Widespread Distribution and Expression of Gamma A (UMB), an Uncultured, Diazotrophic, γ-Proteobacterial nifH Phylotype

Rebecca Langlois; Tobias Großkopf; Matthew Mills; Shigenobu Takeda; Julie LaRoche

doi:10.1371/journal.pone.0128912

Abstract

Marine dinitrogen (N₂) fixation studies have focused nearly exclusively on cyanobacterial diazotrophs; however γ-proteobacteria are an abundant component of the marine community and have been largely overlooked until recently. Here we present a phylogenetic analysis of all nifH γ-proteobacterial sequences available in public databases and qPCR data of a γ-proteobacterial phylotype, Gamma A (UMB), obtained during several research cruises. Our analysis revealed a complex diversity of diazotrophic γ-proteobacteria. One phylotype in particular, Gamma A, was described in several traditional and quantitative PCR studies. Though several γ-proteobacterial nifH sequences have been described as laboratory contaminants, Gamma A is part of a large cluster of sequences isolated from marine environments and distantly related to the clade of contaminants. Using a TaqMan probe and primer set, Gamma A nifH DNA abundance and expression were analyzed in nearly 1000 samples collected during 15 cruises to the Atlantic and Pacific Oceans. The data showed that Gamma A is an active, cosmopolitan diazotroph found throughout oxygenated, oligotrophic waters reaching maximum abundances of 8 x 10⁴ nifH DNA copies l^-1 and 5 x 10⁵ nifH transcript copies l^-1. Gamma A nifH transcript abundances were on average 3 fold higher than nifH DNA abundances. The widespread distribution and activity of Gamma A indicate that it has potential to be a globally important N₂ fixing organism.

Citation: Langlois R, Großkopf T, Mills M, Takeda S, LaRoche J (2015) Widespread Distribution and Expression of Gamma A (UMB), an Uncultured, Diazotrophic, γ-Proteobacterial nifH Phylotype. PLoS ONE 10(6): e0128912. https://doi.org/10.1371/journal.pone.0128912

Academic Editor: Lucas J. Stal, Royal Netherlands Institute of Sea Research (NIOZ), NETHERLANDS

Received: January 30, 2015; Accepted: May 1, 2015; Published: June 23, 2015

Copyright: © 2015 Langlois et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited

Data Availability: All sequence names are available in the Supporting Information files. Sequences are available from the National Center for Biotechnology Information.

Funding: Tasman Sea sample collection was supported by the New Zealand Ministry of Fisheries BRAG Project ZBD200811 & FRST Oceans Outcome-based Investment. This work was supported by the Deutsche Forschungsgemeinschaft (RO2138/5-1), Bundesministerium für Bildung und Forschung (SOPRAN), and National Sciences and Engineering Research Council grants to JL and a Canada Excellence Research Chair in Ocean Sciences Technologies grant to Douglas Wallace.

Competing interests: The authors have declared that no competing interests exist.

Introduction

Biological dinitrogen (N₂) fixation is an important source of fixed nitrogen to many oligotrophic oceanic regions carried out by a diverse group of Bacteria and Archaea called diazotrophs. All diazotrophic organisms contain the nitrogenase enzyme, which is the only enzyme known to catalyze the reduction of N₂ gas to ammonia. Nitrogenase is composed of two subunits, dinitrogenase reductase and dinitrogenase. The former provides the electrons to the active site of N₂ fixation in the latter subunit. The nifH gene, which codes for dinitrogenase reductase, is highly conserved and present in all diazotrophs, making it an ideal candidate for phylogenetic analyses [1].

The presence of oceanic diazotrophs belonging to γ-proteobacteria was first reported by Zehr et al. (1998) [2]. However, subsequent research focused mainly on cyanobacterial diazotrophs, including Trichodesmium, the unicellular cyanobacteria groups A and B (UCYN A and UCYN B) and diatom-diazotroph symbionts [3]. A recent high-throughput sequencing analysis of nifH PCR amplicons from 12 sites world-wide recovered nifH sequences from α, β, δ, and γ-proteobacteria, methanogens, Clostridia and cyanobacteria and indicated that nifH genes belonging to non-cyanobacterial diazotrophs were diverse and dominant. This suggests that non-cyanobacterial diazotrophs may play a bigger role in global N₂ fixation than previously thought [4].

Proteobacteria are the dominant microbes in the oceans [5]. Within the proteobacteria, γ-proteobacteria form an abundant and metabolically diverse class in marine systems and are capable of photoautotrophy, chemoautotrophy, chemoheterotrophy and facultative anaeroby [6]. In the Sargasso Sea, γ-proteobacterial sequences were the second most abundant group following α-proteobacterial sequences [7], accounting for 20% of the sequences recovered. In an analysis of over 30 Sorcerer II Global Ocean Samples, γ-proteobacteria 16S rRNA gene sequences were the third most dominant group (13%) of sequences following α-proteobacteria (32%) and unclassified proteobacteria (16%) [8]. Among the γ–proteobacteria, the SAR86 ribotype has been described as the largest group of uncultured microbes [9]. Given the prominence of γ-proteobacteria in the ocean and the common occurrence of diazotrophy among this phylum, there is a large potential for the presence of diazotrophic γ-proteobacteria in marine systems.

Among cultured representatives, the soil bacterium Azotobacter vinelandii is arguably the best described γ-proteobacterial diazotroph, and has served as a model organism for the description of the nitrogenase enzyme complex [1]. Other cultured representatives include pathogens such as Klebsiella pneumonia, a human pathogen also found in freshwater and Vibrio splendidus, an oyster pathogen [10]. The cultivated marine γ–proteobacterial species include Vibrio natriegens and Vibrio diazotrophicus, found in salt marsh mud and marine and estuarine waters, respectively [11, 12]. Many γ-proteobacterial nifH phylotypes described from the open ocean so far have low similarity (<86%) to cultivated diazotrophic γ-proteobacteria [2, 13, 14].

In an early study on nifH diversity, Bird et al. (2005) [15] described a group of γ-proteobacterial sequences (Uncultured Marine Bacteria, UMB) that was dominant and actively transcribed the nifH gene throughout the water column in the Arabian Sea. Identical and similar sequences have also been described in the Atlantic and Pacific Oceans [2, 14, 16, 17]. These γ-proteobacterial nifH sequences accounted for over 6% of the sum of all nifH DNA sequences (Trichodesmium, UCYN A, B, and C, Cluster III and two γ-proteobacteria phylotypes) detected by qPCR in a study in the Atlantic Ocean and were the most broadly distributed phylotype [18]. This group was also widely distributed in the Pacific Ocean [19]. In the South Pacific Gyre, γ-proteobacteria phylotypes were dominant at all but two stations [13]. Data is accumulating showing that diazotrophic γ-proteobacteria are prevalent and active throughout open ocean systems. Given the paucity of data available presently, one can only speculate about the physiology and N₂ fixing potential of marine diazotrophic γ-proteobacteria.

In an effort to advance our knowledge about diazotrophic γ-proteobacteria we looked at the diversity and distribution of γ-proteobacterial nifH genes, in particular one sequence (Gamma A, AY896371) that has been targeted by quantitative PCR (qPCR) in several studies [16, 18, 20]. We conducted an in-depth phylogenetic analysis of publically available γ-proteobacterial nifH genes and the γ-proteobacterial nifH phylotypes of marine origin used in qPCR studies to date in order assess the phylogenetic relationship of the Gamma A (UMB) clade to the broader diversity of nifH sequences clustering within the γ-proteobacteria. In addition, abundances of Gamma A nifH DNA and transcripts (cDNA) were measured during several cruises to the Atlantic and Pacific Oceans. The spatial and temporal distribution of Gamma A was analyzed with respect to available physical and chemical properties of the water column. Results indicate that Gamma A phylotypes and transcripts are found throughout oligotrophic surface waters in both the Pacific and Atlantic Oceans. The wide distribution and active transcription of the nifH gene indicate that this group may be important for global N₂ fixation.

