0MOLECULAR ECOLOGY
Molecular Ecology (2015) 24, 4931-4942 doi: 10.1111/mec.13365
Macroecological patterns of archaeal ammonia oxidizers in the Atlantic Ocean
EVA SINTES,* DANIELE DE CORTE,* NATASCHA OUILLON* and GERHARD J. HERNDL*j *Department of Limnology and Bio-Oceanography, Center of Ecology, University of Vienna, Althanstrasse 14,1090 Vienna, Austria, ^Department of Biological Oceanography, Royal Netherlands Institute for Sea Research, PO Box 59,1790 Den Burg, The Netherlands
Abstract
Macroecological patterns are found in animals and plants, but also in micro-organisms. Macroecological and biogeographic distribution patterns in marine Archaea, however, have not been studied yet. Ammonia-oxidizing Archaea (AOA) show a bipolar distribution (i.e. similar communities in the northernmost and the southernmost locations, separated by distinct communities in the tropical and gyral regions) throughout the Atlantic, detectable from epipelagic to upper bathypelagic layers (<2000 m depth). This tentatively suggests an influence of the epipelagic conditions of organic matter production on bathypelagic AOA communities. The AOA communities below 2000 m depth showed a less pronounced biogeographic distribution pattern than the upper 2000 m water column. Overall, AOA in the surface and deep Atlantic waters exhibit distance-decay relationships and follow the Rapoport rule in a similar way as bacterial communities and macroorganisms. This indicates a major role of environmental conditions in shaping the community composition and assembly (species sorting) and no, or only weak limits for dispersal in the oceanic thaumarchaeal communities. However, there is indication of a different strength of these relationships between AOA and Bacteria, linked to the intrinsic differences between these two domains.
Keywords: ammonia oxidizers, biogeography, deep ocean, macroecology, Thaumarchaeota
Received 21 October 2014; revision received 29 July 2015; accepted 21 August 2015
Introduction
The application of ecological theories to microbial organisms has gain renewed interest in the last decade (Martiny et al. 2006; Ramette & Tiedje 2007; Fierer 2008; Amend et al. 2013; van der Gast 2015). Although there are some singularities of the microbial ecology patterns due to differences in scale and physiology between macro- and micro-organisms (Fierer et al. 2011; Carbonero et al. 2014), many macroecological patterns, yet quantitatively different, also extend to Bacteria (Soini-nen 2012; Amend et al. 2013). Patterns found in macroorganisms, such as taxa-area relationships and distance-decay patterns (Green et al. 2004; Horner-Devine et al. 2004; Bell 2010; Astorga et al. 2012; Wetzel et al. 2012; Zinger et al. 2014), latitudinal species
Correspondence: Eva Sintes, Fax: +43 14277876401; E-mail: eva.sintes@univie.ac.at
richness gradients (Pommier et al. 2007; Fuhrman et al. 2008) and the Rapoport rule (Amend et al. 2013; Sul et al. 2013) have been also reported for microorganisms.
In this context, the existence of biogeographic patterns for microbes is now widely accepted (Martiny et al. 2006; Ramette & Tiedje 2007; Fierer 2008). However, there is an intense debate on the underlying factors for these patterns, both for macro- and microorganisms. Are the observed patterns generated via selective processes, such as evolutionary adaptation and sorting of species according to the environmental conditions, or nonselective as proposed by neutral theory, such as dispersal and drift (Vellend et al. 2014)? This debate in microbial biogeography has been centred in the last decades on the Baas-Becking principle 'everything is everywhere, but, the environment selects' (Baas-Becking 1934). Recent research argues that the generalizing ubiquitous dispersal hypothesis
© 2015 The Authors. Molecular Ecology Published by John Wiley & Sons Ltd.
This is an open access article under the terms of the Creative Commons Attribution License,
which permits use, distribution and reproduction in any medium, provided the original work is properly cited.
highlighted by this tenet should be rejected (van der Gast 2015) and the relative influence of selective (e.g. niche suitability) and nonselective (stochastic, historical) processes should be considered and disentangled (Barberan et al. 2014).
Although the dark ocean harbours 75% of the prokaryotic biomass (Aristegui et al. 2009), macroeco-logical patterns of microbes, including biogeographic studies, have focused mostly on Protozoa (Finlay & Fenchel 2004; Foissner 2006) and epipelagic bacterial communities (Pommier et al. 2007; Sul et al. 2013) with few exceptions (Ghiglione et al. 2012). Generally, only limited information is available on the archaeal communities inhabiting the deep ocean. There are intrinsic differences between bacterial and archaeal domains such as their cell wall, metabolic pathways, molecular repertoire and diversity (DeLong & Pace 2001; Verhees et al. 2003; Wang et al. 2007; Ellen et al. 2010; Albers & Meyer 2011; Yarza et al. 2014) that could influence both selective processes and dispersal. Only recently, however, archaeal biogeography has been studied across different environments (from symbionts to terrestrial and aquatic habitats) based on 16S rRNA (Auguet et al. 2010) and amoA sequences (Fernandez-Guerra & Casamayor 2012; Cao et al. 2013), or focusing on aquatic ecosystems (Biller et al. 2012) or soils (Pester et al. 2012). These studies focused on differences between habitats. Thus, macroecological patterns, such as distance-decay or latitudinal range, were not inferred for Archaea in contrast to studies on Bacteria (Sul et al. 2013).
The goal of this study was to provide a comprehensive view on the macroecological patterns of ammonia-oxidizing Archaea (AOA) along a latitudinal gradient and throughout the water column of the Atlantic Ocean. Our hypothesis was that the intrinsic lower diversity of Archaea, in particular Thaumarchaeota, together with other intrinsic differences such as differential dispersal rates associated to different population size, etc., results in different latitudinal and depth-related patterns in Archaea as compared to Bacteria and Eukarya.
