Skip to main content

Thank you for visiting You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

Bacterial diversity and community composition from seasurface to subseafloor


We investigated compositional relationships between bacterial communities in the water column and those in deep-sea sediment at three environmentally distinct Pacific sites (two in the Equatorial Pacific and one in the North Pacific Gyre). Through pyrosequencing of the v4–v6 hypervariable regions of the 16S ribosomal RNA gene, we characterized 450 104 pyrotags representing 29 814 operational taxonomic units (OTUs, 97% similarity). Hierarchical clustering and non-metric multidimensional scaling partition the samples into four broad groups, regardless of geographic location: a photic-zone community, a subphotic community, a shallow sedimentary community and a subseafloor sedimentary community (1.5 meters below seafloor). Abundance-weighted community compositions of water-column samples exhibit a similar trend with depth at all sites, with successive epipelagic, mesopelagic, bathypelagic and abyssopelagic communities. Taxonomic richness is generally highest in the water-column O2 minimum zone and lowest in the subseafloor sediment. OTUs represented by abundant tags in the subseafloor sediment are often present but represented by few tags in the water column, and represented by moderately abundant tags in the shallow sediment. In contrast, OTUs represented by abundant tags in the water are generally absent from the subseafloor sediment. These results are consistent with (i) dispersal of marine sedimentary bacteria via the ocean, and (ii) selection of the subseafloor sedimentary community from within the community present in shallow sediment.


Previous studies indicate that microbial community composition varies from one marine environment to another (Delong et al., 2006; Zinger et al., 2011; Hamdan et al., 2013), but can be relatively consistent in similar marine environments separated by long distances (Inagaki et al., 2006; Agogué et al., 2011; Walsh et al., 2015). For example, adjacent marine water masses with different environmental properties have distinct microbial communities (Agogué et al., 2011; Hamdan et al., 2013). Similarly, subseafloor sedimentary environments with different properties separated by a few tens of kilometers also have distinct communities (Inagaki et al., 2006; Hewson et al., 2007; Schauer et al., 2010). Despite these differences over relatively short geographic distances, microbial community composition in individual deep-seawater masses can be relatively constant for thousands of kilometers (Agogué et al., 2011). And broadly, similar microbial communities inhabit similar subseafloor sedimentary environments separated by thousands of kilometers (Inagaki et al., 2006). These observations are consistent with the old adage, ‘Everything is everywhere but the environment selects’ (Baas-Becking, 1934), in which microorganisms are considered to be ubiquitously dispersed because of their small size, large numbers and low extinction rates (Martiny et al., 2006). Despite these intriguing biogeographic patterns, few previous studies have examined (i) the vertical distribution of bacterial taxa throughout the entire open-ocean water column, or (ii) the relationship of microbial communities in marine sediment to those in the overlying water.

Distributions of abundant bacterial taxa in surface marine sediment (to 1.1 meters below seafloor (mbsf)) suggest that the seafloor is colonized via the overlying water column (Hamdan et al., 2013). However, the relationship of microbial communities in deep subseafloor sediment (1.5 mbsf) to those in surface marine sediment and the overlying ocean has not been previously examined.

Vertically sampling the water column and sediment in distinct oceanographic regions provides a means to evaluate the influence of local oceanographic properties such as nutrient availability, sedimentation rate and distance from land (Fuhrman et al., 2006; Kallmeyer et al., 2012) on microbial diversity throughout both water column and sediment. High productivity in surface water can lead to the formation of oxygen-deficient zones in deeper water, where unique communities may form in response to the deoxygenation (Beman and Carolan, 2013). In high-productivity regimes, such as the eastern equatorial Pacific upwelling region (EQP-1) and the moderately productive central equatorial Pacific upwelling region (EQP-8), a relatively high flux of organic debris to the seafloor sustains an active anaerobic subseafloor sedimentary community (D'Hondt et al., 2004). In contrast, low-productivity regions, such as the North Pacific gyre (EQP-11), are characterized by a continuously oxic water column, extremely low rates of sedimentation and organic matter deposition, and a relatively low-activity aerobic subseafloor community (D'Hondt et al., 2009; Røy et al., 2012; D'Hondt et al., 2015).

(i) To document vertical distributions of bacterial diversity and community composition, (ii) to investigate how these communities vary from one oceanographic region to another, and (iii) to test the extent to which bacterial communities of deep subseafloor sediment may originate in the water column, we used 454 pyrosequencing technology and the v4–v6 hypervariable region of the bacterial 16S ribosomal RNA (rRNA) gene to examine bacterial community composition (presence–absence and relative abundance) in the water column, near-seafloor sediment (0–10 centimeters below seafloor (cmbsf)), and subseafloor sediment at three environmentally distinct Pacific sites: the very high-productivity eastern equatorial upwelling region (EQP-1), the moderately high-productivity open-ocean central equatorial upwelling region (EQP-8) and the very low-productivity northern gyre (EQP-11) (Figure 1).