Materials and Methods

Sample collection, genetic material isolation and cDNA synthesis

Samples for nucleic acid extraction were collected during 15 cruises to the Atlantic and Pacific Oceans (Table 1). All samples analyzed during this study were collected in international waters and did not require specific permissions. This study did not involve protected or endangered species. One to eight liters of bulk seawater were collected using either a CTD, underway ship pump or trace metal clean tow-fish. Water was filtered onto 0.2 μm Durapore (Millipore) filters using low vacuum (2 kPa) and immediately frozen at -80°C until extraction in the laboratory. Generally, filtrations were stopped either after 2 l of seawater had been filtered or 2 hours had passed. Filtrations usually lasted 45 min. Samples were processed in a darkened, air-conditioned laboratory.

Download:

Table 1. Summary of cruise information and sample preparation.

https://doi.org/10.1371/journal.pone.0128912.t001

Samples were extracted using one of three methods (see Table 1 description): Qiagen DNA Plant Mini Kit, Qiagen RNA/DNA Mini Kit or automated Maxwell 16 Tissue DNA Kit (Promega). Samples were extracted according to the manufacturers’ instructions after filter preparation. All filters were prepared in the following manner. Filters were first broken up by holding the tube in liquid nitrogen for at least 30 s and then pulverized using either a sterile pipette tip or a plastic pestle. 100 μl of lysozyme (5 mg ml^-1) was added to the filter pieces. It was then vortexed for 30 s and incubated for 10 min at room temperature. The kit lysis buffer was then added and the tube was vortexed again for 30 s. Samples from the D361 cruise and those extracted with the Plant Kit (Table 1) were then put through a QiaShredder column (Qiagen) before proceeding with the manufacturer’s instructions. Extraction of samples using the other kits directly followed filter preparation using the kit manufacturer’s instructions.

DNA and RNA samples extracted using Qiagen kits were eluted with 40–80 μl of TE buffer or RNAse-free water, respectively. Maxwell extracted samples were eluted in 180–250 μl of the elution buffer provided by the manufacturer. Due to uneven evaporation, the amount of elute remaining at the end of the Maxwell automatic extraction protocol was recorded. Blank extractions (no filters added) were routinely conducted to test extraction reagents for contamination. DNA and RNA concentrations were measured in all samples using the Picogreen and Ribogreen Quantitation Kits (Molecular Probes), respectively. No DNA or RNA was detected in the blank extraction samples. Possible contaminant DNA in RNA samples was removed completely by using the Turbo DNA-free kit (Ambion). 6 μl RNA (6–108 ng μl^-1) was then transcribed to cDNA using Superscript III reverse transcriptase (Invitrogen), according to the manufacturer’s instructions. The nested nifH PCR reverse primers nifH2 (5’ ADNGCCATCATYTCNCC 3’) and nifH3 (5’ ATRTTRTTNGCNGCRTA 3’) [21] were used. No reverse-transcription (NRT) control reactions were included to test cDNA for DNA contamination.

Abundance estimations using qPCR

Gamma A nifH abundances were estimated in DNA (992 samples) and cDNA (673 samples) using the TaqMan probe qPCR method described in Langlois et al. (2008); forward primer- 5’-TTATGATGTTCTAGGTGATGTG-3‘, reverse primer- 5‘-AACAATGTAGATTTCCTGAG CCTTATTC-3‘ and probe- 5‘-TTGCAATGCCTATTCG-3‘. A qPCR master mix was made using TaqMan master mix (Applied Biosystems), 5 pmol forward primer, 5 pmol reverse primer, 25 pmol probe, 10 μg BSA and either 1 μl of sample (2–80 ng DNA or 2–27 ng cDNA) or 5 μl of water or plasmid standard. Samples were run either in triplicate or duplicate. Duplicate no template controls (NTCs, water) and 7-point plasmid standards were run on every plate. Samples were run on ABI Real-Time PCR cyclers using the default program with 45 cycles and raw data was analyzed using software from the manufacturer (Applied Biosystems).

No amplification was seen in NTCs, blank extraction samples, NRTs or RNA samples, indicating that all reagents and samples were clean and that there was no DNA contaminating the cDNA. The average slope and intercept of the standard curves were -3.38 +/- 0.08 and 38.15 +/- 0.44, respectively. The average primer efficiency (E = 10^-1/slope-1) was 99.6% (range 94.9–104.2%). The qPCR cycler detection limit was 1 copy reaction^-1. After accounting for elution and filtration volumes, the actual detection limit ranged from 20 to 80 copies l^-1. Samples where detection was not observed in all replicates were considered not quantifiable and removed from the data set (four samples in total).

In order to compare nifH transcript abundances between areas of high and low nifH DNA abundances, cDNA abundances were normalized to corresponding DNA abundances (cDNA:DNA ratios). All samples where nifH abundances were greater than 80 copies l^-1 in both the DNA and cDNA samples were used to calculate cDNA:DNA nifH ratios.

Sequence analysis

The Gamma A target nucleotide sequence (AY896371) in our qPCR assay was compared against the NCBI data bank using a Blastn search with the default maximum number of target sequences changed to 20000. Sample collection information and references were downloaded for all identical and highly similar (>99%) sequences (90 sequences). Sequences that were targets in qPCR studies were downloaded individually and compared using BioEdit Sequence Alignment Editor (v. 7.0.5.3). The 20000 Blastn search result sequences were downloaded and sequences that did not correctly translate into an uninterrupted NifH protein were removed. All non-γ-proteobacterial sequences were also removed, leaving 6240 sequences (S1 Fig, S1 Table). Two comparisons were made using the Blastn search results. First all sequences annotated as ‘uncultured’ or ‘unidentified’, except for sequences used as targets in qPCR studies of marine origin, were removed, allowing the comparison of key sequences in the context of cultured organisms. Then the ‘uncultured’ and ‘unidentified’ sequences (regardless of origin) were added back to the alignment. The following pipeline was used to construct all phylogenetic trees. Sequence identities were compared and operational taxonomic units (OTUs) at 100%, 99%, and 97% identical were formed using h-cd-hit-est [22]. Unless stated otherwise, the OTUs presented are at 97% (> 97% sequence similarity). Sequences were then aligned using the Multiple Sequence Comparison by Log Expectation tool (www.ebi.ac.uk). Alignments were checked to ensure that the correct reading frame was preserved. Two phylogenetic trees of each alignment were compared using FigTree (v1.4.1). Sequences were imported into JalView [23] and a neighbor-joining tree was constructed. RAxML (Randomized Accelerated Maximum Likelihood) Black Box was used to construct a maximum-likelihood tree [24]. Conserved properties between the two tree construction methods were identified in the phylogenetic trees by black branches and bootstrap values.

Statistical analysis

Gamma A nifH abundances were analyzed in the context of a suite of environmental parameters including nutrients, salinity, temperature and oxygen collected concurrently. Surface nifH phylotype distribution and nutrient plots were constructed using Ocean Data View (v 4.3.6). A Principle Components Analysis (PCA) was conducted using only samples (344) for which all metadata (temperature, salinity, oxygen, silicate, nitrate and phosphate) were collected using PRIMER 6 (v 6.1.12). Before the PCA, draftsman’s plots of all variables were made to evaluate the magnitude and skew of the variables and to identify variable pairs with high correlations. After transformation, variable values were within three orders of magnitude of each other and were distributed evenly. Nitrate concentrations were highly correlated with phosphate (R² = 0.98) and oxygen (R² = 0.90) and were not used in the PCA. The N:P ratio did not correlate highly with the other variables and was used instead of the nitrate and phosphate concentrations. Variables were normalized ((x-μ)*σ ^-1, where x is the sample variable, μ is the mean, and σ is the standard deviation) before proceeding with the PCA. A resemblance matrix using the Euclidean distance measure was calculated and a hierarchical agglomerative cluster analysis was performed to investigate grouping of samples. Samples were also assigned to categories based on sample collection depth (surface- <20 m, mid- 20–100 m, or deep- >100 m) and type (coastal- <1000 km to continental margin or open- >1000 km to continental margin). An Analysis of Similarities (ANOSIM) test was performed to determine the statistical significance of the categories at a significance level of 0.1%. The environmental variables were further analyzed by PCA cluster using MiniTab17 software. ANOVA (with Tukey post-hoc) and t-tests were run and considered statistically significant at P < 0.05.

