Time-series metagenomic analysis reveals robustness of soil microbiome against chemical disturbance

Soil microbial communities have great potential for bioremediation of recalcitrant aromatic compounds. However, it is unclear which taxa and genes in the communities, and how they contribute to the bioremediation in the polluted soils. To get clues about this fundamental question here, time-course (up to 24 weeks) metagenomic analysis of microbial community in a closed soil microcosm artificially polluted with four aromatic compounds, including phenanthrene, was conducted to investigate the changes in the community structures and gene pools. The pollution led to drastic changes in the community structures and the gene sets for pollutant degradation. Complete degradation of phenanthrene was strongly suggested to occur by the syntrophic metabolism by Mycobacterium and the most proliferating genus, Burkholderia. The community structure at Week 24 (∼12 weeks after disappearance of the pollutants) returned to the structure similar to that before pollution. Our time-course metagenomic analysis of phage genes strongly suggested the involvement of the ‘kill-the-winner’ phenomenon (i.e. phage predation of Burkholderia cells) for the returning of the microbial community structure. The pollution resulted in a decrease in taxonomic diversity and a drastic increase in diversity of gene pools in the communities, showing the functional redundancy and robustness of the communities against chemical disturbance.


Supplementary Figure S2. Colony-forming units (CFUs) of control (open circle) and polluted (closed circle) soil-residing heterotrophic prokaryotic cells on R2A agar plates.
To count the number of colony-forming heterotrophic prokaryotic cells in the soil, One-gram soil at an appropriate sampling time point was suspended in 9 ml of phosphate-buffered saline, and the suspension was diluted and spread onto Difco TM R2A agar plates (Becton Dickinson, Franklin Lakes, NJ, USA) and incubated at 30 o C for ten days. The data represent the mean values of triplicated measurements with standard deviations.

(c) Mycobacterium
Supplementary Figure S5. Distribution of contig lengths in eleven metagenomic samples.
Metagenomic reads from each sample were separately assembled using the IDBA-UD program. 18 The circle in the box plot represents base length of each contig. The number of reads used for the assembly, the length of contig N50, and the longest contig are indicated under the sample name.    Figure S8. Time-course changes in taxonomic compositions and abundances of genes for aerobic degradation of aromatic compounds in control (a) and polluted (b) soils. Depicted is an overall pathway map for well-known aerobic and anaerobic degradation routes of the four polluted compounds (see Fig. 4 for the simplified pathway map). Boxes located in each route indicate representative reaction steps, and several downstream steps (dashed lines) are omitted for simplicity. Abbreviations of enzymes are shown in Supplementary Table S5. The added contaminants are indicated by red nodes. The bar graph in the map indicates the abundance of the genes governing each reaction step at the phylum level (except in the cases of Archaea and Proteobacteria, for which the abundances at the domain and class levels are shown, respectively). See Materials and Methods for details of the calculation of the gene abundances. Depending on the abundances, the scales of Y-axes of graphs are conventionally categorized into four groups with the following colors: grey, 10 -6 ; blue, 10 -5 ; purple, 10 -4 ; and red, 10 -3 .         Figure S10. Time-course changes in relative abundances of phage genome-derived sequences in soil metagenomic samples. Depicted are the relative abundances of contigs putatively derived from phage genomes in the control (open circle) and the polluted (closed circle) soil metagenomic samples (see Fig. 5b and Supplementary Fig. S11 for Contig70-3). The abundance of the phage genome-derived contig in each sample was based on the numbers of hit reads counted by the BLASTN search of metagenomic reads against the contigs. The hit numbers were normalized by the number of gyrB-added USCGs a in each metagenomic sample. The relative abundance values are expressed by taking the smallest value as 1.  Table S16 for details of the PHAST analysis. (b)

BpB
The nucleotide positions of six PCR primer sets, sets 1 to 6 (see their primer sequences in Supplementary Table S2), are depicted. The primer sets of 5 and 6 were used for PCR detection of (a) circular and/or concatenated forms of the contig and the quantitative PCR analysis, respectively. (c) Agarose gel electrophoresis of PCR-amplified fragments using the polluted soil metagenomic sample that was prepared at week 6. No PCR products were detected when the control soil metagenomic samples were used. (d) Copy number of the phage genome-derived sequence in the control (open circle) and polluted (closed circle) soil metagenomic samples. Quantitative PCR analysis was performed with three replicates.
λ " H i n d ' I I I P r i m e r ' s e t ' 1 ' P r i m e r ' s e t ' 3 ' P r i m e r ' s e t ' 2 ' P r i m e r ' s e t ' 4 ' P r i m e r ' s e t 5 '