Figure 1

Locations of sampling stations. Superimposed chlorophyll a content is annual average data (mean from September 1994–December 2004) from the SeaWiFS satellite uploaded into GeoMapApp (Behrenfeld and Falkowski, 1997; Gregg et al., 2005).

Materials and methods

Sampling methodology

We collected the sediment and seawater samples during a 43-day transect of the R/V Knorr through the eastern equatorial Pacific and the North Pacific gyre (Knorr Expedition 195-3). We collected the water samples using a 24-Niskin bottle CTD rosette (Sea-Bird SBE-911 CTD system plus package). Once the CTD/Niskin system was on deck, we clamped the Niskin bottles and applied compressed air to the vent plugs to enhance filtration. We filtered 5 l from each bottle onto 0.2 μM Supor-200 47 mm filters at a rate of approximately 200 ml min−1, and then stored the filters at −80 °C. We targeted a combination of standard depths and oceanographic horizons (seasurface, O2 minimum, chlorophyll a maximum, thermocline and deep water) and analyzed 7–14 sampling depths at each station from 3 to 5500 m water depth depending on location. We measured water depths using the center depth of the SeaBeam 2112 multibeam. We estimated the average sedimentation rate for each site by dividing the crustal age (Müller et al., 2008) by the sediment thickness in the NDGC global thickness grid (Divins, 2003). We used mean annual average chlorophyll a data (Behrenfeld and Falkowski, 1997; Gregg et al., 2005), collected from September 1994–December 2004, as a proxy for seasurface productivity.

At each station, we collected sediment using a multi-corer (0–30 cmbsf]), gravity corer (up to 4 mbsf) and long-piston coring device (up to 35 mbsf). We subsampled the sediment using sterile 60 cc cut-off syringes and immediately froze the subsamples at −80 °C for later DNA extraction. For each site in this study, we analyzed one sediment sample taken with the multi-corer at the sediment-water interface (0–10 cmbsf) and two to five sediment samples taken at greater depths (between 0.25 and 34 mbsf). In total, this study includes 39 tag-pyrosequencing samples: 28 from the water column, 3 from the sediment-water interface and 8 from subseafloor sediment.

Site descriptions

All three sites are located in the deep open ocean, with fully pelagic sediment.

EQP-1 (1° 48.21’ N, 86° 11.29’ W) was the easternmost site in our equatorial Pacific transect. This site is located in a relatively flat region of 1.6-Ma basement (Müller et al., 2008), with water depth of 2856 m (Figure 1). It is in the eastern equatorial upwelling regime. Consequently, it contains a highly productive ecosystem and is characterized by a high mean sedimentation rate (~75 m Ma–1). Its sediment is predominantly foraminifera-rich nannofossil ooze. Total sediment thickness is estimated to be 120 m (Divins, 2003). We began sampling with the CTD/Niskin cast on 14 January 2009 (bottles on deck 19:33 GMT). The site has a pronounced O2 minimum zone. The chlorophyll maximum was strongly developed at our time of sampling (at 34 m) and located above the depth of the pycnocline. The O2 minimum zone was within the thermocline and contained O2 concentrations as low as 6 μmol kg–1 (Figure 2). In the sediment, dissolved oxygen was limited to the upper 7 mm (Røy et al., 2012).

Figure 2

Scatterplot of sequencing and CTD data including (from left to right): OTUs (97%), oxygen, chlorophyll a, temperature and density.

EQP-8 (0° 0.36’ N, 147° 47.50’ W) was the westernmost site of our equatorial transect (Figure 1). It is located on a flat abyssal plain, with basement age of 80 Ma (Müller et al., 2008) and water depth of 4336 m. Total sediment thickness is estimated to be 380 m (Divins, 2003). This site is in the Central Pacific equatorial upwelling region. Consequently, it has high surface productivity for an open-ocean location. Its sediment is calcareous nannofossil ooze. The mean sedimentation rate at this site (4.8 m Ma–1) is substantially lower than at EQP-1. We conducted water-column sampling here on 5 February 2009 (CTD/Niskin system on deck 4:54 GMT). At this site, the chlorophyll maximum was slightly deeper in the water column than at EQP-1 at the time of sampling (41 m), but in contrast to EQP-1, was above the depth of the pycnocline. The O2 minimum at this site was within the thermocline, with a concentration of 29 μmol kg–1 (Figure 2). In the sediment, dissolved oxygen penetrated ~10 cm below seafloor (Røy et al., 2012).

EQP-11 (30° 21.32’ N, 157° 52.23’ W) is in the North Pacific gyre. It is located on a very slight topographic high with basement age of 88.7 Ma (Müller et al., 2008). With water depth of 5813 m, this is our deepest water site. Total sediment thickness is estimated to be 100 m (Divins, 2003). We began water-column sampling on the 19th of February (CTD on deck 06:12 GMT). The deep chlorophyll maximum (DCM) was in relatively deep water (127 mbsl) and not strongly developed. The O2 minimum at this site (18 μmol kg–1) was also deep in the water column (900 mbsl). As at the other sites, this O2 minimum was located within the thermocline. Mean sedimentation rate is extremely low at EQP-11 (~1.1 m Ma–1). The sediment is pelagic clay and dissolved oxygen penetrates past the greatest depth cored (28.06 mbsf; Røy et al., 2012).