Results and Discussion

Phylogenetic analysis of γ-proteobacterial nifH

Several nifH diversity studies have reported γ-proteobacterial sequences; however each research group used different terminology, which makes it difficult to compare data across studies [13, 19, 25]. In addition different sets of sequences were used to construct phylogenetic trees, leading to conflicting conclusions about the phylogenetic grouping of phylotypes (for example Gamma 1 and Gamma 2 as α-proteobacteria in [25] versus γ-proteobacteria in [13]). In order to compare our data with published work, we conducted a large-scale phylogenetic analysis of γ-proteobacterial nifH sequences and identified similarities and differences in the various target sequences used to identify marine γ-proteobacteria. A search of the public databases revealed a large number of diverse γ–proteobacterial nifH sequences (6235 sequences), encompassing both cultured (90 unique OTUs) and uncultured representatives (1402 unique OTUs) (Fig 1). The uncultured γ–proteobacterial sequences were recovered from a variety of marine, aquatic, and terrestrial environments including soils, invertebrate guts, hot springs, hydrothermal vents, estuaries, and open oceans. Environmental (uncultured) sequences greatly outnumbered sequences from cultures forming 35 distinct clades, which often consisted entirely of environmental sequences distantly related to cultured organisms (Fig 1). This analysis shows that γ–proteobacteria are vastly underrepresented in culture collections and have been barely touched upon in environmental studies.

Download:

Fig 1. Maximum likelihood tree of γ-proteobacterial nifH OTUs.

To improve visibility only OTUs ≥3 sequences are shown (400 of 1492 OTUs) and sequences names are not displayed (see S4 Fig for the sequence names). Clades of uncultured sequences from marine environments are shaded blue and from a variety of environments are shaded green. Clades containing several qPCR target sequences are labeled. Clades encompassing sequences isolated from PCR reagents are colored red. Probes used in qPCR studies are labeled and include Gamma 1, 2, 4, P, ETSP1, and ETSP3, and Zhang (S2 Fig). Tree was constructed using neighbor joining and RAxML methods. Significant branching (agreement between both algorithms and bootstrap >60) is colored black.

https://doi.org/10.1371/journal.pone.0128912.g001

At least ten different uncultured γ–proteobacterial nifH sequences belonging to six separate clades have been used as targets in qPCR studies of marine environments [13, 18, 25–27]. These unique target sequences are all distantly related (<84% similarity) to clades containing sequences demonstrated to be contaminants in PCR reagents (Figs 1 and 2). The Gamma 4 (P8 in [28]) and Zhang sequences [13, 26] had the least similarity (<84 and 83%, respectively) to all other target sequences and formed two distantly-related, separate clades with environmental sequences from sediments, soil and estuaries. The remaining target sequences fell into one of three clades containing large numbers of environmental sequences. Clade Enviro 1 included Gamma 2, P and 1, as well as sequences from the open ocean, sediments, hot springs and seagrasses (Fig 1 and S4 Fig). The Gamma 2 and P sequences [13, 18] had 91% similarity. Gamma 1 had a higher similarity to (86% similar) to cultivated γ–proteobacteria than to Gamma 2 and P [13, 18] (85 and 83% similarity, respectively) (Figs 1 and 2). GammaETSP1 and GammaETSP3 grouped together (87% similarity) [25] with other sequences from marine environments in Clade Marine 2 (Fig 1). These sequences had a 90% similarity to Teredinibacter, a γ-proteobacterial symbiont found in the gills of shipworms [29]. Gamma A, 3, and ETSP2 clustered together in another clade of uncultured sequences from the open ocean [13, 18, 25], distantly related to cultivated organisms (<80% similarity, Figs 1 and 2). Bird et al. (2005) [15] called the Gamma A phylotype, unidentified marine bacteria (UMB). This phylotype is part of a much larger, well-conserved clade of uncultured marine bacteria, which we refer to as Marine 1.

Download:

Fig 2. Maximum-likelihood tree of the nearest cultured neighbors to the γ-proteobacterial nifH OTUs targeted in marine qPCR studies.

Sequences that have been used as targets for qPCR studies are colored blue and labeled as in the original publications: Gamma 1–4 from Halm et al. (2011) [13], Gamma A and P from Langlois et al. (2008) [18], ETSP1-3 from Turk-Kubo et al. 2014 [25], and Zhang et al. (2011) [26]. Tree was constructed using neighbor joining and RAxML methods. Significant branching (agreement between both algorithms and bootstrap >70) is colored black and bootstrap values are shown.

https://doi.org/10.1371/journal.pone.0128912.g002

Using the nifH sequences analyzed in this study, the Marine 1 clade was comprised of 11 OTUs representing 474 sequences (Fig 3). Within this clade, Gamma A was by far the OTU with the largest number of representatives, containing 313 sequences originating from 23 independent studies. For comparison the next largest OTU, Gamma B represented by HQ611810.1, was comprised of 28 sequences from 5 studies. In addition to Gamma A, 3, and ETSP2, potentially five more nifH phylotypes in the Marine 1 clade would be good candidates for further study (Fig 3). These new phylotype targets were described in multiple studies; however they may not be as common as Gamma A based on the number of sequences forming the OTU.

Download:

Fig 3. Maximum-likelihood tree of Marine 1 nifH OTUs.

For visualization, only OTUs of ≥3 sequences are shown. Maximum likelihood and neighbor joining algorithms were compared. Bootstrap values above 70 are shown and significant branches (bootstrap >70 and agreement between both algorithms) are shown in black. Sequences are identified by accession number, first word in title, the number of sequences in the OTU, and then the number of studies where the sequence was found. For comparison purposes, sequences used as probes and the closest related organisms are included. The qPCR phylotype targets in the Marine 1 clade are labeled (black), as are potential phylotype targets (grey).

https://doi.org/10.1371/journal.pone.0128912.g003

Abundances of the Gamma A nifH phylotype were measured using TaqMan probes in several studies in the marine environment (e.g. [16, 18, 30]). Although three different primer-probe sets have been used so far to target Gamma A, sequence comparisons indicate that all probe and primer sets will target all Gamma A sequences while discriminating the other nine γ–proteobacterial nifH phylotypes (S2 Fig, Table 2). The Gamma A sequence (AY896371) was identical to 13 other sequences in our database. When the sequence identity was lowered to 99%, which corresponds to two mismatches in the nifH amplicon, the OTU contained 194 sequences, all recovered from the open ocean. The Gamma A OTU includes all three Gamma A primer-probe type sequences and five of the original ten UMB sequences described by Bird et al. (2005) [15]. Interestingly, the first reported and most commonly occurring Gamma A sequence in the NCBI database, AF059623 [2], has never been directly used to design a probe in either a qPCR or microarray study, but is targeted by the Gamma A qPCR probes and primers.

Download:

Table 2. Comparison of Gamma A primer and probe sequences to the other γ–proteobacterial nifH sequences used as targets in qPCR studies.

https://doi.org/10.1371/journal.pone.0128912.t002

Distribution of Gamma A nifH phylotypes and transcripts

Abundances of Gamma A nifH phylotypes were estimated in 992 samples using TaqMan qPCR (Fig 4a). The samples were collected at 494 stations from 15 cruises between 2000 and 2010 (Table 1). Depths from the surface to 500 m were sampled, but over half (536 samples) were collected at depths shallower than 20 m. Gamma A nifH DNA was detected and quantifiable in 67% of the samples and abundances ranged from the detection limit to 8 x 10⁴ nifH copies l^-1. The highest Gamma A nifH abundances were detected in the tropical Atlantic Ocean between 28°N and 10°S, but were also elevated in the sub-tropical North Pacific Ocean between 10–20°N. It should be noted that only 194 samples were collected in the Pacific Ocean compared to 798 samples in the Atlantic Ocean. Gamma A nifH transcript abundance was estimated in parallel in 673 samples and was detected in 64% of the samples (Fig 4b). Gamma A transcript abundances were correlated with DNA abundances, but were usually higher and ranged from the detection limit to 4.7 x 10⁵ nifH copies l^-1. The highest transcript abundances were detected in the tropical Atlantic Ocean between 20°N and 10°S.

Download:

Fig 4. Surface abundances of Gamma A nifH DNA (a) and cDNA (b) estimated by TaqMan qPCR.