Materials and methods
Sampling was conducted during the GEOTRACES-1 and -2 cruises with R/V Pelagia, between April and June 2010 and GEOTRACES-3 onboard of R/V James Cook between February and April 2011. Water samples were taken at 51 stations (Fig. S1, Supporting information) with 24 x 25L-Niskin bottles mounted in a frame holding also sensors for conductivity-temperature-depth (CTD), salinity, oxygen, fluorescence and optical backscattering. Contextual environmental parameters, such as inorganic nutrients and microbial abundance, were analysed at 24 depth layers from surface to
abyssopelagic waters as detailed in Appendix S1 (Supporting information). Samples for the analyses of AOA were collected at 6-8 depths from the euphotic layer (50 m depth), the oxygen minimum zone and mesopelagic (<1000 m), upper bathypelagic (10002000 m) and lower bathypelagic and abyssopelagic depths (>2000 m). Quantitative PCR (Q-PCR) was used with slight modifications to evaluate 16S rRNA gene abundance of Thaumarchaeota and the abundance of two ecotypes of AOA based on their amoA gene: the low ammonia concentration ecotype (LAC-AOA) and the high ammonia concentration ecotype (HAC-AOA) as previously described (Sintes et al. 2013) (Appendix S1, Supporting information). Six different oceanographic regions were differentiated along this transect based on the description of Longhurst (2007): the North Atlantic Arctic province (ARCT; 70°N-55°N), the North Atlantic Drift province (NADR; 55°N-40°N), the North Atlantic Gyral province (NAG) comprising the North Atlantic Tropical and the Subtropical Gyral province (40°N-12°N), the Western Tropical Atlantic (WTRA; 12°N-6°S) province, the South Atlantic Gyral (SATL; 6°S-40°S) and the Subantarctic province (SANT) comprising the Subtropical Convergence Zone (SSCT; 40°S-45°S) and the Subantarctic Water Ring province (SANT; 45°S-55°S) (Fig. S1, Supporting information). GPS coordinates and sampling date for each station are available on Dryad (doi: 10.5061/dryad.c6688).
To evaluate the community composition of AOA, terminal-restriction fragment length polymorphism (T-RFLP) analysis of the archaeal amoA gene was conducted on all the samples after its amplification using specific primers as described in the Appendix S1 (Supporting information). Briefly, labelled primers cren amo_F-FAM (Hallam et al. 2006) and amoAR-VIC (Francis et al. 2005) were used to amplify archaeal amoA. FAM- and VIC-labelled purified PCR products were digested at 37 °C overnight with three different restriction enzymes (Mbol, Rsal and Haelll). After separation of the labelled fragments with a 3130xL Genetic Analyzer capillary sequencer (Applied Biosystems), the electropherograms were analysed with GelComparII software (Bio-Rad Laboratories). The threshold level to discriminate bands was set at 0.5% of the total peak height. The obtained matrix was analysed by Primer software (Primer-E, Ltd, Ivybridge, UK) to determine the similarity between the different T-RFLP patterns obtained from the samples. Standardized and normalized OTU abundance data obtained by T-RFLP for the different samples are available on Dryad (doi: 10.5061/ dryad.c6688).
Additionally, one sample from each oceanic province and depth layer (Table S1, Fig. S1, Supporting information) was chosen to generate amoA gene sequence
libraries with Sanger sequencing and/or 454-pyrose-quencing (Appendix S1, Supporting information). 454-pyrosequencing of archaeal amoA was performed at IMGM Laboratories GmbH (Germany) on a Roche 454 GS Junior platform based on titanium chemistry. All samples were barcoded using multiplex identifiers and sequenced together in one run. Raw 454 sequences were initially trimmed using LUCY 1.20 (Chou & Holmes 2001) keeping sequences of >250 nt which had an average Phred score of >27. Subsequently, the remaining sequences were screened for the barcode and primer sequences keeping only the sequences that had exact matches.
The sequences selected by the above procedure were processed following a similar pipeline as described elsewhere (Pester et al. 2012). Briefly, sequences were preclustered using the pre.cluster function in MOTHUR (Schloss et al. 2009) with n = 3 (sequence identity >97.6% for sequences >250 nt). Representatives of the pre.cluster step were further grouped using the CD-HIT-454 (http://weizhong-lab.ucsd.edu/cd-hit/servers. php) clustering tool (Huang et al. 2010) at a 98.5% sequence identity level over 97% of the smaller sequence. Thereafter, HMMFrame (Zhang & Sun 2011) was used to screen possible frame shifts in representative sequences of all CD-HIT clusters.
After manual chimera removal (Pester et al. 2012), sequences were grouped based on their sequencing direction (forward or reverse) and rarefaction curves, binning into OTUs, and a-diversity analysis was conducted using MOTHUR (Schloss et al. 2009). The remaining sequences were aligned together with the clone sequences and NCBI reference sequences from N. maritimus, Nitrososphaera gargensis, Nitrosoarchaeum limnia and Cenarchaeum symbiosum to infer their phy-logeny. Raw 454-pyrosequences of amoA have been deposited in NCBI, Accession no. SRP049002. A detailed description of the pyrosequencing approach can be found in the Appendix S1 (Supporting information). AOA OTUs abundance data obtained by pyrose-quencing are available on Dryad (doi:10.5061/ dryad.c6688).
AOA OTUs were assigned as the gene sequences sharing 98% identity (Agogue et al. 2008). Although this sequence similarity of amoA gene might not correspond to different archaeal species (Pester et al. 2012), it was chosen to keep the information on the diversity of the functional group. Although this selection might increase the diversity values obtained for Thaumarchaeota, it will facilitate observing patterns in the functional group of AOA.
T-RFLP fingerprinting, Sanger sequencing and 454-py-rosequencing analyses targeted the entire archaeal amoA harbouring community. T-RFLP allowed a relatively fast
analysis at high spatial resolution (in total, 295 samples were efficiently fingerprinted with this method), while 454-pyrosequencing was used to attain an in-depth phy-logenetic assessment of the AOA community, including the low abundance groups. Additionally, the Sanger sequencing provided a high quality database of archaeal amoA sequences from the same stations where pyrose-quencing was conducted, facilitating the identification of chimeras and the alignment of the pyrosequences. The role of environmental (temperature, depth, nitrite, oxygen concentration, latitude) and biological (Thaumar-chaeota 16S rRNA, LAC- to HAC-amoA ecotype abundance) factors on the AOA community composition (assessed either by T-RLP or 454 pyrosequencing) was evaluated by canonical correspondence analysis (CCA). Partial redundancy analysis (partial RDA) was used to discriminate the contribution of different explanatory variables (environment, space and time) on the variation of the AOA community composition in the Atlantic (Appendix S1, Supporting information).
Results
Diversity of archaeal ammonia oxidizers: latitudinal and depth patterns
Rarefaction curves (Fig. S2, Supporting information) obtained for Sanger sequencing and 454-pyrosequencing libraries indicated that the epipelagic AOA communities exhibited a lower richness than meso- and bathy-pelagic AOA communities of low latitude provinces (Table S2, Supporting information). A trend of increasing richness (Table S2, Supporting information) and diversity (Fig. 1A, Table S2, Supporting information) towards the gyral and tropical regions was evident from the mesopelagic libraries, while the highest richness and diversity of upper bathypelagic AOA communities were found in the NAG province. Richness and diversity of lower bathypelagic (below 2000 m depth) AOA communities were less variable than in the upper bathypelagic realm (1000-2000 m depth) and were highest in the ARCT, WTRA and SANT provinces (Table S2, Figs 1A and S2, Supporting information). Slight differences (compare Figs 1A and B) in the location of the diversity maximums between pyrosequencing and T-RFLP fingerprinting approaches (i.e. maximum for meso- and lower bathypelagic communities at the equator and around 30°N, respectively) might be caused by the different detection levels of the two methods.
