Time Series Genome-Centric Analysis Unveils Bacterial Response to Operational Disturbance in Activated Sludge.

Abstract

Understanding ecosystem response to disturbances and identifying the most critical traits for the maintenance of ecosystem functioning are important goals for microbial community ecology. In this study, we used 16S rRNA amplicon sequencing and metagenomics to investigate the assembly of bacterial populations in a full-scale municipal activated sludge wastewater treatment plant over a period of 3 years, including a 9-month period of disturbance characterized by short-term plant shutdowns. Following the reconstruction of 173 metagenome-assembled genomes, we assessed the functional potential, the number of rRNA gene operons, and the in situ growth rate of microorganisms present throughout the time series. Operational disturbances caused a significant decrease in bacteria with a single copy of the rRNA (rrn) operon. Despite moderate differences in resource availability, replication rates were distributed uniformly throughout time, with no differences between disturbed and stable periods. We suggest that the length of the growth lag phase, rather than the growth rate, is the primary driver of selection under disturbed conditions. Thus, the system could maintain its function in the face of disturbance by recruiting bacteria with the capacity to rapidly resume growth under unsteady operating conditions.IMPORTANCE Disturbance is a key determinant of community assembly and dynamics in natural and engineered ecosystems. Microbiome response to disturbance is thought to be influenced by bacterial growth traits and life history strategies. In this time series observational study, the response to disturbance of microbial communities in a full-scale activated sludge wastewater treatment plant was assessed by computing specific cellular traits of genomes retrieved from metagenomes. It was found that the genomes observed in disturbed periods have more copies of the rRNA operon than genomes observed in stable periods, whereas the in situ mean relative growth rates of bacteria present during stable and disturbed periods were indistinguishable. From these intriguing observations, we infer that the length of the lag phase might be a growth trait that affects the microbial response to disturbance. Further exploration of this hypothesis could contribute to better understanding of the adaptive response of microbiomes to unsteady environmental conditions.