Abundances are presented as log (nifH copies l^-1) where red colors indicate high abundances and purple near detection limit abundances.

https://doi.org/10.1371/journal.pone.0128912.g004

In addition to our study sites, Gamma A sequences (sequences with similarity greater than 98% to Gamma A) have also been reported in a variety of marine ecosystems including the oligotrophic Mediterranean [31], Red [32], and Arabian Seas [15], the edge of the Mekong River Plume [17], South China Sea [16], Indian Ocean [27], South Pacific [19], and Southern California Bight [33]. Gamma A was found at all stations throughout the upper water column in the Arabian Sea [15] and at low abundances in the Red Sea [32]. In the Indian and South Pacific Oceans, South China Sea during inter-monsoon, and oceanic stations in the Mekong River Plume, abundances were similar to those reported in the present study [16, 17, 19, 27]. Gamma A appears to be a cosmopolitan diazotroph more commonly found in tropical and sub-tropical, oligotrophic regions.

A multivariate analysis of the samples supports the conclusion that the Gamma A phylotype has a global distribution throughout tropical to sub-tropical surface oceans. A PCA of the environmental parameters showed that the samples grouped significantly with respect to depth and distance from the continental margin (ANOSIM, P < 0.01; Fig 5a and 5b). A resemblance matrix of the data was constructed to further compare similarities between the samples and a hierarchical cluster analysis revealed three clusters (SIMPROF, Fig 5). Gamma A abundances were significantly higher in Cluster 1. The average dissolved nutrient concentration of Cluster 1 samples was significantly lower than Cluster 2 and 3 samples (ANOVA, P < 0.01, S2 Table). In contrast, the average Cluster 1 oxygen concentration (194+/- 1.6 μM) was nearly double that of Cluster 2 and 3 (100+/-9.7 and 109+/-4.5 μM). Cluster 1 samples were collected at depths shallower than 100 m (average 26 +/- 1.6) and contained the highest concentrations of Gamma A nifH DNA with an average of 3400 +/- 500 nifH copies l^-1. At a shorter cluster distance of 2, cluster 1 separates into two clusters, one containing open ocean samples (1a) and the second containing shelf samples (1b, Fig 5b). Gamma A abundances were significantly higher in open ocean samples (9000 +/- 800 nifH copies l^-1) than in samples collected closer to the coast (2700 +/- 300 nifH copies l^-1; T-test, P = 0.005).

Download:

Fig 5. Principal components analysis of environmental parameters for samples where Gamma A nifH abundances were estimated.

The cluster analysis distance of 3.1 is shown in panels ‘a-c’ and 2 in ‘b’. Samples are identified by collection depth (surface- <20 m, mid- 20–100 m, deep- >100 m) in ‘a’ and distance to the continental margin (coastal- <1000 km or open- >1000 km) in ‘b’. The Gamma A abundances (nifH copies l^-1) are displayed in ‘c’.

https://doi.org/10.1371/journal.pone.0128912.g005

In addition to dissolved nutrient and oxygen concentrations, Cluster 2 and 3 differed from Cluster 1 with respect to depth, temperature, and salinity. Cluster 2 (coastal shelf) and 3 (open ocean) samples were collected at average depths of 66 +/- 5.3 and 177 +/-14 m, respectively. Samples in Cluster 2 had the highest average temperature (36 +/- 0.06°C), but the lowest average salinity (18 +/- 0.4 psu), indicating a freshwater influence. Cluster 3 samples had the highest average salinity (36 +/- 0.04 psu), but the lowest average temperature (16 +/- 0.4°C) of the three clusters. The environmental conditions in Cluster 2 and 3 samples were perhaps not as favorable to Gamma A as in Cluster 1; Gamma A abundance greater than 10² nifH copies l^-1 was detected in only one sample in clusters 2 and 3 (Fig 5c). Average abundances of samples in clusters 2 and 3 were only 580 +/- 500 and 100 +/- 30 nifH copies l^-1, respectively. Although not all types of marine environments have been tested for the presence of Gamma A, this data set shows the highest Gamma A abundances were detected in warm, well-oxygenated, oligotrophic waters, indicating a preference for tropical or sub-tropical oceans.

Environmental Niche for Gamma A

Patterns of nifH transcripts normalized to nifH DNA abundances were observed when compared with environmental parameters (Fig 6 and S3 Fig). Gamma A nifH cDNA:DNA was highest in waters with temperatures ranging from 20–30°C (Fig 6a) and at depths of 100 m and shallower (Fig 6b, S2 and S3 Tables). In the surface and upper water column samples, nutrient concentrations were low to undetectable (S2 Fig). For samples where nitrate and phosphate concentrations were measurable, N* was calculated (NO3 - 16PO4 [34]). Elevated Gamma A nifH cDNA:DNA was observed at N* concentrations from -5 to 5 μM, however the highest cDNA:DNA ratios (>5) were from samples where N* concentrations were between -2.5 to 0 μM (Fig 6d). It should be noted, though, that a large number of samples, particularly cDNA samples, were collected in the tropical oceans (S4 Table). Despite this, research conducted in geographic areas separate from our study support these conclusions. Moisander et al. (2008) also observed Gamma A only in the upper, well-lit and oxygenated water column [16]. In the Indian Ocean, Shiozaki et al. (2014) found the highest abundances of Gamma A in surface ocean waters [27]. Gamma A abundances in the South Pacific correlated positively with temperature and oxygen and negatively with nutrients and depth [19]. Several studies have looked at nifH diversity in oxygen-minimum and-depleted zones and reported γ-proteobacterial sequences such as Gamma 4, ETSP1, or ETSP3 but no sequences similar to Gamma A or any other Marine 1 sequences were present in those clone libraries [35–37]. No Gamma A was found in clone libraries from samples collected world-wide at depths of 500–5900m [38]. The Gamma A nifH phylotype abundance and expression are associated with warm, oligotrophic, oxygenated surface waters that have an N deficit (i.e. the tropical and sub-tropical oligotrophic oceans).

Download:

Fig 6. Activity ratio (cDNA:DNA) where Gamma A nifH cDNA and DNA were detected by TaqMan qPCR as a function of measured environmental parameters.

The cDNA:DNA ratio is graphed versus temperature (°C, a), depth (m, b), O₂ concentration (μM, c) and N* (μM, d). The cDNA:DNA ratio was calculated only for samples in which both Gamma A DNA and cDNA abundances were >80 nifH copies l^-1. Please note that not all parameters were available for all samples.

https://doi.org/10.1371/journal.pone.0128912.g006

Nitrogenase activity is sensitive to oxygen and thus the presence of the Gamma A phylotype and expression of nifH in oxygenated waters is intriguing. Strictly heterotrophic bacteria do not rely on sunlight for energy; however their presence in the oxygenated surface waters may reflect other metabolic preferences, such as an ample supply of organic carbon. In addition, Gamma A could be capable of photoheterotrophy, a metabolic pathway that uses the energy from sunlight to activate a proton gradient which then augments the organism’s ability to take in organic carbon [39, 40]. Photoheterotrophy may be used to offset the energy requirements for N₂ fixation [41]and would explain why this diazotroph is found in well-lit, oxygenated waters. An alternative explanation for the observed higher abundances of Gamma A in the upper water column may be the availability of labile dissolved organic carbon (DOC) that can fuel heterotrophy. DOC is produced in tropical-subtropical oceans by autotrophs and can accumulate in surface waters, where the highest concentrations are measured, due to vertical stratification [42].

Like many questions concerning the physiology of Gamma A, how this diazotroph reconciles the incompatible high oxygen levels with anaerobic N₂ fixation still remains to be elucidated. Diazotrophs living in oxygenated environments have developed many ways to protect the nitrogenase enzyme, ranging from specialized cell structures, called heterocysts and diazocytes, to temporal separation of oxygen producing processes (such as photosynthesis) and nitrogenase activity [43]. Gamma A may reconcile biological N₂ fixation and inhabiting an aerobic environment by living in association with other organism(s), as exemplified by Candidatus Atelocyanobacterium thalassa or Richelia sp. [44, 45], where the host could provide a carbon source and possibly protect the symbiont from harmful oxygen concentrations in return for fixed nitrogen. If this strategy is used, the light requirements of a putative photosynthetic host may explain the observed depth distribution of Gamma A. Alternatively, Gamma A could avoid high oxygen concentrations by living in association with particles, which have been shown to provide diazotrophs with low oxygen environments and high organic carbon [46, 47].