DNA extraction, pyrosequencing and statistical analyses

We applied a high-resolution v4–v6 tag-pyrosequencing approach of the bacterial 16S rRNA gene to a total of 39 water and sediment samples taken vertically from seasurface to subseafloor at our three sites (Figure 1, Table 1). We extracted DNA from filtered water samples using the Power Water DNA Isolation Kit (MoBio, Carlsbad, CA, USA) and from sediment samples using the Power Soil DNA Isolation Kit (MoBio). We modified the extraction protocol for the deep EQP-11 sediment samples (2 and 4 m) to overcome effects of low biomass and DNA-binding clay (Cai et al., 2006; Kallmeyer et al., 2012). These modifications were (i) pooling of 10x the sediment (10 vs 1 g) and (ii) use of extended PCR cycle numbers (40 × vs 30 × ). In addition, we used a phenol–chloroform DNA extraction for water samples from EQP-1. We used fusion primers 518 f and 1064r, targeting the v4–v6 hypervariable regions of the 16S rRNA gene, to construct amplicon libraries using the following PCR conditions: 94 °C for 2 min, 30 cycles of (94 °C for 30 s, 55 °C for 20 s, 72 °C for 1 min). We pooled three separate PCR reactions per sample to eliminate the potential for early cycle PCR-induced error. We ran negative controls during all PCR reactions. All sequenced reactions produced bands, whereas controls did not. We sequenced pooled amplicons using a FLX 454 pyrosequencer (Roche, Branford, CT, USA).

Table 1 Sampling information and diversity estimates for water and sediment collected from our three sites

We discarded low-quality reads, defined as (i) reads with undefined residues and (ii) reads that did not contain the PCR primers at the beginning of each read (Huse et al., 2007). We submitted our sequence data to the GenBank database under study accession no. SRP058974. We taxonomically assigned reads, based on 80% sequence similarity, using the GAST system for sequence identification (Sogin et al., 2006).

To ensure inter-sample comparability for our taxonomic richness estimates and statistical analyses, we utilized the QIIME software (Boulder, CO, USA) to randomly reduce the number of reads in each sample to the lowest number of reads in any individual sample (2880). We also used the QIIME program to calculate taxonomic richness (total operational taxonomic units (OTUs), Chao-1, ACE and phylogenetic diversity (PD); Caporaso et al., 2010).

We used the Primer-E software package (Clarke and Gorley, 2006) for (i) calculation of Bray–Curtis similarity indices and (ii) all ordination and statistical methods including non-metric multidimensional scaling, hierarchical group-average clustering and Spearman rank correlation tests. We used this output to construct figures using the ggplot2 package (Wickham, 2009) in R computer language (R Core Team, 2014).

To test the hypothesis that OTUs in subseafloor sediment are preferentially from the rare biosphere in the water column, rather than a random sample of water-column OTUs, we developed a Matlab simulation to randomly resample the water-column OTU frequency data with replacement. We repeated this sampling 5000 times to create a distribution of the number of OTUs randomly drawn from the rare (rank abundance >100) water-column OTUs. We then used this distribution to estimate the probability of randomly picking ‘n’ rare OTUs from the water-column abundance distribution.


Similarity among bacterial communities

Hierarchical ‘group-average’ clustering and non-metric multidimensional scaling analyses separate our samples into four distinct abundance-weighted community compositions, based on Bray–Curtis similarity (Figure 3a). Regardless of site, water-column samples partition into (i) photic-zone communities, which include samples from at or above the DCM, and (ii) aphotic communities (below the DCM). The GAST taxonomy demonstrated that Cyanobacteria, Flavobacteria and Alphaproteobacteria dominate the photic-zone samples, whereas the aphotic-zone samples are heavily dominated by Alphaproteobacteria, Gammaproteobacteria, Deltaproteobacteria and Deferribacteres (Figure 4). Regardless of site, aphotic-zone community composition exhibits a consistent relationship to water depth, with a clear compositional gradation from epipelagic communities below the DCM to abyssopelagic communities at depths between 3500 and 6000 mbsl (Figure 3b). The photic-zone samples cluster by site, whereas the aphotic-zone samples vary less from site to site. Water-column samples decline in similarity with increasing water depth regardless of site location (Figure 3b).

Figure 3

(a) Non-metric multidimensional scaling (nMDS) plot of the water column and sediment community compositions. Axes do not represent any measured parameter, but define a 2-D space that allows the best spatial representation of sample similarity, based on Bray–Curtis similarity indices. (b) Enlargement of the nMDS plot for the water-column communities with a superimposed color gradient representing depth from seasurface (red) to the abyssopelagic water of EQP-8 and EQP-11 (blue).

