Members of the marine Roseobacter lineage have been characterized as ecological generalists, suggesting that there will be challenges in assigning well-delineated ecological roles and biogeochemical functions to the taxon. To address this issue, genome sequences of 32 Roseobacter isolates were analyzed for patterns in genome characteristics, gene inventory, and individual gene/ pathway distribution using three predictive frameworks: phylogenetic relatedness, lifestyle strategy and environmental origin of the isolate. For the first framework, a phylogeny containing five deeply branching clades was obtained from a concatenation of 70 conserved single-copy genes. Somewhat surprisingly, phylogenetic tree topology was not the best model for organizing genome characteristics or distribution patterns of individual genes/pathways, although it provided some predictive power. The lifestyle framework, established by grouping isolates according to evidence for heterotrophy, photoheterotrophy or autotrophy, explained more of the gene repertoire in this lineage. The environment framework had a weak predictive power for the overall genome content of each strain, but explained the distribution of several individual genes/pathways, including those related to phosphorus acquisition, chemotaxis and aromatic compound degradation. Unassembled sequences in the Global Ocean Sampling metagenomic data independently verified this global-scale geographical signal in some Roseobacter genes. The primary findings emerging from this comparative genome analysis are that members of the lineage cannot be easily collapsed into just a few ecologically differentiated clusters (that is, there are almost as many clusters as isolates); the strongest framework for predicting genome content is trophic strategy, but no single framework gives robust predictions; and previously unknown homologs to genes for H 2 oxidation, proteorhodopsin-based phototrophy, xanthorhodpsin-based phototrophy, and CO 2 fixation by Form IC ribulose-1,5-bisphosphate carboxylase/oxygenase (RuBisCO) expand the possible mechanisms for energy and carbon acquisition in this remarkably versatile bacterial lineage.

About half the carbon fixed by phytoplankton in the ocean is taken up and metabolized by marine bacteria, a transfer that is mediated through the seawater dissolved organic carbon (DOC) pool. The chemical complexity of marine DOC, along with a poor understanding of which compounds form the basis of trophic interactions between bacteria and phytoplankton, have impeded efforts to identify key currencies of this carbon cycle link. Here, we used transcriptional patterns in a bacterial-diatom model system based on vitamin B 12 auxotrophy as a sensitive assay for metabolite exchange between marine plankton. The most highly up-regulated genes (up to 374-fold) by a marine Roseobacter clade bacterium when cocultured with the diatom Thalassiosira pseudonana were those encoding the transport and catabolism of 2,3-dihydroxypropane-1-sulfonate (DHPS). This compound has no currently recognized role in the marine microbial food web. As the genes for DHPS catabolism have limited distribution among bacterial taxa, T. pseudonana may use this sulfonate for targeted feeding of beneficial associates. Indeed, DHPS was both a major component of the T. pseudonana cytosol and an abundant microbial metabolite in a diatom bloom in the eastern North Pacific Ocean. Moreover, transcript analysis of the North Pacific samples provided evidence of DHPS catabolism by Roseobacter populations. Other such biogeochemically important metabolites may be common in the ocean but difficult to discriminate against the complex chemical background of seawater. Bacterial transformation of this diatom-derived sulfonate represents a previously unidentified and likely sizeable link in both the marine carbon and sulfur cycles.

The potential of metatranscriptomic sequencing to provide insights into the environmental factors that regulate microbial activities depends on how fully the sequence libraries capture community expression (that is, sample-sequencing depth and coverage depth), and the sensitivity with which expression differences between communities can be detected (that is, statistical power for hypothesis testing). In this study, we use an internal standard approach to make absolute (per liter) estimates of transcript numbers, a significant advantage over proportional estimates that can be biased by expression changes in unrelated genes. Coastal waters of the southeastern United States contain 1 × 1012 bacterioplankton mRNA molecules per liter of seawater (∼200 mRNA molecules per bacterial cell). Even for the large bacterioplankton libraries obtained in this study (∼500 000 possible protein-encoding sequences in each of two libraries after discarding rRNAs and small RNAs from >1 million 454 FLX pyrosequencing reads), sample-sequencing depth was only 0.00001%. Expression levels of 82 genes diagnostic for transformations in the marine nitrogen, phosphorus and sulfur cycles ranged from below detection (<1 × 106 transcripts per liter) for 36 genes (for example, phosphonate metabolism gene phnH, dissimilatory nitrate reductase subunit napA) to >2.7 × 109 transcripts per liter (ammonia transporter amt and ammonia monooxygenase subunit amoC). Half of the categories for which expression was detected, however, had too few copy numbers for robust statistical resolution, as would be required for comparative (experimental or time-series) expression studies. By representing whole community gene abundance and expression in absolute units (per volume or mass of environment), ‘omics’ data can be better leveraged to improve understanding of microbially mediated processes in the ocean.

Resolving the ecological niches of coexisting marine microbial taxa is challenging due to the high species richness of microbial communities and the apparent functional redundancy in bacterial genomes and metagenomes. Here, we generated over 11 million Illumina reads of protein-encoding transcripts collected from well-mixed southeastern US coastal waters to characterize gene expression patterns distinguishing the ecological roles of hundreds of microbial taxa sharing the same environment. The taxa with highest in situ growth rates (based on relative abundance of ribosomal protein transcripts) were typically not the greatest contributors to community transcription, suggesting strong top-down ecological control, and their diverse transcriptomes indicated roles as metabolic generalists. The taxa with low in situ growth rates typically had low diversity transcriptomes dominated by specialized metabolisms. By identifying protein-encoding genes with atypically high expression for their level of conservation, unique functional roles of community members emerged related to substrate use (such as complex carbohydrates, fatty acids, methanesulfonate, taurine, tartrate, ectoine), alternative energy-conservation strategies (proteorhodopsin, AAnP, V-type pyrophosphatases, sulfur oxidation, hydrogen oxidation) and mechanisms for negotiating a heterogeneous environment (flagellar motility, gliding motility, adhesion strategies). On average, the heterotrophic bacterioplankton dedicated 7% of their transcriptomes to obtaining energy by non-heterotrophic means. This deep sequencing of a coastal bacterioplankton transcriptome provides the most highly resolved view of bacterioplankton niche dimensions yet available, uncovering a spectrum of unrecognized ecological strategies.