Though we have reported a large geographic coverage of Gamma A DNA and cDNA abundances, nifH abundance data cannot be converted into N₂ fixation rates. Studies that have measured both γ-proteobacterial phylotype abundances and N₂ fixation rates have reached opposing conclusions about the N₂ fixing potential of heterotrophic diazotrophs. Halm et al. (2011) attributed much of the N₂ fixation they measured to the dominant γ-proteobacterial phylotypes they detected [13]. Conversely, Turk-Kubo et al. (2014) concluded that the abundances of γ-proteobacteria that they detected were too low to account for the N₂ fixation rates they measured [25]. The Gamma A abundances detected in our study were high enough to produce measurable N₂ fixation rates, according to the calculations in Turk-Kubo et al. (2014) [25]. However, a more thorough assessment of their importance in the ocean will require cell specific N₂ fixation rates for open ocean γ-proteobacterial diazotrophs. Presently, it is unclear how much of the global N₂ fixation can be attributed to heterotrophic diazotrophs.

Conclusions

Heterotrophic bacteria are prevalent throughout oligotrophic oceans [48] and γ–proteobacteria are an abundant constituent of this community [8, 49]. Following cyanobacterial nifH phylotypes, γ-proteobacterial phylotypes were the next most abundant diazotrophic phylotype in the Atlantic Ocean [18]. Although non-cyanobacterial nifH genes can be more abundant and appear to have a larger distribution than cyanobacterial nifH genes [4, 25], they have received very little attention until recently [41]. This study presents extensive data on the abundance and expression of Gamma A, a putatively heterotrophic diazotroph. Our work and that of others [19, 27] indicate that the Gamma A phylotype is a globally distributed and active diazotroph based on the wide distribution of Gamma A nifH DNA and cDNA throughout the tropical, oligotrophic surface oceans. It is not yet clear how much of an impact this diazotroph has on local and global nitrogen- and carbon-cycles, but it is clear that this organism warrants more attention.

Supporting Information

S1 Alignment. Alignment of nifH sequences used in Fig 1.

https://doi.org/10.1371/journal.pone.0128912.s001

(PDF)

S2 Alignment. Alignment of nifH sequences used in Fig 2.

https://doi.org/10.1371/journal.pone.0128912.s002

(PDF)

S3 Alignment. Alignment of nifH sequences used in Fig 3.

https://doi.org/10.1371/journal.pone.0128912.s003

(PDF)

S1 Fig. Phylogenetic tree of nifH from cultured organisms in relation with nifH OTU’s from environmental sequences used as qPCR target phylotypes and nifH contaminants of PCR Reagents.

The tree was constructed using the neighbor joining method. Clades have been collapsed taxonomically by class according to the given color scheme. The number of sequences comprising each clade is shown. Clades that are not collapsed contained sequences originating from multiple taxonomic classes. The clades containing the qPCR phylotype targets are labeled. Clades containing sequences originating from PCR reagents are colored red. Sequences from all clades identified as γ-proteobacteria were used in further analyses.

https://doi.org/10.1371/journal.pone.0128912.s004

(TIF)

S2 Fig. Dot plot alignment of γ–proteobacterial nifH amplicons used as targets in qPCR studies.

The full Gamma A amplicon is shown at the top. The amplicons used in other studies follow, identified by either the first author’s name or probe name and sequence accession number in brackets. The first six sequences (bold font) are in the Marine 1 clade. The Gamma A primers (light grey bars) and probes (dark grey bars) used in qPCR studies are shown.

https://doi.org/10.1371/journal.pone.0128912.s005

(TIF)

S3 Fig. Plot of Gamma A cDNA:DNA versus NO₃ (μM, a), PO₄ (μM, b), and SiO₂ (μM, c) concentrations.

Note that the x-axis scale changes in each panel.

https://doi.org/10.1371/journal.pone.0128912.s006

(TIF)

S4 Fig. Linear maximum likelihood tree of γ-proteobacterial nifH OTUs.

This tree is a reproduction of Fig 1, including the accession numbers, first word in title, the number of sequences in the OTU, and isolation environmental source (if available).

https://doi.org/10.1371/journal.pone.0128912.s007

(PDF)

S1 Table. List of accession numbers for sequences identified as γ-proteobacteria.

https://doi.org/10.1371/journal.pone.0128912.s008

(PDF)

S2 Table. Descriptive statistics summary of PCA clusters.

The sample size of each cluster is given (n). The calculated mean, and standard error (SE), and range of values measured for each parameter are given. Means in bold font are statistically different (ANOVA, P < 0.01).

https://doi.org/10.1371/journal.pone.0128912.s009

(PDF)

S3 Table. Detection Frequency of Gamma A in qPCR Analysis.

The number of samples where Gamma A abundances (nifH copies l^-1) were detected per each depth range (m) are shown. The number of samples per depth range (n) is given and the detection frequency (%) was calculated.

https://doi.org/10.1371/journal.pone.0128912.s010

(PDF)

S4 Table. Sampling frequency arranged by latitudinal bands.

The number of DNA and cDNA samples collected in each latitudinal band (°N) is provided.

https://doi.org/10.1371/journal.pone.0128912.s011

(PDF)

S5 Table. Designation of sequences in the Marine 1 clade into OTUs at 100% similarity.

Representative sequences in Fig 3 are in bold font.

https://doi.org/10.1371/journal.pone.0128912.s012

(PDF)

S6 Table. Designation of sequences in the Marine 1 clade into OTUs at 99% similarity.

Representative sequences in Fig 3 are in bold font.

https://doi.org/10.1371/journal.pone.0128912.s013

(PDF)

S7 Table. Designation of sequences in the Marine 1 clade into OTUs at 97% similarity.

Representative sequences in Fig 3 are in bold font.

https://doi.org/10.1371/journal.pone.0128912.s014

(PDF)

Acknowledgments

The authors are grateful to many people who filtered thousands of liters of water. Mark Moore and Micha Rijkenberg collected samples in the Atlantic Ocean. Cliff Law and Andrew Marriner collected samples in the Tasman Sea. Stefanie Sudhaus, Orly Levitan and Christian Schlosser helped sample during the M68/3 cruise. Without the help of these people, this manuscript would not have been possible.

Author Contributions

Conceived and designed the experiments: RL. Performed the experiments: RL. Analyzed the data: RL. Contributed reagents/materials/analysis tools: RL JL. Wrote the paper: RL JL TG ST MM. Collected samples: RL TG ST MM.