T-RFLP fingerprinting of the AOA revealed distinct communities associated to the oceanographic provinces and depth layers (two-way ANOSIM, R = 0.35 and R = 0.42, respectively, P = 0.001). The Shannon index of diversity obtained from T-RFLP fingerprints was
(A) 7.0
® 6.° c
§ 5.5 c c CO
-c 5.0
4.5 4.0
O Meso O UpperBathy 0 LowerBathy
800N 60' N 400N 200N 0' Latitude
200S 400S 600S
20"N 00
Latitude
Fig. 1 Diversity (Shannon index) of AOA communities in the mesopelagic (Meso), upper (UpperBathy) and lower bathypelagic layer (LowerBathy) in different regions obtained from the 454-pyrosequencing libraries (A), and based on the depth profiles using T-RFLP fingerprinting (B).
40 0 N
generally higher for mesopelagic and upper bathypelagic waters (Fig. 1B) than for lower bathypelagic layers, in agreement with results obtained by cloning and pyrosequencing (Table S2, Supporting information, Fig. 1A). The highest diversity indexes were obtained for the oligotrophic gyral provinces (Fig. 1B).
Considering the communities of different depth layers separately (Fig. S3, Supporting information), AOA communities assessed by T-RFLP fingerprinting grouped according to oceanic provinces, especially in epi- and mesopelagic waters (Fig. S3A and B, Supporting information, respectively). AOA from the ARCT, NADR and SANT provinces tended to cluster together (Fig. S3, Supporting information) and shared a higher number of OTUs among them as assessed by T-RFLP and 454 pyrosequencing than with gyral and tropical provinces (Table S3, Supporting information, Fig. 2A, B). In contrast, AOA communities of the NAG, WTRA and SATL provinces shared a higher number of AOA OTUs (Table S3, Supporting information, Fig. 2C, D) than with high latitude AOA communities. This trend resulted in a decrease in similarity between AOA communities (determined by T-RFLP fingerprinting) from ~50°S to 0°, followed by an increase in similarity towards 64°N (Fig. 3). This latitudinal trend in similarity of AOA communities was more pronounced in epipelagic waters (r = 0.80, P < 0.0001, Fig. 3), than for the mesopelagic (r = 0.68, P < 0.0001) and upper bathypelagic AOA communities (r = 0.66, P < 0.0001). The similarity between AOA communities was higher and less variable in upper bathypelagic waters than in meso- and epipelagic layers (Fig. 3). In contrast to the AOA communities of the upper water layers, the AOA communities in the lower bathypelagic layer showed no significant relation between similarity in community composition and distance (r = 0.41, P > 0.4).
The latitudinal tendency in the similarity of the AOA communities was also apparent from the archaeal amoA pyrosequencing libraries, exhibiting a higher similarity
between ARCT and SANT communities and between NAG and SATL communities (Fig. 4) than to the AOA communities of other provinces, even though they are farther apart than, for example ARCT from NAG, or NAG from WTRA (Fig. 4).
In regions of deep-water mass formation or upwelling (ARCT, WTRA and SANT, Fig. S4C, Supporting information), the different depth layers shared a higher number of OTUs (Fig. S4A, B, D, Supporting information) than in the stratified water column of the NADR and the gyral provinces (Fig. S4A, B, D, Supporting information).
Latitudinal and temperature range of individual members of the AOA community
As revealed by both T-RFLP and pyrosequencing, AOA OTUs from all depth layers and both hemispheres followed the Rapoport rule, that is OTUs from high latitudes exhibited a broader latitudinal range than OTUs from low latitudes (Fig. 5), except the lower bathypelagic OTUs obtained by pyrosequencing (>2000 m depth, Fig. 5C, D). Our results also indicate a broader latitudinal range of abundant OTUs than of less abundant OTUs (Table S4, Supporting information). In terms of temperature range of AOA OTUs, only epipelagic OTUs and surprisingly, upper bathypelagic (10002000 m depth) AOA OTUs exhibited a significant temperature range that mimicked the latitudinal range (Fig. S5, Supporting information).
Environmental factors influencing the distribution of AOA
The role of environmental factors on the AOA community composition was assessed by CCA. The first two coordinates explained 58.1% and 56.3% of the variation in the T-RFLP fingerprints obtained from the forward and the reverse regions, respectively (Figs 6A and S6A, Supporting information). Coordinate 1 was positively
(A) 30
rn D I—
(B) 50
ARCT NADR NAG WTRA SATL SANT
ARCT NADR NAG WTRA SATL SANT
rn D I—
T3 £ .c
ARCT NADR
NAG WTRA SATL SANT Region
ARCT NADR
NAG WTRA SATL SANT Region
Fig. 2 Proportion of shared OTUs of the ARCT region with all the other regions: for the whole water column (A) and for mesopela-gic (Meso), upper (UB) and lower (LB) bathypelagic layers (B) separately. The proportion of OTUs from the WTRA region shared with other regions in the Atlantic for the entire water column (C) and for distinct water layers (D).
related to the ratio LAC-amoA to HAC-amoA and negatively to latitude and nitrite concentration, while coordinate 2 was positively related to temperature, salinity and abundance of Thaumarchaeota 16S rRNA genes and negatively to depth and, to a lesser extent, to oxygen concentration.
Using amoA pyrosequence data, 49.8% and 47.6% of the variation was explained by the first 2 coordinates for the forward- and reverse-sequenced amoA, respectively (Figs 6B and S6B, Supporting information). Coordinate 1 was strongly negatively related to temperature, salinity and abundance of Thaumarchaeota (MCGI) 16S rRNA gene and positively to oxygen concentration for the for-ward-sequenced amoA (Fig. 6B). The AOA communities from the oxygen minimum zone (250 m depth) are separated from the communities from other depth layers by coordinate 1 except the AOA community of the ARCT
and SANT province. Coordinate 2 was positively related to depth and the ratio between low ammonia concentration (LAC-) amoA ecotype and high ammonia concentration (HAC-) amoA ecotype abundance and negatively to latitude and nitrite concentration. The AOA communities from ARCT and SANT are separated from the other regions by coordinate 2. Similar relationships, with reversed axes, were observed for the reverse-sequenced amoA (Fig. S6B, Supporting information).
