“We set out to study postprandial glycemic response variation based on individual differences, focusing on differences in gut microbiome activity, and we were the first to accomplish this using the metatranscriptome,” wrote the researchers from Viome Research Institute, in the journal Diabetes Therapy.
The scientists also identified specific microbial features that influence glycemic response - knowledge that may help elucidate molecular mechanisms of glycemic control.
Controlling postprandial glycemic response is a crucial strategy in mitigating metabolic diseases such as obesity, type 2 diabetes, hypertension, CVD and liver disease.
There is growing evidence to suggest that glycemic response to the same foods varies significantly from one person to the next. Recent studies have shown that postprandial glycemic response is driven not only by the glycemic index of the food but also by individual phenotypes and molecular characteristics, including the gut microbiome.
Metagenomic vs metatranscriptomic
The conclusions in these studies depended on 16S rRNA gene sequencing or metagenomic data from the gut microbiome. According to the authors of the current study, the problem with 16S rRNA gene sequencing is that it provides poor taxonomic resolution of microbiomes. In addition, they said metagenomic methods are unable to identify some microorganisms and can only predict gene expression based on gene content, which can be highly erroneous.
Metatranscriptomic methods, by contrast, offer a comprehensive lens of the microbiome, with a specific focus on genes that are actively transcribed. However, metatranscriptomics has not been widely used in clinical studies due to various challenges and complexities.
For this study, the researchers employed a metatranscriptomic approach to collect gut microbiome activity data, measuring the complete set of gene transcripts (RNA) from a sample.
“In microbiome samples, this technique facilitates quantitative strain level classification and functional pathway analysis of microbes, which we accomplish through sequence alignments to publicly available reference databases,” explained the researchers.
Study details
550 adults (66% female) recruited from the general public across the USA took part in the study. The research team tracked their food intake, sleep, activity and glycemic response for two weeks. Blood glucose measurements were taken every 15 minutes using a continuous glucose sensor. Stool samples were collected at start of the study and processed using a metatranscriptomic method, which enabled microbiome balance scores of low or normal to be assigned.
For breakfast, morning snack and lunch, participants ate pre-designed meals (formulated to ensure a diverse macronutrient intake). Provided meals accounted for 66% of all meals. According to the researchers, this proportion was significantly higher than in previous studies, and allowed them to quantify how different people’s glycemic responses to the same foods varied.
The researchers used a mixed-effects linear regression model to link glycemic response to nutrient characteristics of meals, anthropometric factors such as BMI and age and gut microbiome activity.
“The linear model allows us to provide a concise description of the relationships between nutrients, anthropometrics, microbiome activity and postprandial response. Additionally, it allows us to derive significance statistics, testing whether each predictor is relevant to the determination of the postprandial glycemic response,” they explained.
The researchers also processed the same data using a gradient-boosting machine model for greater predictive accuracy.
Microbiome balance
Their modelling found that several of the significant predictors of glycemic response were microbiome scores. One of these was ‘microbiome balance’. When this was ‘low’, it showed a negative association with glycemic response.
“Microbiome balance scores that were low usually resulted from either an imbalance of relative activities of beneficial vs harmful microbes or from lower quantity and diversity of microbial organisms,” wrote the researchers.
They said the relationship between a suboptimal gut microbiome and higher glycemic responses is in line with current literature, implicating the role of gut health in glycemic regulation.
Other microbiome features that were found to be predictors of glycemic response were: fucose metabolism pathways, indoleacetate production pathways, glutamine production pathways, tyrosine metabolisers and fructose metabolisers.
The researchers said these microbiome features may influence postprandial glycemic response directly or indirectly. They said that although it is challenging to establish causal mechanisms, there may be functional patterns that connect the significant scores with gut health, intestinal barrier integrity and inflammation.
They said this data will be used in future to develop therapeutics that target specific microbial pathways to lower glycemic response. These could include small molecule inhibitors, small molecule supplements, phages, vaccines and probiotics.
“We will seek to confirm and validate these mechanisms. An understanding of which microbiome features are significant will pave the path to precise personalisation of food and supplement recommendations,” they said.
Journal: Diabetes Therapy
Authors: Tily H, Patridge E, Cai Y, Gopu V, Gline S, Genkin M, Lindau H, Sjue A, Slavov J, Perlina A, Klitgord N, Messier H, Vuyisich M, Banavar G
“Gut microbiome activity contributes to prediction of individual variation in glycemic response
in adults"
https://doi.org/10.1007/s13300-021-01174-z