Genomic resolution of linkages in carbon, nitrogen, and sulfur cycling among widespread estuary sediment bacteria

Abstract

Background: Estuaries are among the most productive habitats on the planet. Bacteria in estuary sediments control the turnover of organic carbon and the cycling of nitrogen and sulfur. These communities are complex and primarily made up of uncultured lineages, thus little is known about how ecological and metabolic processes are partitioned in sediments.

Results: De novo assembly and binning resulted in the reconstruction of 82 bacterial genomes from different redox regimes of estuary sediments. These genomes belong to 23 bacterial groups, including uncultured candidate phyla (for example, KSB1, TA06, and KD3-62) and three newly described phyla (White Oak River (WOR)-1, WOR-2, and WOR-3). The uncultured phyla are generally most abundant in the sulfate-methane transition (SMTZ) and methane-rich zones, and genomic data predict that they mediate essential biogeochemical processes of the estuarine environment, including organic carbon degradation and fermentation. Among the most abundant organisms in the sulfate-rich layer are novel Gammaproteobacteria that have genes for the oxidation of sulfur and the reduction of nitrate and nitrite. Interestingly, the terminal steps of denitrification (NO3 to N2O and then N2O to N2) are present in distinct bacterial populations.

Conclusions: This dataset extends our knowledge of the metabolic potential of several uncultured phyla. Within the sediments, there is redundancy in the genomic potential in different lineages, often distinct phyla, for essential biogeochemical processes. We were able to chart the flow of carbon and nutrients through the multiple geochemical layers of bacterial processing and reveal potential ecological interactions within the communities.

Keywords: Anaerobic respiration; Candidate phyla; Carbon; Estuary; Metagenome; Nitrogen; Sediment; Sulfate reduction; Sulfur.

Figure 1

Diversity of organisms from which genomic bins were reconstructed from the White Oak River sediments. Phylogenetic tree inferred from 16 syntenous ribosomal protein genes present within genomic bins from the sediment metagenomic assemblies. Each sequence in bold is from one genomic bin. Genomic bins belonging to novel phyla for which no reference genomes are available (WOR-1, WOR-2, WOR-3, TA06, and KD3-62) were designated based on corresponding 16S rRNA gene phylogenetic analyses (Figure 2). The sample depths from which each of the bins were obtained are delineated by blue (shallow), green (SMTZ), and red (deep). The phylogeny was generated using the PhyML (maximum likelihood) method.

Figure 2

Phylogenetic tree of 16S rRNA genes present in bacterial genomic bins. Top hits from NCBI were included. Many of the White Oak River bacteria are most closely related to sequences recovered from other estuaries and coastal sediments. This tree was generated using the maximum likelihood method in the ARB alignment and phylogeny software package [59]. Closed circles represent maximum likelihood (RAxML, ARB package) bootstraps >75% and open circles are >50% values.

Figure 3

Glycoside hydrolases (GH) identified by CAZy searches of the genomic bins. GH families that contain enzymes that are not specifically involved in degradation were specifically identified by pfam or EC numbers in the annotations, based on Wrighton et al. [22].

Figure 4

Flow diagram of the potential interactions between (left to right) organic carbon utilization, fermentation, and respiration identified in the bacterial genomes reconstructed in this study. Arrows represent metabolic capabilities that were identified in the metagenomic reconstruction from the White Oak River estuary. The dashed lines on the right represent potential electron donors for the anaerobic respiration processes. Note that the Gammaproteobacteria are capable of coupling nitrate reduction to either thiosulfate or sulfide oxidation. Abbreviations in the diagram are as follows; DNRA, dissimilatory nitrate reduction to ammonia; ‘Betaprot’, Betaproteobacteria; ‘Deltaprot’, Deltaproteobacteria; ‘Gemmatio’, Gemmatimonadetes (‘Gemm38-2’ refers specifically to bin 38-2), ‘Gammaprot’, Gammaproteobacteria; ‘Myxococca’, Myxococcales; ‘Plancto’, Planctomycetes.

Figure 5

Diagrams of metabolic potential and electron transport of WOR-1 (bin DG-54-3) and Gammaproteobacteria (bin SG8-45), based on gene content. ATPase, ATP synthetase; FDH, formate dehydrogenase; NiFe-hyd, Ni,Fe-hydrogenase; Cytb/c1, quinone cytochrome oxidoreductase; Cyt c, cytochrome c; nap/nar, nitrate reductase; nir, nitrite reductase; nor, nitric oxide reductase; SAT, sulfate transferase; apr, APS reductase; rdsr, reverse dissimilatory sulfite reductase; Rub, RuBisCO; Q, quinine; SDH/FR, succinate dehydrogenase/fumarate reductase; NDH, NADH dehydrogenase; SOX, sulfur oxidation multienzyme complex. The sulfide quinone oxidoreductase (sqr) gene was not identified in this particular genomic bin, however many other closely related Gammaproteobacteria do have sqr genes.

Figure 6

Operons for sulfur oxidation and nitrate reduction present in the dominant Gammaproteobacteria genotypes. Those shown here are present in the SG8-45 bin. However, syntenous operons are also present in several other Gammaproteobacteria bins (SG8-11, SG8-15, SG8-45, SG8-47, SG8-50, STMZ1-46, and SMTZ-46).