Discussion
Factors driving the distribution and diversity of AOA throughout the Atlantic
Several environmental variables have been identified as potentially important in regulating AOA community
• Epi a Meso ▼ UpperBathy ■ LowerBathy
Il % ■
• » ■
5000 10 000 15 000 20
Distance (km)
Fig. 3 Pairwise similarity between AOA community fingerprints vs. distance (km) for the epipelagic (Epi), mesopelagic (Meso), upper (UpperBathy) and lower bathypelagic (LowerBathy) throughout the Atlantic Ocean, all related to the southernmost station. Each point represents the Bray-Curtis similarity index between the community composition from a specific depth layer in the southernmost station and the community composition at the same depth layer from a station located at a distance x.
J5 ü x 0 c o
A Meso ▼ UpperBathy ■ LowerBathy
A WTRA
NAG*.-
NADR aaSAN ARCT
SATL a NADR
-0.25 -0.20 -0.15 -0.10 -0.05 0.00 0.05 0.10 0.15 0.20 P1 - Percent variation explained 58.80%
Fig. 4 Principal coordinates analysis of the 18 forward-pyrose-quenced archaeal amoA libraries. Different colours denote depth layers (Meso: mesopelagic, UpperBathy: upper bathypelagic, LowerBathy: lower bathypelagic), oceanographic regions are noted next to the symbols, for abbreviations see legend of Fig. S1 (Supporting information).
composition and distribution (Erguder et al. 2009), such as salinity (Francis et al. 2005; Mosier & Francis 2008; Sahan & Muyzer 2008; Abell et al. 2010), temperature
(Biller et al. 2012), pH (Pester et al. 2012) in soils, nitrite (Herfort et al. 2007), dissolved oxygen (Santoro et al. 2008), light (Merbt et al. 2012), latitude (Biller et al. 2012; Pester et al. 2012) or depth (Biller et al. 2012), leading to the habitat-phylogeny association of the microbial ammonia oxidizers (Fernández-Guerra & Casamayor 2012). Our data indicate a strong influence of latitude and depth on the AOA community composition (Figs 6 and S6, Supporting information), but the underlying factors for this influence include nitrite concentration, temperature and dissolved oxygen concentration. Although it is not clear how temperature and oxygen concentration might affect AOA, it has been shown that temperature relates to archaeal amoA community composition (Biller et al. 2012). Also, Thaumarchaeota are frequently associated to oxygen minimum zones (Agogue et al. 2008; Beman et al. 2008), where ammonia is released by remineralization processes (Wuchter et al. 2006). This suggests an indirect effect of dissolved oxygen concentration on AOA community composition. Ammonium concentration has been suggested to determine the abundance and oxidation rates not only of ammonia-oxidizing Bacteria (AOB), but also of AOA (Christman et al. 2011). Although we have no ammonium measurements, the end product of ammonia oxidation (NO2 ), and the ratio between the two ecotypes of AOA, which is linked to ammonia and nitrite concentrations (Sintes et al. 2013), strongly relates to the AOA community composition (Figs 6 and S6, Supporting information), supporting the significance of this nutrient in shaping the ammonia-oxidizing community.
The variation of the AOA community was explained not only by environmental factors, but also by spatial and temporal factors, explaining altogether 44% as estimated by RDA (Appendix S1, Supporting information). Variation partitioning supported the main role of the spatial and environmental factors on the community composition (Fig. S7, Supporting information) as compared to the possible temporal effect associated to the different sampling times.
Latitudinal ranges of archaeal ammonia oxidizers
AOA OTUs from high latitudes exhibit a wider latitudinal range than OTUs from low latitudes at all depth layers, except the lower bathypelagic OTUs determined by 454-pyrosequencing (Fig. 5). This finding is in agreement with studies on bacterial taxa (Amend et al. 2013; Sul et al. 2013). The broader range in AOA diversity at high than at low latitudes follows the Rapoport rule (Stevens 1989), found in plants and animals though with many exceptions (Rohde et al. 1993; Gaston et al. 1998; Rohde 1999). Also, abundant OTUs exhibit a broader latitudinal range as compared to less abundant
(A) 120
0.43, P < 0.005 Meso S = 0.26, rz = 0.14, P = 0.08
- UB S = 0.27, r -
- LB S = 0.41, r2 =
20 30 40 Absolute latitude (°
0.20, P < 0.05 0.34, P < 0.001
(B) 120
Epi S = 0.38, r2 = 0.38, P < 0.01 Meso S = 0.27, r2 = 0.66, P < 0.001
UB S LB S =
0.34, r2 0.17, r2--
■ 0.67, P < 0.001 0.52, P < 0.001
10 20 30 40 50 Absolute latitude (°)
Meso S = 1.17, r2 = 0.997, P = 0.03 UB S = 0.37, r 2 = 0.91, P = 0.20 LB S = -0.17x, r2 = 0.12, P = 0.78
10 20 30 40 50 Absolute latitude (°)
e ct c TO
T5 40 H di tu
q 20 -
Meso S = 0.39x, r2 = 0.86, P = 0.24 UB S = 0.45, r 2 = 0.96, P = 0.12 LB S = 0.03, r 2= 0.02, P = 0.91
10 20 30 40 Absolute latitude (°
Fig. 5 Relationship between AOA average latitudinal range size and latitude. Each symbol represents the average latitudinal range for a sample, based on T-RFLP OTUs for the (A) northern hemisphere, (B) southern hemisphere of the Atlantic, and based on 454 pyrosequences for the northern (C) and southern (D) hemispheres. Different symbols denote the different depth layers: epipelagic (Epi), mesopelagic (Meso), upper (UB) and lower (LB) bathypelagic.
OTUs (Sul et al. 2013). The mesopelagic AOA communities, also following the Rapoport rule, do not show a relation between temperature range and temperature (Fig. S5, Supporting information). This suggests that both latitudinal distance (as an indicator of dispersal limitation) and current environmental conditions (an indicator of species sorting) determine mesopelagic archaeal biogeography (Sul et al. 2013).
Bipolar distribution of archaeal ammonia oxidizers in the ocean
A distinct latitudinal gradient in alpha-diversity of the AOA community was detectable (Fig. 1), with higher diversity in low latitude than in high latitude regions, similar to the diversity pattern reported for Bacteria (Pommier et al. 2007; Fuhrman et al. 2008; Sul et al. 2013) and macroorganisms for terrestrial and marine ecosystems (Hillebrand 2004). The slope of latitudinal
diversity in the southern versus the northern hemisphere for the epi-, meso- and upper bathypelagic waters was not significantly different (Table S5, Supporting information, ANCOVA, P > 0.9) indicating a bipolar distribution of the AOA communities in these water layers (Sul et al. 2013). Although we do not have samples collected at strictly 'polar' but subpolar locations (64°N-40°N, 55°S-40°S), the trend observed corresponds to what is usually termed 'bipolar' distribution (i.e. high latitudes present similar communities and differ from low latitudinal communities). Additional support for the bipolar distribution is provided by the higher amount and percentage of shared OTUs between regions with similar environmental conditions, that is within high and within low latitude regions, than between high and low latitude regions (Fig. 2, Table S3, Supporting information).