References

1. Postgate J. Nitrogen Fixation. 3rd ed. Boston: Cambridge University Press; 1998.
2. Zehr JP, Mellon MT, Zani S. New Nitrogen-Fixing Microorganisms Detected in Oligotrophic Oceans by Amplification of Nitrogenase (nifH) Genes. Applied and Environmental Microbiology. 1998;64(9):3444–50. pmid:9726895
- View Article
- PubMed/NCBI
- Google Scholar
3. Luo YW, Doney SC, Anderson LA, Benavides M, Berman-Frank I, Bode A, et al. Database of diazotrophs in global ocean: abundances, biomass and nitrogen fixation rates. Earth Syst Sci Data. 2012;4(1):47–73.
- View Article
- Google Scholar
4. Farnelid H, Andersson AF, Bertilsson S, Al-Soud WA, Hansen LH, Sorensen S, et al. Nitrogenase Gene Amplicons from Global Marine Surface Waters are Dominated by Genes of Non- Cyanobacteria. PLoS ONE. 2011;6(4):e19223. Epub 2011/05/12. pmid:21559425; PubMed Central PMCID: PMC3084785.
- View Article
- PubMed/NCBI
- Google Scholar
5. Giovannoni SJ, Vergin KL. Seasonality in Ocean Microbial Communities. Science. 2012;335(6069):671–6. pmid:22323811
- View Article
- PubMed/NCBI
- Google Scholar
6. Zavarzin GA, Stackebrandt E, Murray RG. A correlation of phylogenetic diversity in the Proteobacteria with the influences of ecological forces. Can J Microbiol. 1991;37(1):1–6. Epub 1991/01/01. pmid:1708691.
- View Article
- PubMed/NCBI
- Google Scholar
7. Venter JC, Remington K, Heidelberg JF, Halpern AL, Rusch D, Eisen JA, et al. Environmental Genome Shotgun Sequencing of the Sargasso Sea. Science. 2004;304(5667):66–74. pmid:15001713
- View Article
- PubMed/NCBI
- Google Scholar
8. Rusch DB, Halpern AL, Sutton G, Heidelberg KB, Williamson S, Yooseph S, et al. The Sorcerer II Global Ocean Sampling Expedition: Northwest Atlantic through Eastern Tropical Pacific. PLoS Biol. 2007;5(3):e77. pmid:17355176
- View Article
- PubMed/NCBI
- Google Scholar
9. Dupont CL, Rusch DB, Yooseph S, Lombardo MJ, Richter RA, Valas R, et al. Genomic insights to SAR86, an abundant and uncultivated marine bacterial lineage. ISME J. 2012;6(6):1186–99. Epub 2011/12/16. pmid:22170421; PubMed Central PMCID: PMC3358033.
- View Article
- PubMed/NCBI
- Google Scholar
10. Grimes DJ, Johnson CN, Dillon KS, Flowers AR, Noriea NF III, Berutti T. What Genomic Sequence Information Has Revealed About Vibrio Ecology in the Ocean- A Review. Microb Ecol. 2009;58:447–60. pmid:19727929
- View Article
- PubMed/NCBI
- Google Scholar
11. Guerinot ML, West PA, Lee JV, Colwell RR. Vibrio diazotrophicus sp. nov., a Marine Nitrogen-Fixing Bacterium. International Journal of Systematic Bacteriology. 1982;32(3):350–7.
- View Article
- Google Scholar
12. Payne WJ, Eagon RG, Williams AK. Some observations on the physiology of Pseudomonas natriegens nov. spec. Antoine van Leeuwenhoek. 1961;27(1):121–8.
- View Article
- Google Scholar
13. Halm H, Lam P, Ferdelman TG, Lavik G, Dittmar T, LaRoche J, et al. Heterotrophic organisms dominate nitrogen fixation in the South Pacific Gyre. ISME J. 2012;6(6):1238–49. http://www.nature.com/ismej/journal/v6/n6/suppinfo/ismej2011182s1.html. pmid:22170429
- View Article
- PubMed/NCBI
- Google Scholar
14. Langlois RJ, LaRoche J, Raab PA. Diazotrophic Diversity and Distribution in the Tropical and Subtropical Atlantic Ocean. Applied and Environmental Microbiology. 2005;71(12):7910–9. pmid:16332767
- View Article
- PubMed/NCBI
- Google Scholar
15. Bird C, Martinez Martinez J, O'Donnell AG, Wyman M. Spatial Distribution and Transcriptional Activity of an Uncultured Clade of Planktonic Diazotrophic y-Proteobacteria in the Arabian Sea. Applied and Environmental Microbiology. 2005;71(4):2079–85. pmid:15812041
- View Article
- PubMed/NCBI
- Google Scholar
16. Moisander PH, Beinart RA, Voss M, Zehr JP. Diversity and abundance of diazotrophic microorganisms in the South China Sea during intermonsoon. ISME J. 2008;2(9):954–67. Epub 2008/06/06. pmid:18528417.
- View Article
- PubMed/NCBI
- Google Scholar
17. Bombar D, Moisander P, Dippner J, Foster R, Voss M, Karfeld B, et al. Distribution of diazotrophic microorganisms and nifH gene expression in the Mekong River plume during intermonsoon. Marine Ecology Progress Series. 2011;424:39–52.
- View Article
- Google Scholar
18. Langlois RJ, Hümmer D, LaRoche J. Abundances and Distributions of the Dominant nifH Phylotypes in the Northern Atlantic Ocean. Applied and Environmental Microbiology. 2008;74(6):1922–31. pmid:18245263
- View Article
- PubMed/NCBI
- Google Scholar
19. Moisander PH, Serros T, Paerl RW, Beinart RA, Zehr JP. Gammaproteobacterial diazotrophs and nifH gene expression in surface waters of the South Pacific Ocean. ISME J. 2014;8(10):1962–73. pmid:24722632
- View Article
- PubMed/NCBI
- Google Scholar
20. Church MJ, Jenkins BD, Karl DM, Zehr JP. Vertical distributions of nitrogen-fixing phylotypes at Stn ALOHA in the oligotrophic North Pacific Ocean. Aquatic Microbial Ecology. 2005;38(1):3–14. ISI:000226947900002.
- View Article
- Google Scholar
21. Zehr JP, Crumbliss LL, Church MJ, Omoregie EO, Jenkins BD. Nitrogenase genes in PCR and RT-PCR reagents: implications for studies of diversity of functional genes. BioTechniques. 2003;35:996–1005. pmid:14628674
- View Article
- PubMed/NCBI
- Google Scholar
22. Huang Y, Niu B, Gao Y, Fu L, Li W. CD-HIT Suite: a web server for clustering and comparing biological sequences. Bioinformatics. 2010;26(5):680–2. pmid:20053844
- View Article
- PubMed/NCBI
- Google Scholar
23. Waterhouse AM, Procter JB, Martin DMA, Clamp M, Barton GJ. Jalview Version 2—a multiple sequence alignment editor and analysis workbench. Bioinformatics. 2009;25(9):1189–91. pmid:19151095
- View Article
- PubMed/NCBI
- Google Scholar
24. Stamatakis A, Hoover P, Rougemont J. A Rapid Bootstrap Algorithm for the RAxML Web Servers. Systematic Biology. 2008;57(5):758–71. pmid:18853362
- View Article
- PubMed/NCBI
- Google Scholar
25. Turk-Kubo KA, Karamchandani M, Capone DG, Zehr JP. The paradox of marine heterotrophic nitrogen fixation: abundances of heterotrophic diazotrophs do not account for nitrogen fixation rates in the Eastern Tropical South Pacific. Environmental Microbiology. 2014;16(10):3095–114. pmid:24286454
- View Article
- PubMed/NCBI
- Google Scholar
26. Zhang Y, Zhao Z, Sun J, Jiao N. Diversity and distribution of diazotrophic communities in the South China Sea deep basin with mesoscale cyclonic eddy perturbations. FEMS Microbiology Ecology. 2011;78(3):417–27. pmid:22066702
- View Article
- PubMed/NCBI
- Google Scholar
27. Shiozaki T, Ijichi M, Kodama T, Takeda S, Furuya K. Heterotrophic bacteria as major nitrogen fixers in the euphotic zone of the Indian Ocean. Global Biogeochemical Cycles. 2014:2014GB004886.
- View Article
- Google Scholar
28. Loescher CR, Groskopf T, Desai FD, Gill D, Schunck H, Croot PL, et al. Facets of diazotrophy in the oxygen minimum zone waters off Peru. ISME J. 2014;8(11):2180–92. pmid:24813564
- View Article
- PubMed/NCBI
- Google Scholar
29. Yang JC, Madupu R, Durkin AS, Ekborg NA, Pedamallu CS, Hostetler JB, et al. The Complete Genome of Teredinibacter turnerae T7901: An Intracellular Endosymbiont of Marine Wood-Boring Bivalves (Shipworms). PLoS ONE. 2009;4(7):e6085. pmid:19568419
- View Article
- PubMed/NCBI
- Google Scholar
30. Church MJ, Björkman KM, Karl DM, Saito MA, Zehr JP. Regional distribution of nitrogen-fixing bacteria in the Pacific Ocean. Limnology and Oceanography. 2008;53(1):63–77.
- View Article
- Google Scholar
31. Man-Aharonovich D, Kress N, Zeev EB, Berman-Frank I, Beja O. Molecular ecology of nifH genes and transcripts in the eastern Mediterranean Sea. Environ Microbiol. 2007;9(9):2354–63. Epub 2007/08/10. pmid:17686031.
- View Article
- PubMed/NCBI
- Google Scholar
32. Foster RA, Paytan A, Zehr JP. Seasonality of N2 fixation and nifH gene diversity in the Gulf of Aqaba (Red Sea). Limnology and Oceanography. 2009;54(1):219–33.
- View Article
- Google Scholar
33. Hamersley MR, Turk KA, Leinweber A, Gruber N, Zehr JP, Gunderson T, et al. Nitrogen fixation within the water column associated with two hypoxic basins in the Southern California Bight. Aquatic Microbial Ecology. 2011;63:193–205. Epub 15 April 2011.
- View Article
- Google Scholar
34. Moore CM, Mills MM, Achterberg EP, Geider R, La Roche J, Lucas MI, et al. Large-scale distribution of Atlantic nitrogen fixation controlled by iron availability. Nature Geoscience. 2009;2:867–71.
- View Article
- Google Scholar
35. Jayakumar A, Al-Rshaidat MMD, Ward BB, Mulholland MR. Diversity, distribution, and expression of diazotroph nifH genes in oxygen-deficient waters of the Arabian Sea. FEMS. 2012;82(3):597–606. pmid:22697171
- View Article
- PubMed/NCBI
- Google Scholar
36. Bonnet S, Dekaezemacker J, Turk-Kubo KA, Moutin T, Hamersley RM, Grosso O, et al. Aphotic N₂ Fixation in the Eastern Tropical South Pacific Ocean. PLoS ONE. 2013;8(12):e81265. pmid:24349048
- View Article
- PubMed/NCBI
- Google Scholar
37. Fernandez C, Farías L, Ulloa O. Nitrogen Fixation in Denitrified Marine Waters. PLoS ONE. 2011;6(6):e20539. pmid:21687726
- View Article
- PubMed/NCBI
- Google Scholar
38. Hewson I, Moisander PH, Achilles KM, Carlson CA, Jenkins BD, Mondragon EA, et al. Characteristics of diazotrophs in surface to abyssopelagic waters of the Sargasso Sea. Aquatic Microbial Ecology. 2007;46(1):15–30. ISI:000244194200002.
- View Article
- Google Scholar
39. Kolber ZS, Plumley FG, Lang AS, Beatty JT, Blankenship RE, VanDover CL, et al. Contribution of aerobic photoheterotrophic bacteria to the carbon cycle in the ocean. Science. 2001;292(5526):2492–5. ISI:000169658200042. pmid:11431568
- View Article
- PubMed/NCBI
- Google Scholar
40. Yurkov VV, Beatty JT. Aerobic anoxygenic phototrophic bacteria. Microbiol Mol Biol Rev. 1998;62(3):695–724. Epub 1998/09/08. pmid:9729607; PubMed Central PMCID: PMC98932.
- View Article
- PubMed/NCBI
- Google Scholar
41. Riemann L, Farnelid H, Steward GF. Nitrogenase genes in non-cyanobacterial plankton: prevalence, diversity and regulation in marine waters. Aquatic Microbial Ecology. 2010;61(3):235–47.
- View Article
- Google Scholar
42. Hansell DA, Carlson CA, Repeta DJ, Schlitzer R. Dissolved organic matter in the ocean: A controversy stimulates new insights. Oceanography. 2009;22(4):202–11.
- View Article
- Google Scholar
43. Berman-Frank I, Lundgren P, Chen Y-B, Kupper H, Kolber Z, Bergman B, et al. Segregation of Nitrogen Fixation and Oxygenic Photosynthesis in the Marine Cyanobacterium Trichodesmium. Science. 2001;294(5546):1534–7. pmid:11711677
- View Article
- PubMed/NCBI
- Google Scholar
44. Foster RA, Subramaniam A, Mahaffey C, Carpenter EJ, Capone D, Zehr JP. Influence of the Amazon River plume on distributions of free-living and symbiotic cyanobacteria in the western tropical north Atlantic Ocean. Limnology and Oceanography. 2007;52(2):517–32.
- View Article
- Google Scholar
45. Thompson AW, Foster RA, Krupke A, Carter BJ, Musat N, Vaulot D, et al. Unicellular Cyanobacterium Symbiotic with a Single-Celled Eukaryotic Alga. Science. 2012;337(6101):1546–50. pmid:22997339
- View Article
- PubMed/NCBI
- Google Scholar
46. Paerl HW. Microzone Formation: Its Role in the Enhancement of Aquatic N₂ fixation. Limnology and Oceanography. 1985;30(6):1246–52.
- View Article
- Google Scholar
47. Paerl HW, Prufert LE. Oxygen-Poor Microzones as Potential Sites of Microbial N₂ Fixation in Nitrogen-Depleted Aerobic Marine Waters. AEM. 1987;53(5):1078–87.
- View Article
- Google Scholar
48. Duarte CM, Regaudie-de-Gioux A, Arrieta JM, Delgado-Huertas A, Agustí S. The Oligotrophic Ocean Is Heterotrophic. Annual Review of Marine Science. 2013;5(1):551–69. pmid:22809189.
- View Article
- PubMed/NCBI
- Google Scholar
49. Poretsky RS, Hewson I, Sun S, Allen AE, Zehr JP, Moran MA. Comparative day/night metatranscriptomic analysis of microbial communities in the North Pacific subtropical gyre. Environmental Microbiology. 2009;11(6):1358–75. pmid:19207571
- View Article
- PubMed/NCBI
- Google Scholar
50. Rijkenberg MJA, Langlois RJ, Mills MM, Patey MD, Hill PG, Nielsdóttir MC, et al. Environmental Forcing of Nitrogen Fixation in the Eastern Tropical and Sub-Tropical North Atlantic Ocean. PLoS ONE. 2011;6(12):e28989.
- View Article
- Google Scholar