Figure 4

Heatmap clustering visualization of water column and sediment bacterial communities (x axis) and class-level GAST taxonomy (y axis). The heatmap is color coded by the percentage contributed by a given taxon to any given sample. Hierarchal clustering is based on the Morista–Horn similarity index.

We also observed two distinct groups in the sediment: a near seafloor (0–10 cmbsf) community and a subseafloor (0.25–34 mbsf) community. The common members of the shallow sedimentary community overlap with those of the deep-water community. The shallow sedimentary community is dominated by Alphaproteobacteria, Gammaproteobacteria and Deltaproteobacteria, as well as Nitrospira and Planctomycetes (Figure 4). The deep sedimentary community primarily comprises members of the obsidian pool (OP) candidate phylum, including OP9, OP8 and OP3, as well as Chloroflexi (Dehalococcoidetes and Anaerolineae), Actinobacteria and another phylum of unknown lineage (Figure 4).

The rank abundance and distribution of shared OTUs

To examine the extent to which OTUs are shared between the previously described communities (photic, aphotic, shallow sediment and deep sediment) (Figure 3a), we generated rank abundance histograms of the 100 most abundant OTUs (97% similarity) in each community. Figures 5a and c highlight the OTUs shared between the water column (photic and aphotic combined) and shallow sediment communities, between shallow and deep (1.5 mbsf) sediment communities, and between the deep sedimentary and water-column communities. Although our clustering results binned an EQP-1 sample from 0.25 mbsf with the deep sedimentary community (Figure 4), we excluded this sample from this shared-OTU analysis because it was taken close to the seafloor and we wish to ensure that only OTUs truly selected for subseafloor conditions are compared.

Figure 5

Rank abundance histograms of community overlap between (a) water column and shallow sediment (0–10 cmbsf), (b) shallow sediment and deep sediment (≥1.5 mbsl) and (c) water column and deep sediment (≥1.5 mbsl). OTUs highlighted in pink are shared between compared environments, whereas OTUs highlighted in blue are unique to each environment.

Many of the abundant OTUs in our shallow and deep sedimentary communities are present in the water at trace levels (<0.025% of total reads) (Table 2, Figures 5a and c). To quantify the statistical significance of so many shared OTUs being rare in the water column, we randomly resampled with replacement the water-column OTU frequency data until 24 unique OTUs were selected. These 24 randomly selected OTUs represent the 24 OTUs in the 100 most abundant deep-sediment OTUs that are shared with the water column (pink bars in lower panel of Figure 5c). Of these 24 shared OTUs, 22 are rare in the water column. Given the water-column abundance distribution and 5000 repetitions of our random resampling procedure, the probability of randomly selecting 22 of 24 OTUs from the pool of OTUs rarer than the top 100 in the water column is vanishingly small (P<0.0001).

Table 2 Shared OTUs in the 100 most abundant OTUs found in the pooled data from each environment (water column, shallow sediment and deep sediment)

These shared OTUs belong to Dehalococcoidetes, OP9, OP8 and other taxa. Furthermore, the OTUs that are abundant in the deep (subseafloor) sediment are also often abundant in the shallow sediment (Figure 5b). In contrast, most of the OTUs abundant in the water column are not present in our subseafloor sediment tags. The most abundant water-column OTUs that appear in our deep sedimentary tags include members of the Actinobacteria, Planctomycetes and Firmicutes (Figure 4). In short, the individual OTUs that are abundant in the deep sediment are also often abundant in the shallow sediment and often present but very rare (<0.025% of the tag population) in the overlying water column (Figure 6). The exceedingly small probability of getting so many rare water-column OTUs (22 of 24) by randomly sampling the water column (P<0.0001) indicates that water-column OTUs surviving in subseafloor sediment are preferentially from the rare community in the water column.

Figure 6

Rank abundance histograms for the top OTUs shared (x axis) between the oceanic and sedimentary environments. The y axis indicates the percentage of the total number of reads per OTU within our subseafloor sediment samples (1.5 mbsf, red), shallow sediment samples (orange) and water-column samples (blue).

Taxonomic richness, evenness and PD

Vertical trends in bacterial diversity (# of observed OTUs, Chao-1, ACE and PD) and evenness (H’) are similar at all three sites, with all measurements of richness and evenness peaking within the mesopelagic zone (Table 1). Overall, diversity is highest in the water column and lowest in the deep subseafloor sediment (Figure 2a). Within the water column, the vertical profile of bacterial diversity mirrors the O2 profile and matches the profile of apparent oxygen utilization at each site (Figures 2a and b, Table 1). Diversity in the water column is generally lowest in the surface ocean and highest at the O2 minima (Figures 2a and b). In the sediment, richness is highest near the seafloor (Figure 2a). Bacterial diversity in the water column varies from site to site; for example, richness is highest in the oxygen minimum zone of North Pacific gyre site EQP-11 and much lower in the oxygen minimum zone of eastern equatorial upwelling Site EQP-1 (Figure 2a).


