Abyssal Zone, Metagenomics of
KeywordsAutonomous Underwater Vehicle Photic Zone Marine Group North Pacific Subtropical Gyre Abyssal Depth
The set of tools and methods used to characterize entire communities of deep-sea organisms by extracting and sequencing their DNA without isolation and culturing. In oceanography “abyssal” is defined as the zone of the cold deep sea at depths between 4,000 and 6,000 m, but the term is used here in a less strict sense to include any sample taken at >4,000 m, including the hadal zone.
In recent years rapid advancements in DNA sequencing technology have allowed the characterization of the collective genome sequence of entire communities of organisms. This has opened new avenues for understanding the genetic basis for environmental adaptations. This is particularly relevant to the study of organisms living in extreme or difficult-to-sample environments, where individual members of the community are often difficult or impossible to grow under laboratory conditions. Metagenome sequences are now available for environments spanning extremes of temperature, pH, radiation, and salinity. However, despite the fact that most of the ocean’s volume is at depth >1,000 m, deep-sea metagenomics is only making its first steps. The abyssal (also known as abyssopelagic) environment is characterized by the absence of light, which limits primary production and nutrient availability, and by low temperature and high hydrostatic pressure which affects to a large extent enzymatic activities and physiologies. Within this relatively barren environment, other small isolated areas exist with a different biogeochemistry, such as hydrothermal vents, cold seeps, and whale falls (Lauro and Bartlett 2008). This review focuses on the bulk deep-sea environment and does not cover these smaller niche zones.
Sample Collection and DNA Sequencing
The development of environmental genomics of the abyssal zone has been largely impaired by the availability of easy and cost-effective sampling methodology. Issues such as sample decompression, sample amount, and recovery time (defined as the time elapsed between sample collection and processing) have largely restricted the study of deep-sea metagenomics to the microbial domains of Bacteria and Archaea. These organisms lack any gas-filled spaces which could expand during decompression and therefore can tolerate large isothermal pressure changes for short periods of time (Chastain and Yayanos 1991).
Most samples have been collected using Niskin bottles (Martín-Cuadrado et al. 2007), but this approach is limited to the amount of wire cable that can be carried on oceanographic vessels (usually 6,000–8,000 m). Moreover the abyssal deployment of Niskin rosettes is largely dependent on weather conditions. Further, the time between sample collection and availability for processing can be up to several hours as the ship’s winches recover the bottles from the abyssal depths.
Over the last few years some technological advancements have improved this situation. For example, recent sampling ventures have benefited (albeit not without considerable expense) from the use of autonomous underwater vehicles (e.g., Takami et al. 1997), free vehicles capable of descent to the abyssal ocean, sample retrieval, and rapid return to the surface after a predetermined amount of time at the bottom (e.g., Eloe et al. 2011b), and, more recently, manned submarines capable of reaching the deepest depths of the ocean such as the bottom of the Mariana Trench (http://deepseachallenge.com/).
The throughput and cost-effectiveness of sequencing have also improved with technological advances: from the initial Sanger end-sequencing of fosmid clones (DeLong et al. 2006; Martín-Cuadrado et al. 2007) to 454 pyrotag sequencing of 16S rRNA genes (Agogué et al. 2011; Brown et al. 2009) to shotgun pyrosequencing of whole communities (Eloe et al. 2011a). This trend is expected to continue with studies being currently undertaken with the latest generation Illumina’s Solexa or SOLiD platforms (Schuster 2008) and sequencers based on newer technologies (e.g., single-molecule real-time sequencing) appearing on the market (Korlach et al. 2010).
