4.3 Microbial community functional structure
For the 23 water samples collected from CBM wells, 5 samples located in the runoff areas and 18 samples in the stagnant areas (Fig.5a). We generated a bacterial community profile for each sample via PCR amplification of the 16S ribosomal RNA (rRNA) gene targeting regions V3-V4 using primers 341F and 805R, followed by Illumina sequencing. 1966435 high-quality sequences from 23 samples were obtained (average, 85497; range, 53843-147971 reads per sample). High quality reads were analyzed with USEARCH, removing chimeric and organelle sequences, to produce 8264 operational taxonomic units (OTUs).
In all the samples, at the class level, Gammaproteobacteria, Betaproteobacteria, Deltaproteobacteria and Alphaproteobacteria were the most abundant bacteria, made up more than 85% of all microorganisms. Clostridia, Bacteroidia, Flavobacteriia, Bacilli, Methanobacteria, Actinobacteria were also present at higher relative abundance, but much lower than the four most abundant classes (Fig.8a).
90 most highly abundant microorganisms related to methanogenesis, denitrification, sulfate reduction and methane oxidation were selected. Maximum likelihood phylogenetic was constructed based on 16s V3-V4 sequences, shown in Fig.8b. There were three orders of Methanogens in study area, including Methanobacteriales, Methanomicrobiales and Methanosarcinales. Methane oxidizing bacteria included Methylococcales and Methylophilales. Denitrifying bacteria included Burkholderiales and Rhodobacterales. Sulfate reduction bacteria included Desulfobacterales and Desulfovibrionales. Aerobic ammonia oxidation bacteria (AOB) included Nitrosomonadales. Nitrification bacteria included Nitrospirales and Nitrosomonadales. The order of anammox bacteria was belonged to Candidatus Brocadiales.
Consistent with this expectation, the microbial community functional gene structure was different between runoff area and stagnant area. All of the nonparametric multivariate statistical tests of dissimilarity (MRPP, ANOSIM, and Adonis) showed that the overall functional structure of 3# coal production water’s microbial communities was different between runoff area and stagnant area (p < 0.005, Table 1). Cluster analysis of C-N-S function genes showed that the samples from runoff area and stagnant area were completely clustered into two groups, which further indicated the functional differences of microorganisms between the two groups (Fig.11a).
The relative abundance of genes associated to C, N and S cycling was increased from the runoff area to stagnant area (Fig. 9). In contrast, only few functional genes, which mainly belong to functions related to assimilatory nitrate reduction, were significantly (p < 0.005) decreased in relative abundance in the stagnant area. Next, it will be discussed in detail.