Vertical trends in bacterial diversity

Water-column bacterial communities at all sites exhibit a clear depth profile, with samples declining in similarity from the seasurface to the abyssopelagic waters, in accordance with previously described trends (for example, DeLong et al., 2006; Brown et al., 2009; Treusch et al., 2009). Although changes in community composition with depth have previously been well documented, studies of taxonomic richness with depth have yielded mixed results. Some studies have reported declines in taxonomic richness and/or PD with increasing ocean depth (Brown et al., 2009; Agogué et al., 2011; Bryant et al., 2012), whereas others have observed higher richness deeper in the water column (for example, Pommier et al., 2010, Kembel et al., 2011; Ghiglione et al., 2012). By sampling the water column more comprehensively than previous studies, we find that bacterial richness (taxonomic richness, PD and Chao-1) and evenness in the water column is lowest at the surface and highest within the O2 minima of the mesopelagic zone at each of our sites. These results are consistent from site to site regardless of the diversity metric applied. Consequently, they provide clear evidence of vertical changes in bacterial richness across a range of environments. Some previous studies also reported elevated bacterial richness within the mesopelagic zone (Kembel et al., 2011; Jing et al., 2013; Ladau et al., 2013), but did not explicitly associate this result with the O2 minimum zone. The relationship we observe between oxygen and taxonomic richness suggests that respiration may have a significant role in shaping bacterial richness in the open ocean.

Only a few studies have examined bacterial diversity in both the water column and marine sediment (Feng et al., 2009; Zinger et al., 2011; Hamdan et al., 2013). None of these studies included sediment deeper than 1.1 mbsf. Most previous studies of seafloor bacterial communities agree that those communities are highly diverse (Ravenschlag et al., 1999; Madrid et al., 2001; Luna et al., 2004; Hewson et al., 2007; Schauer et al., 2010). Some studies have suggested that seafloor communities are taxonomically richer than communities in the water column (Lozupone and Knight, 2007; Feng et al., 2009; Zinger et al., 2011). In our study, taxonomic richness was generally higher in the water and lower in the sediment. This difference from previous studies may be partly due to the absence of subseafloor (>1 mbsf) sedimentary communities in those studies (>1 mbsf communities at our sites tend to be consistently less diverse than seafloor communities). It may also be due to the different oceanographic context of our studies relative to some previous studies (for example, Feng et al., 2009). It is possible that the open-ocean seafloor sediment is less taxonomically rich than seafloor sediment of continental margins and estuaries. This difference may also partly result from different sampling approaches. Our study and Feng et al. (2009) sample both seawater communities and sediment communities at each location, while Luzopone and Knight (2007) and Zinger et al. (2011) compared seawater communities from diverse locations with sediment communities from very different locations.

Within the sediment at all of our sites, taxonomic richness is highest at the seafloor and declines with increasing sediment depth (Figure 2). There is extensive taxonomic overlap between the communities of the deep sediment and those of the shallow sediment (Figure 5). These results suggest that communities in deep marine sediment are derived from a subset of the community that lived in the sediment when it was near the seafloor.

Taxonomic overlap between the ocean and sediment

Zinger et al. (2011) examined the bacterial beta-diversity of seafloor and seawater ecosystems and demonstrated that pelagic communities and epibenthic communities differed greatly at all taxonomic levels. Our study supports this observation and extends it to subseafloor sediment, as the abundance-weighted bacterial community compositions of the sediment are clearly distinct from those in the overlying water column (Figure 3).

To examine the extent of taxonomic overlap between the ocean communities and the sediment communities, we separately pooled the DNA sequences from the water samples of all three sites and the DNA sequences from the sediment samples of all three sites. We pooled the data for these comparisons, rather than doing the comparisons site by site, because sinking particles may be advected large distances, and in different directions at different water depths, before settling to the seafloor. Consequently, the material that settles to the seafloor rarely originates in the immediately overlying water. Although our three sites are separated by thousands of kilometers, spanning over 70 degrees of longitude and 30 degrees of latitude, we think it is justifiable and necessary to pool the samples because (i) intense sequencing at a single site has demonstrated that the marine bacterial rare biosphere is highly cosmopolitan (Gibbons et al., 2013) and (ii) pelagic communities from the same oceanographic horizon (the DCM) are highly similar for thousands of kilometers (Walsh et al., 2015).