[ref1] 1. Postgate J. Nitrogen Fixation. 3rd ed. Boston: Cambridge University Press; 1998.

[ref2] 2. Zehr JP, Mellon MT, Zani S. New Nitrogen-Fixing Microorganisms Detected in Oligotrophic Oceans by Amplification of Nitrogenase (nifH) Genes. Applied and Environmental Microbiology. 1998;64(9):3444–50. pmid:9726895
View Article
PubMed/NCBI
Google Scholar

[3] View Article

[4] PubMed/NCBI

[5] Google Scholar

[ref3] 3. Luo YW, Doney SC, Anderson LA, Benavides M, Berman-Frank I, Bode A, et al. Database of diazotrophs in global ocean: abundances, biomass and nitrogen fixation rates. Earth Syst Sci Data. 2012;4(1):47–73.
View Article
Google Scholar

[7] View Article

[8] Google Scholar

[ref4] 4. Farnelid H, Andersson AF, Bertilsson S, Al-Soud WA, Hansen LH, Sorensen S, et al. Nitrogenase Gene Amplicons from Global Marine Surface Waters are Dominated by Genes of Non- Cyanobacteria. PLoS ONE. 2011;6(4):e19223. Epub 2011/05/12. pmid:21559425; PubMed Central PMCID: PMC3084785.
View Article
PubMed/NCBI
Google Scholar

[10] View Article

[11] PubMed/NCBI

[12] Google Scholar

[ref5] 5. Giovannoni SJ, Vergin KL. Seasonality in Ocean Microbial Communities. Science. 2012;335(6069):671–6. pmid:22323811
View Article
PubMed/NCBI
Google Scholar

[14] View Article

[15] PubMed/NCBI

[16] Google Scholar

[ref6] 6. Zavarzin GA, Stackebrandt E, Murray RG. A correlation of phylogenetic diversity in the Proteobacteria with the influences of ecological forces. Can J Microbiol. 1991;37(1):1–6. Epub 1991/01/01. pmid:1708691.
View Article
PubMed/NCBI
Google Scholar

[18] View Article

[19] PubMed/NCBI

[20] Google Scholar

[ref7] 7. Venter JC, Remington K, Heidelberg JF, Halpern AL, Rusch D, Eisen JA, et al. Environmental Genome Shotgun Sequencing of the Sargasso Sea. Science. 2004;304(5667):66–74. pmid:15001713
View Article
PubMed/NCBI
Google Scholar

[22] View Article

[23] PubMed/NCBI

[24] Google Scholar

[ref8] 8. Rusch DB, Halpern AL, Sutton G, Heidelberg KB, Williamson S, Yooseph S, et al. The Sorcerer II Global Ocean Sampling Expedition: Northwest Atlantic through Eastern Tropical Pacific. PLoS Biol. 2007;5(3):e77. pmid:17355176
View Article
PubMed/NCBI
Google Scholar

[26] View Article

[27] PubMed/NCBI

[28] Google Scholar

[ref9] 9. Dupont CL, Rusch DB, Yooseph S, Lombardo MJ, Richter RA, Valas R, et al. Genomic insights to SAR86, an abundant and uncultivated marine bacterial lineage. ISME J. 2012;6(6):1186–99. Epub 2011/12/16. pmid:22170421; PubMed Central PMCID: PMC3358033.
View Article
PubMed/NCBI
Google Scholar