The bipolar distribution of AOA extending from the epi- to the upper bathypelagic realm is further
ÇT Mr t' m. \
250 250 j 250 " ^MCGm^w 4500 ■ 3500 ■ 4500 4000 ■ • y
250 1250 250 * VST] .°XV 2500 ^ 1250 2500' ' 9 1250 1250 250
• • ■ ■ • • ■
Fig. 6 Canonical correspondence analysis (CCA): for (A) T-RFLP fingerprints of the forward region, and (B) 454-pyrosequencing libraries sequenced with the forward primer. Symbols represent different oceanic provinces, for abbreviations see legend of Fig. S1 (Supporting information). Next to the symbols, the depth of the individual sample is indicated. Arrows indicate selected environmental variable (Latitude, Depth, Sal: salinity, Temp: temperature, oxy: dissolved oxygen concentration, NO2: nitrite concentration, MCGI/mL SW: 16S rRNA gene abundance of Thaumarchaeota, LAC/HACamoA: ratio between LAC and HAC-amoA gene).
supported by the latitudinal variation in beta-diversity (Fig. 3). This variation results from a decrease in similarity between communities from polar to tropical regions (Fig. 7A) followed by an increase in similarity towards Antarctica or Arctic (Fig. 7B). The decrease in similarity from polar AOA communities towards the equator mimics the distance-decay patterns of macroor-ganisms. A weaker (or insignificant) slope of the distance-decay in micro-organisms as compared to macroorganisms is probably attributable to the differences in dispersal limitations between micro- and macroorganisms (Soininen et al. 2007; Soininen 2012). Indeed, the few studies comparing the patterns of both micro- and macroorganisms at the same location support this hypothesis (Mazaris et al. 2010; Soininen et al. 2011). Within micro-organisms, eukaryotes exhibit stronger decay patterns than Bacteria (Soininen et al. 2011). Although we cannot directly compare our results
with those of macroorganisms as we only evaluated AOA in this study, it is still noteworthy that AOA in the Atlantic showed a similar slope in the distance-decay pattern in epipelagic waters as that found for trees, shrubs and bryophytes (Nekola & White 1999), but a steeper slope than that for Bacteria in the International Census of Marine Microbes (ICoMM) study (Zin-ger et al. 2014). However, comparable to Zinger et al. (2014), AOA also show a steeper slope in the distance-decay relationship in surface waters than in the deep waters. Few studies compared the distance-decay relationship for Bacteria and Archaea in the same samples (Barreto et al. 2014). In the latter study, the methano-genic Archaea targeted by the functional gene mcrA showed the steeper slope, followed by the 16S rRNA signature of Archaea and Bacteria (based on 16 rRNA gene abundance), in agreement with our results. This finding might indicate a lower dispersal of AOA as
(A) Towards the equator
(B) From the equator
<u Meso -
Fig. 7 Slope of the decay of similarity with distance from the southernmost AOA community towards the north (black bars) and from the northernmost (grey bars) towards the south, until the Equator (A) and from the Equator on (B). Slope values are in units of ln(similarity) per 1000 km distance. NS, nonsignificant; * significant (P < 0.05).
0.00 0.10 0.20
compared to Bacteria, probably explainable by their lower diversity, abundance and growth rate.
The bipolar distribution of AOA supports the Baas-Becking principle (Baas-Becking 1934) 'everything is everywhere, but the environment selects' and the species-sorting concept of metacommunity theory (Leibold et al. 2004) with weak, albeit stronger, dispersal limits for AOA than for Bacteria. This conclusion is also in agreement with recent findings suggesting the presence of a persistent microbial seed bank in the ocean (Gibbons et al. 2013).
The distribution of AOA throughout the Atlantic might also be explained by the Snowball Earth hypothesis (Ashkenazy et al. 2013). The Snowball Earth hypothesis (Hoffman & Schrag 2002) is based on the evidence that marine ice extended to the equator during the Neo-proterozoic era (~750-635 million years ago) (Ashkenazy et al. 2013). In this framework, Ashkenazy et al. (2013) suggest that the ocean would have been well-mixed and was characterized by a dynamic circulation. These dynamic conditions might have resulted in an either well-mixed or only weakly stratified water column (Ashkenazy et al. 2013). Under these conditions, a uniform distribution of microbes can be expected. The following re-treat of the sea ice and the increasing temperatures in the tropical and temperate ocean might have led to a diversification of the organisms present in these areas, while the high latitudinal organisms would remain relatively unaltered. This hypothesis might also explain the larger latitudinal ranges of high latitude microbes, the higher diversity in low latitude microbes and the higher similarity between Arctic and Southern Ocean polar and subpolar communities.
454-pyrosequencing and T-RLP fingerprinting of AOA communities resulted in similar latitudinal and depth-related patterns, even though these two methods have different resolution level. Similar patterns such as the bipolar distribution, the Rapoport rule or the decrease of beta-diversity with distance from both northern and southern microbial communities towards the equator were found with both techniques.
In contrast to the epipelagic to upper bathypelagic AOA communities, the AOA communities of the lower bathypelagic (>2000 m) do not show a clear bipolar distribution pattern (Figs 3, 4 and 5, Table S5, Supporting information). The lack of a bipolar distribution pattern in the lower bathypelagic AOA communities might be due the generally small population size restricting dispersal as compared to more abundant populations (Martiny et al. 2006). Another possible explanation for the absence of a biogeographic distribution pattern in lower bathypelagic AOA communities might be the influence of resuspension from the sediment on the deepest water masses (Lampitt 1985; Bogucki et al.
1997), the presence of a nepheloid layer [i.e. the turbidity layer above the seafloor containing significant amounts of suspended sediment and/or sediment particles (Biscaye & Eittreim 1977)], or the influence of hydrothermal vent plumes (Lupton et al. 1985). All these would induce changes in the environmental conditions favouring specific AOA OTUs and resulting in a lower similarity of the communities in this depth layer (Fig. 3) as compared to upper bathypelagic communities. The apparent discrepancy between the lower similarity and the relatively high percentage of shared OTUs (Fig. 2) between lower bathypelagic communities from different regions might be explained by the lower abundance of the shared OTUs. In the lower bathy-pelagic communities, 69% of the shared OTUs had more than one sequence per sample as compared to 70% and 79% in the upper bathypelagic and mesopela-gic communities, respectively. The more variable conditions in near-bottom waters are also reflected by higher prokaryotic abundance or heterotrophic production (De Corte et al. 2012) (see representative profiles for the different regions in Fig. S8, Supporting information) than in the overlying waters.