Phylogenetic and Functional Diversity Gleaned from Abyssal Metagenomics
Analyses of phylogenetic diversity have been performed (1) over a depth profile at the Hawaii Ocean Time-Series Station, ALOHA, in the North Pacific Subtropical Gyre (NPSG), and from the ocean’s surface to abyssal depths (4,000–4,400 m) (DeLong et al. 2006; Lauro and Bartlett 2008; Brown et al. 2009); (2) from the Puerto Rico Trench (PRT) from a depth in excess of 6,000 m (Eloe et al. 2010, 2011a); and (3) along a North Atlantic Ocean transect down to 4,500 m (Agogué et al. 2011). Metagenomic studies of the abyss have shown that microbes at these depths exhibit a remarkable phylogenetic diversity. The microbiota is dominated by members of the Proteobacteria, especially Alphaproteobacteria and to a lesser extent Gammaproteobacteria, and Bacteroidetes. Other bacterial clades have also been detected at abyssal depths including Beta-, Delta-, and Epsilonproteobacteria, Actinobacteria, Verrucomicrobia, Planctomycetes, Chloroflexi, Acidobacteria, Firmicutes, and Gemmatimonadetes. Additionally, archaeal clades were detected at abyssal depths, belonging to Euryarchaeota and Crenarchaeota.
Some bacterial groups show a distinct change in abundance with depth. Alphaproteobacteria, Gammaproteobacteria, Cyanobacteria, and Bacteroidetes are typically the dominant bacterial clades at the ocean surface. Whereas Cyanobacteria decline rapidly with depth (consistent with a phototrophic physiology), members of Alphaproteobacteria, Gammaproteobacteria, and Bacteroidetes were shown to persist throughout the water column to abyssal depths and therefore inferred to be major contributors to biogeochemical cycling at abyssal depths (DeLong et al. 2006; Brown et al. 2009). Further, the relative abundance of these particular groups remained comparable across various depths, from surface to abyssal (Brown et al. 2009). However, North Atlantic metagenome studies showed that Gammaproteobacteria abundance increased with depth (Agogué et al. 2011). The Euryarchaeota Marine Group II showed a decline from surface (epipelagic) to abyssal depths of NPSG, whereas Euryarchaeota Marine Group III exhibited a much lower frequency that was fairly constant across all depths (DeLong et al. 2006; Brown et al. 2009). Crenarchaeota Marine Group I (=Thaumarchaeota) accounted for nearly half of Archaea tag sequences at 4,400 m of NPSG, whereas Euryarchaeota Marine Group II and III made up around 37 % and 11 %, respectively. A small subunit ribosomal survey of the PRT similarly recovered a dominance of Crenarchaeota Marine Group I compared to Euryarchaeota Marine Group II (Eloe et al. 2010). However, a subsequent survey of archaeal ribosomal genes from the PRT recovered Euryarchaeota Marine Group II as the dominant archaeal clade compared to Crenarchaeota Marine Group I (Eloe et al. 2011a).
When particle-associated (>3 mm) and free-living (3–0.22 mm) microbes at PRT were assessed separately, the former was dominated by Bacteroidetes, Planctomycetes, Rhodobacterales, Rhizobiales (both Alphaproteobacteria), and Myxococcales (Deltaproteobacteria) (Eloe et al. 2010). As with Bacteroidetes at the ocean surface, abyssal members of this clade of heterotrophs likely prefer polymeric organic matter contained within detrital particles. Particle-associated Rhizobiales were most closely related to sequences derived from soil and sediments, as observed in 4,400 m samples from the NPSG (Brown et al. 2009), which suggests a similar mode of metabolism between abyssal and soil/sediment Rhizobiales, with these bacteria perhaps targeting similar refractory compounds (Eloe et al. 2010). Around 40 % of the bacterial sequences recovered from the two size fractions belonged to Alphaproteobacteria, with the highest numbers of sequences for the ubiquitous SAR11 clade (Eloe et al. 2010). This attests to the success of the SAR11 clade at all levels of the water column. At the surface, SAR11 are oligotrophs that scavenge nanomolar concentrations of labile solutes and contain ion-translocating photoproteins (proteorhodopsins) (Giovannoni et al. 2005). However, this latter ability is unlikely to function in SAR11 outside the photic zone. Epipelagic members of Euryarchaeota Marine Group II also have proteorhodopsin, but this is absent from members of this group below the photic zone (Frigaard et al. 2006; Martin-Cuadrado et al. 2007). The lack of genes involved in photosynthetic processes and photoactive proteins is more broadly characteristic of deep-water communities at depths below the photic zone. The photic zone (or epipelagic zone) extends to a depth of around 200 m below the surface of the ocean and is that part of the water column where there is adequate sunlight for photosynthesis to occur. However, some solar illumination can penetrate into the mesopelagic zone (or “twilight zone”) (~200–1,000 m below the ocean surface), although it is inadequate for photosynthesis. The bathypelagic zone (or midnight zone), which is the upper layer of the deep sea, extends from a depth of 1,000–4,000 m and sunlight is completely absent. Thus, in the abyssal deep sea, all primary production is independent of sunlight.