When we consider the presence–absence of individual taxa, we observe overlap between environments. The most prominent example of this overlap is the presence in seawater and shallow sediment of many OTUs that are abundant in the deep-sediment samples (Figures 5 and 6). The overlap is very lopsided considering relative abundance; OTUs represented by abundant sequences in the sediment tag populations are often present but represented by rare sequences in the seawater tag population, whereas OTUs with abundant tags in the seawater population are rarely present in the sediment populations. This result is consistent with microbes being carried through the ocean and deposited in the sediment. The absence of known phototrophs (for example, Prochlorococcus, Synechococcus) in our subseafloor tag populations indicates that taxonomic overlap between the seawater and subseafloor sedimentary communities does not simply result from preservation of detritus. Instead, the subseafloor conditions appear to select for preferential survival of taxa that were represented by relatively few sequences in the seawater populations. The relatively low abundance of these sequences in the seawater populations is not surprising, as subseafloor sedimentary conditions are very different from water-column conditions; for example, anoxic conditions and low energy availability are much more ubiquitous in subseafloor sediment than in the overlying ocean (Jørgensen and D'Hondt, 2006). It is possible that microbes associated with the sedimentary environment are transported long distances through the ocean as components of the rare biosphere, from which they may or may not re-colonize the seafloor. Although resuspension and downslope sediment transport may be important sedimentary depositional processes in some marine environments, they are not significant processes at our sites; all three sites are located on flat abyssal regions or topographic highs and the cores from all three sites show no evidence of turbidites, indicating little to no redeposition of sediment. Consequently, advective transport through the water column provides the simplest explanation of microbial input to this sediment, consistent with discovery of thermophiles in marine sediment far from hydrothermal vents (Lee et al., 2005; Hubert et al., 2009), and with Hamdan et al.'s (2013) identification of hydrography as a forcing factor for OTU composition in near-seafloor marine sediment.

Microbial transport through the water column may explain the incidence of large-scale biogeographic patterns in both near-seafloor sediment (Hamdan et al., 2013) and subseafloor sediment. A previous study (Inagaki et al., 2006) observed that similar taxa (90% similarity, for example, Dehalococcoidetes, OP9 and Desulfobacteriales) are present in environmentally similar subseafloor sedimentary environments separated by large geographic distances. Our results show this similarity at a much finer taxonomic resolution (97% similarity) and indicate that these clades may dominate in a broad range of subseafloor sedimentary environments. Transport through the water column is far easier than through a sedimentary matrix where both advection and biologically accessible electron donors are limited.


Our study provides the first vertical profiles of microbial diversity from seasurface to subseafloor sediment (down to 34 mbsf). Bacterial richness is generally higher in the water column than in the sediment. It is highest in the O2-minimum zone of the water column and lowest in the subseafloor sediment. Despite the differences in abundance-weighted community compositions in the water column, the shallow sediment and the subseafloor sediment, the most abundant taxa in the subseafloor sedimentary communities are also often abundant in the shallow sedimentary communities and present but rare in the water column. This result suggests that deep subseafloor communities (i) constitute a subset of the diverse taxa that were present in the sediment at the time of burial and (ii) are ultimately seeded via the water column.

Accession codes




  1. Agogué H, Lamy D, Neal PR, Sogin ML, Herdnel GJ . (2011). Water mass-specificity of bacterial communities in the North Atlantic revealed by massively parallel sequencing. Mol Ecol 20: 258–274.

    Article  Google Scholar 

  2. Baas-Becking LGM . (1934) Geobiologie of Inleiding tot de Milieukund. WP Van Stockum and Zoom (in Dutch): : The Hague, The Netherlands.

    Google Scholar 

  3. Behrenfeld MJ, Falkowski PG . (1997). Photosynthetic rates derived from satellite-based chlorophyll concentration. Limnol Oceanogr 42: 1–20.

    CAS  Article  Google Scholar 

  4. Beman JM, Carolan MT . (2013). Deoxygenation alters bacterial diversity and community composition in the ocean’s largest oxygen minimum zone. Nat Commun 4: 2705.

    Article  Google Scholar 

  5. Brown MV, Philip GK, Bunge JA, Smith MC, Bissett A, Lauro FM et al. (2009). Microbial community structure in the North Pacific Ocean. ISME J 3: 1374–1386.

    CAS  Article  Google Scholar 

  6. Bryant JA, Stewart FJ, Eppley JM, DeLong EF . (2012). Microbial community phylogenetic and trait diversity declines with depth in a marine oxygen minimum zone. Ecology 93: 1659–1673.

    Article  Google Scholar 

  7. Cai P, Huang Q, Zhang X, Chen H . (2006). Adsorption of DNA on clay minerals and various colloidal particles from an Alfisol. Soil Biol Biochem 38: 471–476.

    CAS  Article  Google Scholar 

  8. Caporaso GJ, Kuczynski J, Stombaugh J, Bittinger K, Bushman FD, Costello EK et al. (2010). QIIME allows analysis of high-throughput community sequencing data. Nat Methods 7: 335–336.

    CAS  Article  Google Scholar 

  9. Clarke KR, Gorley RN . (2006). PRIMER v6: User Manual/Tutorial.: PRIMER-E, : Plymouth.

  10. D'Hondt S, Jørgensen BB, Miller DJ, Batzke A, Blake R, Cragg BA et al. (2004). Distributions of microbial activities in deep subseafloor sediments. Science 306: 2216–2221.

