Changes in Gut Microbiome Can Help Predict Diabetes Risk

July 8, 2020

New research suggests the 24-hour changes in gut microbiome profile could help predict risk of type 2 diabetes.

A new study examining the impact and associations of gut microbiome profiles on various disease states has uncovered a new link that could aid clinicians in the fight against type 2 diabetes.

Performed by investigators at the ZIEL Institute for Food and Health at Technical University of Munich in Germany, results of the analysis indicate fluctuations of the gut microbiome could help improve clinicians’ ability to predict which patients will develop type 2 diabetes.

"When certain gut bacteria do not follow a day-night rhythm, so if their number and function does not change over the course of the day, this can be an indicator for a potential type 2 diabetes disease. Knowing this can improve diagnosis and outlook of type 2 diabetes," said study investigator Silke Kiessling, PhD, a chronobiologist and research group leader at the ZIEL Institute for Food and Health, in a statement.

While research into gut microbiome profiles has ballooned in recent years, relatively little is understood in regard to its impact and how this newfound knowledge can be applied in real-world settings. In previous studies, researchers established a link between lifestyle, obesity, and gut microbiome as risk factors for metabolic disorders. With this in mind, Kiessling and colleagues from the ZIEL Institute for Food and Health designed a study using subjects from the KORA cohort.

Of note, KORA was designed as a prospective cohort study examining impact of genetic, lifestyle, and environmental factors on disease progression. As part of KORA, 1976 patients provided stool samples, which underwent sequencing with use of 16S rRNA gene amplicon sequencing.

Upon initial analysis, microbiota compositions were dominated by Firmicutes and Bacteroidetes—accounting for a cumulative mean relative abundance of 91%. Investigators also pointed out an average individual richness of 348±77 operational taxonomic units (OTUs) and 118±37 Shannon effective number of species. In total, investigators used results to identify 40 features related to physiology, lifestyle, environment, disease-associated parameters, and medication, which accounted for 9.1% of the observed variability in microbiome.

From the study cohort, investigators identified 1340 subjects sampled in 2013 and another 699 subjects samples from a 5-year follow-up. To assess whether they could successfully predict type 2 diabetes using arrhythmic microbial signatures, investigators identified 13 arrhythmic OTUs associated with disrupted rhythmicity in type 2 diabetes, which was used to create cross-validated prediction models.

Using these models, investigators were able to predict type 2 diabetes in 699 KORA subjects 5 years after initial sampling—noting the most effective predictions occurred when combining factors with BMI. Additionally, shotgun metagenomic analysis successfully linked 25 metabolic pathways to diurnal oscillation of gut bacteria.

“Apart from bacteria and their variations over the course of the day, other parameters such as the body mass index play a role in being able to better predict a person's future medical conditions,” said study investigator Dirk Haller, a professor for Nutrition and Immunology at the Technical University of Munich, in the aforementioned statement.

This study, “Arrhythmic Gut Microbiome Signatures Predict Risk of Type 2 Diabetes,” was published in Cell Host and Microbe.