However, hot spots for dispersal of the AOA community in the bathypelagic Atlantic exist as well. The higher amount and percentage of shared OTUs between depth layers (Fig. 5D) in regions characterized by deep-water mass formation (ARCT and SANT) or upwelling (WTRA) indicate that these regions might act as hot spots for dispersion of micro-organisms in the bathy-pelagic realm. The notion of higher dispersal of microorganisms in these areas is also supported by the higher diversity of the lower bathypelagic AOA communities (Fig. 1) in the ARCT and SANT as compared to the NADR and gyral provinces.
Taken together, our results indicate that epipelagic and deep-water AOA follow biogeographic distribution patterns, similar to Bacteria and macroorganisms, though with some differences in the shape of the relationships. More specifically, AOA exhibit a bipolar distribution of AOA communities from epi- to upper bathypelagic waters (<2000 m). The bipolar distribution of surface and deep ocean AOA might be interpreted in the frame of the Baas-Becking principle and the species-sorting theory, or as a result of an historical homogeneous period (Snowball Earth hypothesis) followed by changing environmental conditions that favoured speci-ation and diversification in low latitudes. Moreover, distinct hot spots for archaeal dispersal have been identified, such as the formation sites of deep-water masses and upwelling zones, pointing to the importance of these areas to sustain the diversity of the deep ocean microbial communities. The contrasting biogeographic pattern of the AOA communities inhabiting the deepest
layers (>2000 m) might be a consequence of sporadic events affecting near-bottom water layers and introducing variability in the environmental conditions, as well as by their small population size.
Acknowledgements
We thank the captain and crew of R/V Pelagia and James Cook R/V for their support and splendid atmosphere on board. T. Yokokawa, B. Husain, K. Bergauer, H. Gruber and L. Haberleitner collected or extracted the DNA samples. H. Gruber also contributed to the T-RFLP fingerprinting laboratory work. J. Garcia developed scripts that facilitated the management of the pyrosequence data set. ES was supported by the Austrian Science Fund (FWF) project P27696-B22, DDC received a fellowship of the University of Groningen. Shiptime was provided by the Geotraces project (Dutch Science Foundation ALW-NWO). Laboratory work was supported by the Austrian Science Fund (FWF) projects: I486-B09 and P23234-B11 and by the European Research Council under the European Community's Seventh Framework Program (FP7/2007-2013)/ ERC grant agreement No. 268595 (MEDEA project) to GJH.
References
Abell GC, Revill AT, Smith C et al. (2010) Archaeal ammonia oxidizers and nirS-type denitrifiers dominate sediment nitrifying and denitrifying populations in a subtropical macroti-dal estuary. Isme Journal, 4, 286-300. Agogue H, Brink M, Dinasquet J, Herndl GJ (2008) Major gradients in putatively nitrifying and non-nitrifying Archaea in the deep North Atlantic. Nature, 456, 788-791. Albers SV, Meyer BH (2011) The archaeal cell envelope. Nature
Reviews Microbiology, 9, 414-426. Amend AS, Oliver TA, Amaral-Zettler LA et al. (2013) Macroe-cological patterns of marine bacteria on a global scale. Journal of Biogeography, 40, 800-811. Arístegui J, Gasol JM, Duarte CM, Herndl GJ (2009) Microbial oceanography of the dark ocean's pelagic realm. Limnology and Oceanography, 45, 1501-1529. Ashkenazy Y, Gildor H, Losch M et al. (2013) Dynamics of a
Snowball Earth ocean. Nature, 495, 90-93. Astorga A, Oksanen J, Luoto M et al. (2012) Distance decay of similarity in freshwater communities: do macro- and microorganisms follow the same rules? Global Ecology and Biogeography, 21, 365-375. Auguet JC, Barberan A, Casamayor EO (2010) Global ecological
patterns in uncultured Archaea. Isme Journal, 4, 182-190. Baas-Becking LGM (1934) Geobiologie of Inleiding tot de Milieukunde. Van Stockum, WP & Zoon, NV, The Hague, The Netherlands.
Barberan A, Casamayor EO, Fierer N (2014) The microbial contribution to macroecology. Frontiers in microbiology, 5, 203. Barreto DP, Conrad R, Klose M, Claus P, Enrich-Prast A (2014) Distance-decay and taxa-area relationships for Bacteria, Archaea and methanogenic Archaea in a tropical lake sediment. PLoS ONE, 9, e110128. Bell T (2010) Experimental tests of the bacterial distance-decay relationship. Isme Journal, 4, 1357-1365.
Beman JM, Popp BN, Francis CA (2008) Molecular and biogeochemical evidence for ammonia oxidation by marine Crenarchaeota in the Gulf of California. Isme Journal, 2, 429-441.
Biller SJ, Mosier AC, Wells GF, Francis CA (2012) Global biodiversity of aquatic ammonia-oxidizing Archaea is partitioned by habitat. Frontiers in microbiology, 3, 252.
Biscaye PE, Eittreim SL (1977) Suspended particulate loads and transports in the nepheloid layer of the abyssal Atlantic Ocean. Marine Geology, 23, 155-172.
Bogucki D, Dickey T, Redekopp LG (1997) Sediment resuspension and mixing by resonantly generated internal solitary waves. Journal of Physical Oceanography, 27, 1181-1196.
Cao HL, Auguet JC, Gu JD (2013) Global ecological pattern of ammonia oxidizing Archaea. PLoS ONE, 8, e52853.
Carbonero F, Oakley BB, Purdy KJ (2014) Metabolic flexibility as a major predictor of spatial distribution in microbial communities. PLoS ONE, 9, e85105.
Chou HH, Holmes MH (2001) DNA sequence quality trimming and vector removal. Bioinformatics, 17, 1093-1104.
Christman GD, Cottrell MT, Popp BN, Gier E, Kirchman DL
(2011) Abundance, diversity, and activity of ammonia-oxidizing prokaryotes in the coastal Arctic ocean in summer and winter. Applied and Environmental Microbiology, 77, 2026-2034.
De Corte D, Sintes E, Yokokawa T, Reinthaler T, Herndl GJ
(2012) Links between viruses and prokaryotes throughout the water column along a North Atlantic latitudinal transect. Isme Journal, 6, 1566-1577.