    CAS  Article  Google Scholar 

  11. D'Hondt S, Inagaki F, Zarikian CA, Abrams LJ, Dubois N, Engelhardt T et al. (2015). Presence of oxygen and aerobic communities from sea floor to basement in deep-sea sediments. Nature Geosci 8: 299–304.

    CAS  Article  Google Scholar 

  12. D'Hondt S, Spivack A, Pockalny R, Ferdleman T, Fisher J, Kallmeyer J et al. (2009). Subseafloor sedimentary life in the South Pacific Gyre. Proc Natl Acad Sci USA 106: 11651–11656.

    CAS  Article  Google Scholar 

  13. DeLong EF, Preston CM, Mincer T, Rich V, Hallam SJ, Frigaard NU et al. (2006). Community genomics among stratified microbial assemblages in the ocean's interior. Science 311: 496–503.

    CAS  Article  Google Scholar 

  14. Divins DL . (2003). Total Sediment Thickness of the World's Oceans & Marginal Seas, vol. Boulder, CO.

  15. Feng B-W, Li X-R, Wang J-H, Hu Z-Y, Meng H, Xiang L-Y et al. (2009). Bacterial diversity of water and sediment in the Changjiang estuary and coastal area of the East China Sea. FEMS Microbiol Ecol 70: 236–248.

    CAS  Article  Google Scholar 

  16. Fuhrman JA, Hewson I, Schwalbach MS, Steele JA, Brown MV, Naeem S . (2006). Annually reoccurring bacterial communities are predictable from ocean conditions. Proc Nat Acad Sci USA 103: 13104–13109.

    CAS  Article  Google Scholar 

  17. Ghiglione J-F, Galand PE, Pommier T, Pedrós-Alió C, Maas EW, Bakker K et al. (2012). Pole-to-pole biogeography of surface and deep marine bacterial communities. Proc Nat Acad Sci USA 109: 17633–17638.

    CAS  Article  Google Scholar 

  18. Gibbons SM, Caporosa JG, Pirrung M, Field D, Knight R, Gilbert JA . (2013). Evidence for a persistent microbial seed bank throughout the global ocean. Proc Nat Acad Sci USA 110: 4651–4655.

    CAS  Article  Google Scholar 

  19. Gregg WW, Casey NW, McClain CR . (2005). Recent trends in global ocean chlorophyll. Geophys Res Lett 32: L03606.

    Google Scholar 

  20. Hamdan LJ, Coffin RB, Sikaroodi M, Greinert J, Treude T, Gillevet PM . (2013). Ocean currents shape the microbiome of Arctic marine sediments. ISME J 7: 685–696.

    CAS  Article  Google Scholar 

  21. Hewson I, Jacobson/Meyers ME, Fuhrman JA . (2007). Diversity and biogeography of bacterial assemblages in surface sediments across the San Pedro Basin, Southern California Borderlands. Environ Microbiol 9: 923–933.

    CAS  Article  Google Scholar 

  22. Hubert C, Loy A, Nickel M, Arnosti C, Baranyi C, Brüchert V et al. (2009). A constant flux of diverse thermophilic bacteria into the Cold Arctic Seabed. Science 325: 1541–1544.

    CAS  Article  Google Scholar 

  23. Huse SM, Huber JA, Morrison HG, Sogin ML, Welch DM . (2007). Accuracy and quality of massively parallel DNA pyrosequencing. Genome Biol 8: R143.

    Article  Google Scholar 

  24. Inagaki F, Nunoura T, Nakagawa S, Teske A, Lever M, Lauer A et al. (2006). Biogeographical distribution and diversity of microbes in methane hydrate-bearing deep marine sediments on the Pacific Ocean Margin. Proc Nat Acad Sci USA 103: 2815–2820.

    CAS  Article  Google Scholar 

  25. Jing H, Xia X, Suzuki K, Liu H . (2013). Vertical profiles of bacteria in the tropical and subarctic oceans revealed by pyrosequencing. PLoS One 8: e79423.

    Article  Google Scholar 

  26. Jørgensen BB, D'Hondt S . (2006). A starving majority deep beneath the seafloor. Science 314: 932–934.

    Article  Google Scholar 

  27. Kallmeyer J, Pockalny R, Adhikari RR, Smith DC, D'Hondt S . (2012). Global distribution of microbial abundance and biomass in subseafloor sediment. Proc Nat Acad Sci USA 109: 16213–16216.

    CAS  Article  Google Scholar 

  28. Kembel SW, Eisen JA, Pollard KS, Green JL . (2011). The phylogenetic diversity of metagenomes. PLoS One 6: e23214.

    CAS  Article  Google Scholar 

  29. Ladau J, Sharpton TJ, Finucane MM, Jospin G, Kembel SW, O'Dwyer J et al. (2013). Global marine bacterial diversity peaks at high latitudes in winter. ISME J 7: 1669–1677.

