Monthly dynamics of microbial communities and variation of nitrogen-cycling genes in an industrial-scale expanded granular sludge bed reactor

INTRODUCTION: The expanded granular sludge bed (EGSB) is a major form of anaerobic digestion system during wastewater treatment. Yet, the dynamics of microbial and viral communities and members functioning in nitrogen cycling along with monthly changing physicochemical properties have not been well elucidated. METHODS: Here, by collecting the anaerobic activated sludge samples from a continuously operating industrial-scale EGSB reactor, we conducted 16S rRNA gene amplicon sequencing and metagenome sequencing to reveal the microbial community structure and variation with the ever-changing physicochemical properties along within a year. RESULTS: We observed a clear monthly variation of microbial community structures, while COD, the ratio of volatile suspended solids (VSS) to total suspended solids (TSS) (VSS/TSS ratio), and temperature were predominant factors in shaping community dissimilarities examined by generalized boosted regression modeling (GBM) analysis. Meanwhile, a significant correlation was found between the changing physicochemical properties and microbial communities (p <0.05). The alpha diversity (Chao1 and Shannon) was significantly higher (p <0.05) in both winter (December, January, and February) and autumn (September, October, and November) with higher organic loading rate (OLR), higher VSS/TSS ratio, and lower temperature, resulting higher biogas production and nutrition removal efficiency. Further, 18 key genes covering nitrate reduction, denitrification, nitrification, and nitrogen fixation pathways were discovered, the total abundance of which was significantly associated with the changing environmental factors (p <0.05). Among these pathways, the dissimilatory nitrate reduction to ammonia (DNRA) and denitrification had the higher abundance contributed by the top highly abundant genes narGH, nrfABCDH, and hcp. The COD, OLR, and temperature were primary factors in affecting DNRA and denitrification by GBM evaluation. Moreover, by metagenome binning, we found the DNRA populations mainly belonged to Proteobacteria, Planctomycetota, and Nitrospirae, while the denitrifying bacteria with complete denitrification performance were all Proteobacteria. Besides, we detected 3,360 non-redundant viral sequences with great novelty, in which Siphoviridae, Podoviridae, and Myoviridae were dominant viral families. Interestingly, viral communities likewise depicted clear monthly variation and had significant associations with the recovered populations (p <0.05). DISCUSSION: Our work highlights the monthly variation of microbial and viral communities during the continuous operation of EGSB affected by the predominant changing COD, OLR, and temperature, while DNRA and denitrification pathways dominated in this anaerobic system. The results also provide a theoretical basis for the optimization of the engineered system.