The Deepwater Horizon blowout released a massive amount of oil and gas into the deep ocean between April and July 2010, stimulating microbial blooms of petroleum-degrading bacteria. To understand the metabolic response of marine microorganisms, we sequenced B66 million community transcripts that revealed the identity of metabolically active microbes and their roles in petroleum consumption. Reads were assigned to reference genes from B2700 bacterial and archaeal taxa, but most assignments (39%) were to just six genomes representing predominantly methane-and petroleum-degrading Gammaproteobacteria. Specific pathways for the degradation of alkanes, aromatic compounds and methane emerged from the metatranscriptomes, with some transcripts assigned to methane monooxygenases representing highly divergent homologs that may degrade either methane or short alkanes. The microbial community in the plume was less taxonomically and functionally diverse than the unexposed community below the plume; this was due primarily to decreased species evenness resulting from Gammaproteobacteria blooms. Surprisingly, a number of taxa (related to SAR11, Nitrosopumilus and Bacteroides, among others) contributed equal numbers of transcripts per liter in both the unexposed and plume samples, suggesting that some groups were unaffected by the petroleum inputs and blooms of degrader taxa, and may be important for re-establishing the pre-spill microbial community structure.

Quantitative PCR (qPCR) analysis revealed elevated relative abundance (1.8% of prokaryotes) of marine group 1 Crenarchaeota (MG1C) in two samples of southeastern US coastal bacterioplankton, collected in August 2008, compared with samples collected from the same site at different times (mean 0.026%). We analyzed the MG1C sequences in metatranscriptomes from these samples to gain an insight into the metabolism of MG1C population growing in the environment, and for comparison with ammonia-oxidizing bacteria (AOB) in the same samples. Assemblies revealed low diversity within sequences assigned to most individual MG1C open reading frames (ORFs) and high homology with 'Candidatus Nitrosopumilus maritimus' strain SCM1 genome sequences. Reads assigned to ORFs for ammonia uptake and oxidation accounted for 37% of all MG1C transcripts. We did not recover any reads for Nmar_1354-Nmar_1357, proposed to encode components of an alternative, nitroxyl-based ammonia oxidation pathway; however, reads from Nmar_1259 and Nmar_1667, annotated as encoding a multicopper oxidase with homology to nirK, were abundant. Reads assigned to two homologous ORFs (Nmar_1201 and Nmar_1547), annotated as hypothetical proteins were also abundant, suggesting that their unknown function is important to MG1C. Superoxide dismutase and peroxiredoxin-like transcripts were more abundant in the MG1C transcript pool than in the complete metatranscriptome, suggesting an enhanced response to oxidative stress by the MG1C population. qPCR indicated low AOB abundance (0.0010% of prokaryotes), and we found no transcripts related to ammonia oxidation and only one RuBisCO transcript among the transcripts assigned to AOB, suggesting they were not responding to the same environmental cues as the MG1C population.

Significance The microbial community of the Amazon River Plume determines the fate of the world’s largest input of terrestrial carbon and nutrients to the ocean. By benchmarking with internal standards during sample collection, we determined that each liter of plume seawater contains 1 trillion genes and 50 billion transcripts from thousands of bacterial, archaeal, and eukaryotic taxa. Gene regulation by taxa inhabiting distinct microenvironments provides insights into micron-scale patterns of transformations in the marine carbon, nitrogen, phosphorus, and sulfur cycles in this globally important ecosystem.

The ecological significance of the marine bacterial populations distinguishable by flow cytometry on the basis of the fluorescence (FL) of their nucleic acid (NA) content and proxies of cell size (such as side scatter, SSC) remains largely unknown. Some studies have suggested that cells with high NA (HNA) content and high SSC (HS) represent the active members of the community, while the low NA (LNA) cells are inactive members of the same phylogenetic groups. But group-specific activity measurements and phylogenetic assignment after cell sorting have suggested this is not be the case, particularly in open-ocean communities. To test the extent to which the different NA subgroups are similar, and consequently the extent to which they likely have similar ecological and biogeochemical roles in the environment, we analysed the phylogenetic composition of three populations after cell sorting [high NA-high SC (HNA-HS), high NA-low SC (HNA-LS), low NA (LNA)] by 454 pyrosequencing in two contrasting periods of the year in NW Mediterranean coastal waters (BBMO, Blanes Bay Microbial Observatory) where these three populations have recurrent seasonal patterns. Statistical analyses showed that summer and winter samples were significantly different and, importantly, the sorted populations within a sample were composed of different taxa. The majority of taxa were associated with one NA fraction only, and the degree of overlap (i.e. OTUs present simultaneously in 2 fractions) between HNA and LNA and between summer and winter communities was very small. Rhodobacterales, SAR116 and Bacteroidetes contributed primarily to the HNA fraction, whereas other groups such as SAR11 and SAR86 contributed largely to the LNA fractions. Gammaproteobacteria other than SAR86 showed less preference for one particular NA fraction. An increase in diversity was observed from the LNA to the HNA-HS fraction for both sample dates. Our results suggest that, in Blanes Bay, flow cytometric signatures of natural communities track their phylogenetic composition.