley intake and their gut bacteria. The results of the 16S rRNA gene sequencing showed that the fecal samples of responders (n = 22) were richer in Bifidobacterium, Faecalibacterium, Ruminococcus 1, Subdoligranulum, Ruminococcaceae UCG-013, and Lachnospira than those of non-responders (n = 43), who had high barley intake but symptoms of dyslipidemia. These results indicate the presence of certain gut bacteria that define barley responders. Therefore, we attempted to generate a gut bacteria-based responder classification model through machine learning using random forest. The area under the curve value of the classification model in estimating the effect of barley on the occurrence of dyslipidemia in the host was 0.792 and the Matthews correlation coefficient was 0.56. Our findings connect gut bacteria to individual differences in the effects of barley on lipid metabolism, which could assist in developing personalized dietary strategies.

Keywords