DeLong EE, Pace NR (2001) Environmental diversity of Bacteria and Archaea. Systematic Biology, 50, 470-478.
Ellen AF, Zolghadr B, Driessen AMJ, Albers SV (2010) Shaping the Archaeal Cell Envelope. Archaea, 2010, 608243.
Erguder TH, Boon N, Wittebolle L, Marzorati M, Verstraete W (2009) Environmental factors shaping the ecological niches of ammonia-oxidizing Archaea. Fems Microbiology Reviews, 33, 855-869.
Fernandez-Guerra A, Casamayor EO (2012) Habitat-associated phylogenetic community patterns of microbial ammonia oxi-dizers. PLoS ONE, 7, e47330.
Fierer N (2008) Microbial biogeography: patterns in microbial diversity across space and time. In: Accessing Uncultivated Microorganisms: From the Environment to Organisms and Genomes and Back (ed. Zengler K), pp. 95-115. ASM Press, Washington, District of Columbia.
Fierer N, McCain CM, Meir P et al. (2011) Microbes do not follow the elevational diversity patterns of plants and animals. Ecology, 92, 797-804.
Finlay BJ, Fenchel T (2004) Cosmopolitan metapopulations of free-living microbial eukaryotes. Protist, 155, 237-244.
Foissner W (2006) Biogeography and dispersal of micro-organisms: a review emphasizing protists. Acta Protozoologica, 45, 111 -136.
Francis CA, Roberts KJ, Beman JM, Santoro AE, Oakley BB (2005) Ubiquity and diversity of ammonia-oxidizing Archaea in water columns and sediments of the ocean. Proceedings of the National Academy of Sciences of the United States of America, 102, 14683-14688.
Fuhrman JA, Steele JA, Hewson I et al. (2008) A latitudinal diversity gradient in planktonic marine bacteria. Proceedings of the National Academy of Sciences of the United States of America, 105, 7774-7778.
van der Gast CJ (2015) Microbial biogeography: the end of the ubiquitous dispersal hypothesis? Environmental Microbiology, 17, 544-546.
Gaston KJ, Blackburn TM, Spicer JI (1998) Rapoport's rule: time for an epitaph? Trends in Ecology & Evolution, 13, 70-74.
Ghiglione JF, Galand PE, Pommier T et al. (2012) Pole-to-pole biogeography of surface and deep marine bacterial communities. Proceedings of the National Academy of Sciences of the United States of America, 109, 17633-17638.
Gibbons SM, Caporaso JG, Pirrung M et al. (2013) Evidence for a persistent microbial seed bank throughout the global ocean. Proceedings of the National Academy of Sciences of the United States of America, 110, 4651-4655.
Green JL, Holmes AJ, Westoby M et al. (2004) Spatial scaling of microbial eukaryote diversity. Nature, 432, 747-750.
Hallam SJ, Konstantinidis KT, Putnam N et al. (2006) Genomic analysis of the uncultivated marine Crenarchaeote Cenar-chaeum symbiosum. Proceedings of the National Academy of Sciences of the United States of America, 103, 18296-18301.
Herfort L, Schouten S, Abbas B et al. (2007) Variations in spatial and temporal distribution of Archaea in the North Sea in relation to environmental variables. Fems Microbiology Ecology, 62, 242-257.
Hillebrand H (2004) On the generality of the latitudinal diversity gradient. American Naturalist, 163, 192-211.
Hoffman PF, Schrag DP (2002) The snowball Earth hypothesis: testing the limits of global change. Terra Nova, 14, 129-155.
Horner-Devine MC, Lage M, Hughes JB, Bohannan BJM (2004) A taxa-area relationship for bacteria. Nature, 432, 750-753.
Huang Y, Niu B, Gao Y, Fu L, Li W (2010) CD-HIT Suite: a web server for clustering and comparing biological sequences. Bioinformatics, 26, 680-682.
Lampitt RS (1985) Evidence for the seasonal deposition of detritus to the deep-sea floor and its subsequent resuspension. Deep-Sea Research Part a-Oceanographic Research Papers, 32, 885-897.
Leibold MA, Holyoak M, Mouquet N et al. (2004) The metacommunity concept: a framework for multi-scale community ecology. Ecology Letters, 7, 601-613.
Longhurst A (2007) Ecological Geography of the Sea. Academic Press, Elsevier Inc., Burlington.
Lupton JE, Delaney JR, Johnson HP, Tivey MK (1985) Entrain-ment and vertical transport of deep-ocean water by buoyant hydrothermal plumes. Nature, 316, 621-623.
Martiny JBH, Bohannan BJM, Brown JH et al. (2006) Microbial biogeography: putting microorganisms on the map. Nature Reviews Microbiology, 4, 102-112.
Mazaris AD, Moustaka-Gouni M, Michaloudi E, Bobori DC (2010) Biogeographical patterns of freshwater micro- and macroorganisms: a comparison between phytoplankton, zooplankton and fish in the eastern Mediterranean. Journal of Biogeography, 37, 1341-1351.
Merbt SN, Stahl DA, Casamayor EO et al. (2012) Differential photoinhibition of bacterial and archaeal ammonia oxidation. Fems Microbiology Letters, 327, 41-46.
Mosier AC, Francis CA (2008) Relative abundance and diversity of ammonia-oxidizing Archaea and Bacteria in the San Francisco Bay estuary. Environmental Microbiology, 10, 3002-3016.
Nekola JC, White PS (1999) The distance decay of similarity in biogeography and ecology. Journal of Biogeography, 26, 867-878.
Pester M, Rattei T, Flechl S et al. (2012) amoA-based consensus phylogeny of ammonia-oxidizing archaea and deep sequencing of amoA genes from soils of four different geographic regions. Environmental Microbiology, 14, 525-539.
Pommier T, Canback B, Riemann L et al. (2007) Global patterns of diversity and community structure in marine bacterio-plankton. Molecular Ecology, 16, 867-880.
Ramette A, Tiedje JM (2007) Biogeography: an emerging cornerstone for understanding prokaryotic diversity, ecology, and evolution. Microbial Ecology, 53, 197-207.
Rohde K (1999) Latitudinal gradients in species diversity and Rapoport's rule revisited: a review of recent work and what can parasites teach us about the causes of the gradients? Ecography, 22, 593-613.
Rohde K, Heap M, Heap D (1993) Rapoports rule does not apply to marine teleosts and cannot explain latitudinal gradients in species richness. American Naturalist, 142, 1-16.
