Dead or Alive? Molecular life-dead distinction in human stool samples reveals significantly different composition of the microbial community

Abstract

The gut microbiome is strongly interwoven with human health. Conventional gut microbiome analysis generally involves 16S rRNA gene targeting next generation sequencing (NGS) of stool microbial communities, and correlation of results with clinical parameters. However, some microorganisms may not be alive at the time of sampling, and thus their impact on the human health is potentially less significant. As conventional NGS methods do not differentiate between viable and dead microbial components, retrieved results provide only limited information.Propidium monoazide (PMA) is frequently used in food safety monitoring and other disciplines to discriminate living from dead cells. PMA binds to free DNA and masks it for subsequent procedures. In this article we show the impact of PMA on the results of 16S rRNA gene-targeting NGS from human stool samples and validate the optimal applicable concentration to achieve a reliable detection of the living microbial communities.Fresh stool samples were treated with a concentration series of zero to 300 µM PMA, and were subsequently subjected to amplicon-based NGS. The results indicate that a substantial proportion of the human microbial community is not intact at the time of sampling. PMA treatment significantly reduced the diversity and richness of the sample depending on the concentration and impacted the relative abundance of certain important microorganisms (e.g. Akkermansia, Bacteroides). Overall, we found that a concentration of 100 µM PMA was sufficient to quench signals from disrupted microbial cells.The optimized protocol proposed here can be easily implemented in classical microbiome analyses, and helps to retrieve an improved and less blurry picture of the microbial community composition by excluding signals from background DNA.