    CAS  Article  Google Scholar 

  30. Lee Y-J, Wagner ID, Brice ME, Kevbrin VV, Mills GL, Romanek CS, Wiegel J . (2005). Thermosediminibacter oceani gen. nov., sp. nov. and Thermosediminibacter litoriperuensis sp. nov., new anaerobic thermophilic bacteria isolated from Peru Margin. Extremophiles 9: 375–383.

    CAS  Article  Google Scholar 

  31. Lozupone CA, Knight R . (2007). Global patterns in bacterial diversity. Proc Nat Acad Sci USA 104: 11436–11440.

    CAS  Article  Google Scholar 

  32. Luna GM, Dell’Anno A, Giuliano L, Danovaro R . (2004). Bacterial diversity in deep Mediterranean sediments: relationship with the active bacterial fraction and substrate availability. Environ Microbiol 6: 745–753.

    CAS  Article  Google Scholar 

  33. Madrid VM, Aller JY, Aller RC, Chistoserdov AY . (2001). High prokaryote diversity and analysis of community structure in mobile mud deposits off French Guiana: identification of two new bacterial candidate divisions. FEMS Microbiol Ecol 37: 197–209.

    CAS  Article  Google Scholar 

  34. Martiny JBH, Bohannan BJM, Brown JH, Colwell RK, Fuhrman JA, Green JL et al. (2006). Microbial biogeography: putting microorganisms on the map. Nat Rev Micro 4: 102–112.

    CAS  Article  Google Scholar 

  35. Müller RD, Sdrolias M, Gaina C, Roest WR . (2008). Age, spreading rates, and spreading asymmetry of the world's ocean crust. Geochem Geophys Geosyst 9: Q04006.

    Article  Google Scholar 

  36. Pommier T, Neal PR, Gasol JM, Coll M, Acinas SG, Pedrós-Alió C . (2010). Spatial patterns of bacterial richness and evenness in the NW Mediterranean Sea explored by pyrosequencing of the 16S rRNA. Aquat Microb Ecol 61: 221–233.

    Article  Google Scholar 

  37. R Core Team. (2014) R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing: Vienna, Austria URL

  38. Ravenschlag K, Sahm K, Pernthaler J, Amann R . (1999). High bacterial diversity in permanently cold marine sediments. Appl Environ Microbiol 65: 3982–3989.

    CAS  PubMed  PubMed Central  Google Scholar 

  39. Røy H, Kallmeyer J, Adhikari RR, Pockalny R, Jørgensen BB, D’Hondt S . (2012). Aerobic microbial respiration in 86-million year old deep-sea red clay. Science 336: 922–925.

    Article  Google Scholar 

  40. Schauer R, Bienhold C, Ramette A, Harder J . (2010). Bacterial diversity and biogeography in deep-sea surface sediments of the South Atlantic Ocean. ISME J 4: 159–170.

    CAS  Article  Google Scholar 

  41. Sogin M, Morrison H, Welch D, Huse S, Neal P, Arrieta J et al. (2006). Microbial diversity in the deep-sea and the unexplored 'rare Biosphere'. Proc Nat Acad Sci USA 103: 12115–12120.

    CAS  Article  Google Scholar 

  42. Treusch AH, Vergin KL, Finlay LA, Donatz MG, Burton RM, Carlson CA et al. (2009). Seasonality and vertical structure of microbial communities in an ocean gyre. ISME J 3: 1148–1163.

    Article  Google Scholar 

  43. Walsh EA, Smith DC, Sogin ML, D’Hondt SL . (2015). Bacterial and archaeal biogeography of the deep chlorophyll maximum in the South Pacific Gyre. Aquatic Microbial Ecology 75: 1–13. doi:10.3354/ame01746.

    Article  Google Scholar 

  44. Wickham H . (2009) ggplot2: Elegant Graphics for Data Analysis. Springer: : New York.

    Book  Google Scholar 

  45. Zinger L, Amaral-Zettler LA, Furhman JA, Horner-Devine MC, Huse SM, Welch DBM et al. (2011). Global patterns of bacterial beta-diversity in seafloor and seawater ecosystems. PLoS One 6: e24570.

    CAS  Article  Google Scholar 

Download references


We thank Captain Kent Sheasley, and the crew and scientific party of R/V Knorr Expedition 195-3 for our samples. We also thank Robert Pockalny for mapping and helping to select the Expedition 195-3 sampling sites. This study was funded by the Biological Oceanography Program of the US National Science Foundation (grant OCE-0752336) and by the NSF-funded Center for Dark Energy Biosphere Investigations (grant NSF-OCE-0939564). This is C-DEBI publication 269.

Author information



Corresponding author

Correspondence to Emily A Walsh.

Ethics declarations

Competing interests

The authors declare no conflict of interest.

Rights and permissions

This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in the credit line; if the material is not included under the Creative Commons license, users will need to obtain permission from the license holder to reproduce the material. To view a copy of this license, visit

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Walsh, E., Kirkpatrick, J., Rutherford, S. et al. Bacterial diversity and community composition from seasurface to subseafloor. ISME J 10, 979–989 (2016).

Download citation

Further reading


Quick links