Sahan E, Muyzer G (2008) Diversity and spatio-temporal distribution of ammonia-oxidizing Archaea and Bacteria in sediments of the Westerschelde estuary. Fems Microbiology Ecology, 64, 175-186.
Santoro AE, Francis CA, de Sieyes NR, Boehm AB (2008) Shifts in the relative abundance of ammonia-oxidizing Bacteria and Archaea across physicochemical gradients in a subterranean estuary. Environmental Microbiology, 10, 1068-1079.
Schloss PD, Westcott SL, Ryabin T et al. (2009) Introducing mothur: open-source, platform-independent, community-supported software for describing and comparing microbial communities. Applied and Environmental Microbiology, 75, 7537-7541.
Sintes E, Bergauer K, De Corte D, Yokokawa T, Herndl GJ (2013) Archaeal amoA gene diversity points to distinct bio-geography of ammonia-oxidizing Crenarchaeota in the ocean. Environmental Microbiology, 15, 1647-1658.
Soininen J (2012) Macroecology of unicellular organisms - patterns and processes. Environmental Microbiology Reports, 4, 10-22.
Soininen J, McDonald R, Hillebrand H (2007) The distance decay of similarity in ecological communities. Ecography, 30, 3-12.
Soininen J, Korhonen JJ, Karhu J, Vetterli A (2011) Disentangling the spatial patterns in community composition of prokaryotic and eukaryotic lake plankton. Limnology and Oceanography, 56, 508-520.
Stevens GC (1989) The latitudinal gradient in geographical range - how so many species coexist in the tropics. American Naturalist, 133, 240-256.
Sul WJ, Oliver TA, Ducklow HW, Amaral-Zettler LA, Sogin ML (2013) Marine bacteria exhibit a bipolar distribution. Proceedings of the National Academy of Sciences of the United States of America, 110, 2342-2347.
Vellend M, Lajoie G, Bourret A et al. (2014) Drawing ecological inferences from coincident patterns of population- and community-level biodiversity. Molecular Ecology, 23, 2890-2901.
Verhees CH, Kengen SWM, Tuininga JE et al. (2003) The unique features of glycolytic pathways in Archaea. Biochemical Journal, 375, 231-246.
Wang M, Yafremava LS, Caetano-Anolles D, Mittenthal JE, Caetano-Anolles G (2007) Reductive evolution of architectural repertoires in proteomes and the birth of the tripartite world. Genome Research, 17, 1572-1585.
Wetzel CE, Bicudo Dde C, Ector L et al. (2012) Distance decay of similarity in neotropical diatom communities. PLoS ONE, 7, e45071.
Wuchter C, Abbas B, Coolen MJL et al. (2006) Archaeal nitrification in the ocean. Proceedings of the National Academy of Sciences of the United States of America, 103, 12317-12322.
Yarza P, Yilmaz P, Pruesse E et al. (2014) Uniting the classification of cultured and uncultured bacteria and archaea using 16S rRNA gene sequences. Nature Reviews Microbiology, 12, 635-645.
Zhang Y, Sun Y (2011) HMM-FRAME: accurate protein domain classification for metagenomic sequences containing frame-shift errors. BMC Bioinformatics, 12, 198.
Zinger L, Boetius A, Ramette A (2014) Bacterial taxa-area and distance-decay relationships in marine environments. Molecular Ecology, 23, 954-964.
E.S. and G.J.H. designed the work and wrote the study. E.S., D.D.C. and N.O. performed research. E.S. performed all the data analysis.
Data accessibility
Sequence information obtained by Sanger sequencing has been deposited in GenBank, Accession nos KF727022-KF727275. Raw 454-pyrosequences were submitted to the Sequence Read Archive (SRA) at NCBI under the Accession number SRP049002. Station GPS coordinates and sampling date, OTUs abundance obtained by T-RFLP and 454-pyrosequencing have been deposited on Dryad (doi:10.5061/dryad.c6688).
Supporting information
Additional supporting information may be found in the online version of this article.
Fig. S1 Location of the sampling stations along the cruise track in the Atlantic Ocean.
Fig. S2 Rarefaction curves at a 98% identity cutoff for OTU assignment showing the relative richness of the amoA gene in different regions and depth layers of the Atlantic: (A) obtained from cloning and sequencing, (B) obtained from 454-forward pyrosequenced libraries, (C) obtained from 454-reverse pyrose-quenced libraries.
Fig. S3 MDS ordination of similarity (Bray-Curtis) for the archaeal ammonia oxidizer community: from (A) epipelagic, (B) mesopelagic, (C) upper bathypelagic and (D) lower bathy-pelagic waters throughout the Atlantic obtained by T-RFLP fingerprinting.
Fig. S4 Shared OTUs between depth layers.
Fig. S5 Relationship between AOA average temperature range and temperature. Each symbol represents the average temperature range for a sample based on the TRFLP OTUs present at different depth layers: Epi, epipelagic, Meso, mesopelagic, UB, upper bathypelagic, LB, lower bathypelagic.
Fig. S6 Canonical correspondence analysis: for the (A) T-RFLP fingerprints from the reverse region, and (B) 454-pyrosequenc-ing libraries sequenced with the reverse primer.
Fig. S7 Venn diagram showing the contribution of environmental, spatial and temporal factors to the explained variation of the AOA community composition in the Atlantic ocean.
Fig. S8 Depth profiles of one representative station from each oceanographic region in the Atlantic (named as in Fig. S1): of (A) prokaryotic abundance and (B) heterotrophic production. Prokaryotic abundance and production of the entire northern transect is given in De Corte et al. (2012).
Table S1 Stations and depths where samples for cloning and/ or 454-pyrosequencing of archaeal amoA were collected during the Geotraces cruises.
Table S2 Chao richness index (Chao), ACE richness index (Ace), Shannon diversity index (H') and Simpson diversity index (S) of OTUs (defined at 98% similarity) obtained from the AOA clone libraries (cloning) and from the 454 forward (454-F) and reverse (454-R) pyro-sequenced libraries throughout the Atlantic.
Table S3 Number of shared OTUs (98% similarity) between regions for the whole water column and for specific depths for the forward and reverse 454 pyro-sequenced amoA gene.
Table S4 Mean latitudinal range for low abundant AOA OTUs (>0.5% of the total peak height in T-RFLP fingerprints, LA) vs. high abundant OTUs (>5% of the total peak height, HA) and vs. the 454-pyrosequenced OTUs (454) obtained for different depth layers.
Table S5 Adjusted mean and slope of the variation of the Shannon index of diversity versus latitude in the northern (N) and southern (S) hemispheres obtained with T-RFLP and 454-pyrosequencing (454-pyro).
Appendix S1 Material and methods.
