Rare Taxa Exhibit Disproportionate Cell-Level Metabolic Activity in Enriched Anaerobic Digestion Microbial Communities

Abstract

Variation in microbial activity levels is increasingly being recognized as both an important dimension in community function and a complicating factor in sequencing-based survey methods. This study extends previous reports that rare taxa may contribute disproportionately to community activity in some natural environments, showing that this may also hold in artificially maintained model communities with well-described inputs, outputs, and biochemical functions. These results demonstrate that assessment of activity levels using the rRNA/rDNA ratio is robust across taxonomic unit formation methods and is independently corroborated by omics methods. The results also provide insight into the comparative advantages and disadvantages of different taxonomic unit formation methods in amplicon sequencing studies, showing that UNOISE3 provides comparable microbial diversity, structure, and activity information as the 97% sequence similarity method but potentially loses some phylogenetic diversity and creates more “phantom taxa” (which are present in the RNA pool but not the corresponding DNA pool).

ABSTRACT

Microbial communities are composed of populations with vastly different abundances and levels of metabolic and replicative activity, ranging from actively metabolizing and dividing to dormant or nonviable. The 16S rRNA/rDNA ratio is an emerging tool for evaluating cell-level metabolic activity independent of abundance. In this study, we used five long-term enriched model anaerobic digestion (AD) communities to investigate community composition, diversity, structure, and in particular activity based on the rRNA/rDNA ratio. We cross-validated the 16S amplicon-based results using two alternative operational taxonomic unit (OTU) formation methods (conventional 97% sequence similarity and 100% sequence similar zero-radius OTUs by UNOISE3) and compared these to metagenome-derived population genomes and metatranscriptomes. Significant positive correlations were observed between microbial total activity and abundance with both the amplicon- and omic-based methods. All three methods revealed disproportionately high transcription/abundance ratios for some rare taxa but lower ratios for most abundant taxa for all the communities, which was further corroborated by the high replication rate (iRep) of most low-abundance population genomes.

IMPORTANCE Variation in microbial activity levels is increasingly being recognized as both an important dimension in community function and a complicating factor in sequencing-based survey methods. This study extends previous reports that rare taxa may contribute disproportionately to community activity in some natural environments, showing that this may also hold in artificially maintained model communities with well-described inputs, outputs, and biochemical functions. These results demonstrate that assessment of activity levels using the rRNA/rDNA ratio is robust across taxonomic unit formation methods and is independently corroborated by omics methods. The results also provide insight into the comparative advantages and disadvantages of different taxonomic unit formation methods in amplicon sequencing studies, showing that UNOISE3 provides comparable microbial diversity, structure, and activity information as the 97% sequence similarity method but potentially loses some phylogenetic diversity and creates more “phantom taxa” (which are present in the RNA pool but not the corresponding DNA pool).