Metabologenomic data may predict probiotic effects before intake, study suggests
A randomized, double-blind controlled crossover trial found that Bifidobacterium longum BB536 supplements led to different effects on bowel movement frequency in people who tended to be constipated. Machine learning analysis showed that strong responders could be predicted from features of the intestinal environment before the first probiotic was taken.
“These findings have interesting implications for personalized treatment of chronic constipation. In addition, the effect of the intestinal environment on the response to intervention should be considered not only for probiotics but also for all health effects, adverse effects, and side effects for all orally administered interventions,” wrote researchers from Japan.
“When the effect of probiotics, diets, or drugs is dependent on the intestinal environment before intake, as it is in the present study, the demand for intestinal environment alterations to enhance the effect may also increase.
“Furthermore, the development of personalized health care and medical businesses targeting the intestinal environment is expected to use this intervention effect quantification and responder prediction system.”
Study details
The Japan-based scientists recruited 20 adults who tended to be constipated to participate in their study. The participants were randomly assigned to receive either B. longum BB536 in acid-resistant seamless capsules (five billion CFUs) or similarly encapsulated placebo for two weeks. This was followed by a four-week “washout period” before the participants crossed over to the other group. All capsules were provided by Morishita Jintan Co.
The data showed that, for the 19 people who completed the study, three were classified as Strong Responders, nine as Weak Responders, and seven as Non-Responders.
Machine learning based on the microbiome and metabolome features of fecal samples collected before B. longum supplementation showed significant differences between responders and non-responders for eight bacterial genera. Specifically, there was a higher abundance of g__Ruminococcus_E in the responders, compared to the non-responders, while there was a lower abundance of g__Agathobacter, g__Alistipes, g__Bilophila, g__Butyricimonas, g__Dorea, g__Escherichia and g__Parabacteroides in responders versus non-responders.
“In the present study, we accurately quantified individual differences in the increase in bowel movement frequency in response to intake of B. longum BB536 contained in acid-resistant seamless capsules using a statistical model and defined the responders as those with a noticeably increased bowel movement frequency,” explained the researchers.
“In addition, the machine learning analysis revealed that responders could be predicted from features of the intestinal environment before the initiation of B. longum BB536 supplement intake, and the predictive performance was improved when using both microbiome and metabolome data.
“Based on these results, accurate quantification of the individual response intensity and machine learning predictions may enable companion diagnostics for the response to B. longum BB536 supplementation based on the intestinal environment.”
Source: Computational and Structural Biotechnology Journal
Volume 20, 2022, Pages 5847-5858
“Integrated gut microbiome and metabolome analyses identified fecal biomarkers for bowel movement regulation by Bifidobacterium longum BB536 supplementation: A RCT”
Authors: Y. Nakamura et al.