The continuity of certain clades from the ocean surface down to abyssal depths raises the question of whether the physiologies known or inferred for non-deep-water members of microbial clades are useful for reconstructing the physiologies of abyssal clades. The association of Bacteroidetes with abyssal particulate matter is consistent with the preference for complex polymeric substrates which (based on surface and human commensal species) is typical for this group. Nevertheless, the genomic potential of an abyssal member of this clade for assimilatory nitrate reduction (Eloe et al. 2011a) sets this member of Bacteroidetes apart from typical epipelagic members of this group that derive nitrogen exclusively from organic sources. Piezophilic Gammaproteobacteria have been isolated and cultured that relate to the genera Colwellia, Moritella, Photobacterium, Psychromonas, and Shewanella (Lauro and Bartlett 2008; Lauro et al. 2009); thus, their physiologies have been determined experimentally.
To date, neither Euryarchaeota Marine Groups II nor III has been cultured, and genomic fragments provide few clues about the physiologies of deep-water members of these groups (Martin-Cuadrado et al. 2007). Crenarchaeota Marine Group I accounted for nearly half of Archaea tag sequences at 4,400 m of NPSG, with sequences closely related to Nitrosopumilus maritimus and Candidatus Cenarchaeum symbiosum making up around 5 % and 12 % of Archaea sequences, respectively. N. maritimus is capable of aerobic autotrophic carbon fixation via the 3-hydroxypropionate/4-hydroxybutyrate cycle coupled to ammonia oxidation (Walker et al. 2010), and the same metabolism has been inferred for Ca. C. symbiosum (Hallam et al. 2006). Thus, this physiology may persist throughout the water column down to abyssal depths. Crenarchaeota Marine Group I were not detected at the ocean surface of NPSG (DeLong et al. 2006; Brown et al. 2009), suggesting that “dark autotrophs” within this group are critically important in carbon fixation in the absence of light. Finally, Methanopyri (another group within Euryarchaeota) were also detected at 4,400 m of NPSG; these too were also undetected at the surface (Brown et al. 2009). Abyssal Methanopyri are likely to be anaerobic methanogens based on the close phylogenetic affinities they share with cultivated Methanopyri.
At 4,400 m at NPSG, the Novel Alveolate Groups I and II of Eucarya collectively contributed 21.2 % of tag sequence abundance. These clades include endoparasitic dinoflagellates that target phylogenetically disparate Eucarya; if the abyssal sequences similarly represent organisms that exhibit a potentially lethal parasitic lifestyle, then Novel Alveolate Groups I and II may be major contributors to the microbial loop, via the mass release of spores from an infected or dead host (Brown et al. 2009). Overall, much of the apparent eucaryal diversity is yet to be adequately described, as indicated by the degree of novelty shown by Eucarya tag sequences compared with those of the Bacteria and Archaea (Brown et al. 2009).
