Transient exposure to oxygen or nitrate reveals ecophysiology of fermentative and sulfate-reducing benthic microbial populations

Abstract

For the anaerobic remineralization of organic matter in marine sediments, sulfate reduction coupled to fermentation plays a key role. Here, we enriched sulfate-reducing/fermentative communities from intertidal sediments under defined conditions in continuous culture. We transiently exposed the cultures to oxygen or nitrate twice daily and investigated the community response. Chemical measurements, provisional genomes and transcriptomic profiles revealed trophic networks of microbial populations. Sulfate reducers coexisted with facultative nitrate reducers or aerobes enabling the community to adjust to nitrate or oxygen pulses. Exposure to oxygen and nitrate impacted the community structure, but did not suppress fermentation or sulfate reduction as community functions, highlighting their stability under dynamic conditions. The most abundant sulfate reducer in all cultures, related to Desulfotignum balticum, appeared to have coupled both acetate- and hydrogen oxidation to sulfate reduction. We describe a novel representative of the widespread uncultured candidate phylum Fermentibacteria (formerly candidate division Hyd24-12). For this strictly anaerobic, obligate fermentative bacterium, we propose the name '^U Sabulitectum silens' and identify it as a partner of sulfate reducers in marine sediments. Overall, we provide insights into the function of fermentative, as well as sulfate-reducing microbial communities and their adaptation to a dynamic environment.

Figure 1

Sulfide concentrations in the replicate cultures treated with oxygen (A), treated with nitrate (B) and in the untreated control with sulfate as sole electron acceptor (C). Duplicate measurements of each culture are shown as triangles and circles, the red line depicts the mean of four measurements, the red ribbon represents standard error of the mean. The start of the treatments is indicated by dashed lines, sampling time points for metagenomics and metatranscriptomics are indicated by dotted lines.

Figure 2

Estimated relative abundances of bins in cultures treated with oxygen (Oxy‐1, Oxy‐2), with nitrate (Nit‐1, Nit‐2) and in untreated controls (Con‐1, Con‐2). The bins were classified to genus level. Populations that affiliated with genera lacking a cultured representative are marked with #. For these bins, we report the closest taxonomically assigned, phylogenetic level, for example, bin K affiliated with an uncultured genus in the family Defluviitaleaceae. Taxonomic assignment was based on the SILVA small ribosomal subunit reference database (SSURef, v123). Relative abundances were obtained by mapping metagenomic sequence reads to the assembled contigs of each bin. The phylogeny of most bins is provided (bin O: Fig. 6, bin A‐N: Supporting Information Fig. S6‐S9).

Figure 3

Metabolic capabilities of the bins (A, B, D‐I, K‐P) based on key genes. Circle size represents relative transcriptional activity averaged over all ten transcriptomes. The absence of a circle shows that gene transcription was not detected in any of the treatments and cultures. Note: For reasons of visualization, all relative transcriptional activities above 3.0 are shown as ≥3.0. Metabolic capabilities were not assessed for the bins C, J, Q and R due to scarce metagenomic data (Table 1).

Figure 4

Schematic of the trophic network of key populations (A, B, D‐I, K‐P), and transcriptional changes of stress response genes. A. Most abundant obligate fermentative heterotrophs (Fermenters), sulfate‐reducing bacteria (Respirers I) and associated respiratory heterotrophs (Respirers II) in the three different conditions. The network is based on metagenomic and/–transcriptomic data. All 14 shown bins were present in all cultures. Circle size represents estimated relative abundance. Only one organism (bin D) was abundant in all cultures. Arrows depict key pathways that occur in all (grey), two (blue) or one condition (red). Bins C, J, Q and R were not included due to scarce metagenomic data (Table 1). B. Change of gene transcription caused by the treatment with oxygen or nitrate. Transcriptional changes of genes involved in redox‐ and general stress response were among the highest of all detected genes. Values are log2‐transformed ratios of gene transcription in replicate cultures after and before the treatment, that is, a value of 1 means that gene transcription was twice as high after the treatment than before the treatment; a value of −2 means a fourfold decrease in transcription. The guilds are represented by the 11 most abundant bins for clarity.

Figure 5

Metabolic map of ^USabulitectum silens (bin O) showing central pathways that the organism transcribed in the sulfate‐only treatment (Con‐5, Con‐6). Transcribed genes are shown as blue arrows, genes of annotated pathways that were not detected as red arrows. Enzymes are abbreviated with letters, the full list as well as further metabolic pathways are provided in Supporting Information Table S3. Dashed blue circles depict additional pathways that were detected.

Figure 6

Phylogeny of candidate phylum Fermentibacteria, showing the affiliation of all publicly available, non‐redundant 16S rRNA gene sequences, including the provisional species ^USabulitectum silens and Ca. Fermentibacter daniensis (black). The phylum comprises one class (at a threshold sequence identity of 78.5%), one order (at 82%), four families (at 86.5%) and at least nine genera (at 94.5%). The origin of the sequences is color‐coded (red: methane seeps; red bold: anaerobic methanotrophic enrichment cultures; orange: hypersaline mats; pink: springs; light blue: dolphin, dark blue: anaerobic digesters: grey: other) and indicates niche‐differentiation among Fermentibacteria. An extensive list of ecosystems harbouring Fermentibacteria is provided in the Supplementary Results. Phylogeny is based on the SILVA small subunit ribosomal database SSURef 123.1 (released 03/2016). The scale bar shows estimated sequence divergence. Fermentibacteria sequence alignments and phylogeny are provided as an ARB database (Dataset 3). The parameters that were used to compile the sequence database are described in the Supporting Information.