A microbial gene catalog of anaerobic digestion from full-scale biogas plants

Abstract

Background: Biogas production with anaerobic digestion (AD) is one of the most promising solutions for both renewable energy production and resolving the environmental problem caused by the worldwide increase in organic waste. However, the complex structure of the microbiome in AD is poorly understood.

Findings: In this study, we constructed a microbial gene catalog of AD (22,840,185 genes) based on 1,817 Gb metagenomic data derived from digestate samples of 56 full-scale biogas plants fed with diverse feedstocks. Among the gene catalog, 73.63% and 2.32% of genes were taxonomically annotated to Bacteria and Archaea, respectively, and 57.07% of genes were functionally annotated with KEGG orthologous groups. Our results confirmed the existence of core microbiome in AD and showed that the type of feedstock (cattle, chicken, and pig manure) has a great influence on carbohydrate hydrolysis and methanogenesis. In addition, 2,426 metagenome-assembled genomes were recovered from all digestate samples, and all genomes were estimated to be ≥80% complete with ≤10% contamination.

Conclusions: This study deepens our understanding of the microbial composition and function in the AD process and also provides a huge number of reference genome and gene resources for analysis of anaerobic microbiota.

Keywords: anaerobic digestion; full-scale biogas plant; manure waste; metagenome; metagenome-assembled genomes; methanogenesis.

Figure 1:

The constructed microbial gene catalog of anaerobic digestion (MGCA). a, Rarefaction curve of detected genes from the whole set of 59 digestate samples. The curve approaches saturation as sample number increases. The gene number of a given number of samples was calculated after 100 random samplings with replacement and plotted with a box plot. Box plots show the median ± interquartile range (IQR) and 1.5 IQR ranges (whiskers), with outliers denoted by circles. b, Venn diagram of shared genes among 4 groups of non-redundant genes from MCA, MCH, MPI, and OTH. Only a small proportion of genes were unique for each group. MCA: cattle manure BGPs; MCH: chicken manure BGPs; MPI: pig manure BGPs; OTH: BGPs with other substrates.

Figure 2:

Taxonomic annotation of the microbial gene catalog of anaerobic digestion (MGCA). a, Taxonomic annotation of the gene catalog at the superkingdom and phylum levels. A total of 73.63% and 2.32% of genes in the gene catalog were assigned to Bacteria and Archaea, respectively. b, Percentage of genes assigned to the top 10 methanogenic archaea at genus level.

Figure 3:

KEGG functional profile of the microbial gene catalog of anaerobic digestion (MGCA). Genes without functional annotations were excluded.

Figure 4:

Distributions of feedstock-associated core genera among the 4 groups, MCA, MCH, MPI, and OTH. The area of each circle represents the median value of relative abundance of the corresponding genus in each group, and the non-core genera are not shown. “Core microbes” were defined as the genera most abundantly (top 30 bacterial genera and top 5 archaeal genera) detected in all studied samples. MCA: cattle manure BGPs; MCH: chicken manure BGPs; MPI: pig manure BGPs; OTH: BGPs with other substrates.

Figure 5:

Comparisons of taxonomic and functional profiles among different biogas plants (BGPs). a, Principal coordinate analysis (PCoA) based on Bray-Curtis dissimilarity at the species level. The digestate samples were separated into 3 clusters (MCA, MCH, and MPI). MCA: cattle manure BGPs; MCH: chicken manure BGPs; MPI: pig manure BGPs; OTH: BGPs with other substrates. b, Relative abundance of genes involved in the hydrolysis of starch, oligosaccharide, polysaccharide, and lignocellulose (lignin, hemicellulose, and cellulose). c, Relative abundance of genes involved in protein hydrolysis. d, Relative abundance of genes involved in acetate, propionate, and butyrate oxidation. e, Relative abundance of genes involved in methanogenesis. Box plots show the median ± interquartile range (IQR) and 1.5 IQR ranges (whiskers), with outliers denoted by circles. Wilcoxon rank-sum test among different groups was performed. *P < 0.05.