One complicating factor in the culture-independent studies of deep-sea samples is the problem of discriminating against allochthonous, surface-derived microbes attached to sinking particles that fall all the way to the sea floor. An understanding of biogeochemical cycling at abyssal depths requires a means of distinguishing between autochthonous (i.e., indigenous) and allochthonous (i.e., introduced) members of communities. Abyssal metagenomes have chloroplast and cyanobacterial sequences, which are clearly not active members of the indigenous community (Brown et al. 2009; Eloe et al. 2010). One possible solution is to prefilter the collected water samples in order to remove surface-derived detritus; but this also removes those autochthonous microbes that target and/or attach to particulate matter, such as Bacteroidetes (see above). It has been suggested that the higher abundances of Gammaproteobacteria in the North Atlantic versus NPSG abyssal samples might have been partly due to the latter sample being prefiltered, which may have resulted in particle-attached members of the Gammaproteobacteria being filtered out but Alphaproteobacteria passing through the prefilter (Lauro and Bartlett 2008).
In general, the phylogenetic diversity of Bacteria and Archaea points to diverse metabolic strategies in the abyss. This is consistent with a functional analysis of the PRT metagenome (Eloe et al. 2011a). Gene-encoding enzymes involved in the major autotrophic pathways used by bacteria and archaea were detected: Calvin-Benson-Bassham cycle, reductive tricarboxylic acid cycle, 3-hydroxypropionate cycle, reductive acetyl-CoA pathway, 3-hydroxypropionate/4-hydroxybutyrate cycle, and dicarboxylate/4-hydroxybutyrate cycle. However, genes for key enzymes in these pathways were missing or poorly represented, leading to the hypothesis that autotrophic carbon fixation pathways play a minor role compared to heterotrophic metabolic strategies at the PRT (Eloe et al. 2011a).
The PRT metagenome was also enriched in genes for aerobic carbon monoxide (CO) oxidation, associated with the use of CO as an energy source, with CO possibly derived from the anaerobic metabolism of organic matter (Eloe et al. 2011a). An enrichment of the PRT metagenome in transcriptional regulators and genes with signal transduction domains supports the hypothesis that deep-ocean microbial assemblages possess functions to cope with resource scarcity and a high diversity of molecular substrates. The enrichment of both proton- and ATP-driven efflux systems indicates diverse mechanisms to deal with elevated concentrations of trace metals. A high abundance of genes for sulfatases for the degradation of sulfated polysaccharides was also present in the PRT (Eloe et al. 2011a), which is consistent with an enrichment of genes associated with sulfur metabolism and methionine biosynthesis at abyssal depths of the NPSG (Brown et al. 2009).
As well as transposases, abyssal sequences were relatively enriched for genes involved in protein folding and processing (Eloe et al. 2011a). Protein synthesis is inhibited by hydrostatic pressure, as a result of the dissociation of the ribosomal subunits, rather than pressure denaturation of proteins (Lauro and Bartlett 2008). This impaired ribosomal function at extreme depth leads to truncated and misfolded proteins (Hörmann et al. 2006; Lauro and Bartlett 2008). Chaperones involved in guiding the elongation and correct folding of polypeptides and refolding misfolded proteins might be expected to be an important adaptation to life at abyssal depths.
Surprisingly little is known about the organisms thriving in the abyssal zones and all metagenomic studies to date have been restricted to Bacteria and Archaea. This is largely due to not having economical methods for collecting and rapidly processing the samples. However, it is now clear that the growth and survival of organisms in the deep sea is determined to a great extent by the gradients of physicochemical factors that covary with depth. With the current trends in decreasing cost of sequencing and the availability of new technologies for sampling and exploration of the abyssal zone, ongoing and future metagenomic-based studies will ensure that a better understanding of deep-sea physiology, biochemistry, and nutrient cycling in an ecological context is obtained. By combining these types of studies with targeted functional studies, the next few years will witness a revolution from knowing “who is there” to “who is doing what.”
- Frigaard NU, Martine A, Mincer TJ, De long EF proteorhodopsin lateral gene transfer between marine planktonic bacteria and arciraea nature. 2006; 439:847–850.Google Scholar