[30] View Article

[31] PubMed/NCBI

[32] Google Scholar

[ref10] 10. Grimes DJ, Johnson CN, Dillon KS, Flowers AR, Noriea NF III, Berutti T. What Genomic Sequence Information Has Revealed About Vibrio Ecology in the Ocean- A Review. Microb Ecol. 2009;58:447–60. pmid:19727929
View Article
PubMed/NCBI
Google Scholar

[34] View Article

[35] PubMed/NCBI

[36] Google Scholar

[ref11] 11. Guerinot ML, West PA, Lee JV, Colwell RR. Vibrio diazotrophicus sp. nov., a Marine Nitrogen-Fixing Bacterium. International Journal of Systematic Bacteriology. 1982;32(3):350–7.
View Article
Google Scholar

[38] View Article

[39] Google Scholar

[ref12] 12. Payne WJ, Eagon RG, Williams AK. Some observations on the physiology of Pseudomonas natriegens nov. spec. Antoine van Leeuwenhoek. 1961;27(1):121–8.
View Article
Google Scholar

[41] View Article

[42] Google Scholar

[ref13] 13. Halm H, Lam P, Ferdelman TG, Lavik G, Dittmar T, LaRoche J, et al. Heterotrophic organisms dominate nitrogen fixation in the South Pacific Gyre. ISME J. 2012;6(6):1238–49. http://www.nature.com/ismej/journal/v6/n6/suppinfo/ismej2011182s1.html. pmid:22170429
View Article
PubMed/NCBI
Google Scholar

[44] View Article

[45] PubMed/NCBI

[46] Google Scholar

[ref14] 14. Langlois RJ, LaRoche J, Raab PA. Diazotrophic Diversity and Distribution in the Tropical and Subtropical Atlantic Ocean. Applied and Environmental Microbiology. 2005;71(12):7910–9. pmid:16332767
View Article
PubMed/NCBI
Google Scholar

[48] View Article

[49] PubMed/NCBI

[50] Google Scholar

[ref15] 15. Bird C, Martinez Martinez J, O'Donnell AG, Wyman M. Spatial Distribution and Transcriptional Activity of an Uncultured Clade of Planktonic Diazotrophic y-Proteobacteria in the Arabian Sea. Applied and Environmental Microbiology. 2005;71(4):2079–85. pmid:15812041
View Article
PubMed/NCBI
Google Scholar

[52] View Article

[53] PubMed/NCBI

[54] Google Scholar

[ref16] 16. Moisander PH, Beinart RA, Voss M, Zehr JP. Diversity and abundance of diazotrophic microorganisms in the South China Sea during intermonsoon. ISME J. 2008;2(9):954–67. Epub 2008/06/06. pmid:18528417.
View Article
PubMed/NCBI
Google Scholar

[56] View Article

[57] PubMed/NCBI

[58] Google Scholar

[ref17] 17. Bombar D, Moisander P, Dippner J, Foster R, Voss M, Karfeld B, et al. Distribution of diazotrophic microorganisms and nifH gene expression in the Mekong River plume during intermonsoon. Marine Ecology Progress Series. 2011;424:39–52.
View Article
Google Scholar

[60] View Article

[61] Google Scholar

[ref18] 18. Langlois RJ, Hümmer D, LaRoche J. Abundances and Distributions of the Dominant nifH Phylotypes in the Northern Atlantic Ocean. Applied and Environmental Microbiology. 2008;74(6):1922–31. pmid:18245263
View Article
PubMed/NCBI
Google Scholar

[63] View Article

[64] PubMed/NCBI

[65] Google Scholar

[ref19] 19. Moisander PH, Serros T, Paerl RW, Beinart RA, Zehr JP. Gammaproteobacterial diazotrophs and nifH gene expression in surface waters of the South Pacific Ocean. ISME J. 2014;8(10):1962–73. pmid:24722632
View Article
PubMed/NCBI
Google Scholar

[67] View Article

[68] PubMed/NCBI

[69] Google Scholar

[ref20] 20. Church MJ, Jenkins BD, Karl DM, Zehr JP. Vertical distributions of nitrogen-fixing phylotypes at Stn ALOHA in the oligotrophic North Pacific Ocean. Aquatic Microbial Ecology. 2005;38(1):3–14. ISI:000226947900002.
View Article
Google Scholar

[71] View Article

[72] Google Scholar

[ref21] 21. Zehr JP, Crumbliss LL, Church MJ, Omoregie EO, Jenkins BD. Nitrogenase genes in PCR and RT-PCR reagents: implications for studies of diversity of functional genes. BioTechniques. 2003;35:996–1005. pmid:14628674
View Article
PubMed/NCBI
Google Scholar

[74] View Article

[75] PubMed/NCBI

[76] Google Scholar

[ref22] 22. Huang Y, Niu B, Gao Y, Fu L, Li W. CD-HIT Suite: a web server for clustering and comparing biological sequences. Bioinformatics. 2010;26(5):680–2. pmid:20053844
View Article
PubMed/NCBI
Google Scholar

[78] View Article

[79] PubMed/NCBI

[80] Google Scholar

[ref23] 23. Waterhouse AM, Procter JB, Martin DMA, Clamp M, Barton GJ. Jalview Version 2—a multiple sequence alignment editor and analysis workbench. Bioinformatics. 2009;25(9):1189–91. pmid:19151095
View Article
PubMed/NCBI
Google Scholar

[82] View Article

[83] PubMed/NCBI

[84] Google Scholar

[ref24] 24. Stamatakis A, Hoover P, Rougemont J. A Rapid Bootstrap Algorithm for the RAxML Web Servers. Systematic Biology. 2008;57(5):758–71. pmid:18853362
View Article
PubMed/NCBI
Google Scholar

[86] View Article

[87] PubMed/NCBI

[88] Google Scholar

[ref25] 25. Turk-Kubo KA, Karamchandani M, Capone DG, Zehr JP. The paradox of marine heterotrophic nitrogen fixation: abundances of heterotrophic diazotrophs do not account for nitrogen fixation rates in the Eastern Tropical South Pacific. Environmental Microbiology. 2014;16(10):3095–114. pmid:24286454
View Article
PubMed/NCBI
Google Scholar

[90] View Article

[91] PubMed/NCBI

[92] Google Scholar

[ref26] 26. Zhang Y, Zhao Z, Sun J, Jiao N. Diversity and distribution of diazotrophic communities in the South China Sea deep basin with mesoscale cyclonic eddy perturbations. FEMS Microbiology Ecology. 2011;78(3):417–27. pmid:22066702
View Article
PubMed/NCBI
Google Scholar

[94] View Article

[95] PubMed/NCBI

[96] Google Scholar

[ref27] 27. Shiozaki T, Ijichi M, Kodama T, Takeda S, Furuya K. Heterotrophic bacteria as major nitrogen fixers in the euphotic zone of the Indian Ocean. Global Biogeochemical Cycles. 2014:2014GB004886.
View Article
Google Scholar

[98] View Article

[99] Google Scholar

[ref28] 28. Loescher CR, Groskopf T, Desai FD, Gill D, Schunck H, Croot PL, et al. Facets of diazotrophy in the oxygen minimum zone waters off Peru. ISME J. 2014;8(11):2180–92. pmid:24813564
View Article
PubMed/NCBI
Google Scholar

[101] View Article

[102] PubMed/NCBI

[103] Google Scholar

[ref29] 29. Yang JC, Madupu R, Durkin AS, Ekborg NA, Pedamallu CS, Hostetler JB, et al. The Complete Genome of Teredinibacter turnerae T7901: An Intracellular Endosymbiont of Marine Wood-Boring Bivalves (Shipworms). PLoS ONE. 2009;4(7):e6085. pmid:19568419
View Article
PubMed/NCBI
Google Scholar

[105] View Article

[106] PubMed/NCBI

[107] Google Scholar

[ref30] 30. Church MJ, Björkman KM, Karl DM, Saito MA, Zehr JP. Regional distribution of nitrogen-fixing bacteria in the Pacific Ocean. Limnology and Oceanography. 2008;53(1):63–77.
View Article
Google Scholar

[109] View Article

[110] Google Scholar

[ref31] 31. Man-Aharonovich D, Kress N, Zeev EB, Berman-Frank I, Beja O. Molecular ecology of nifH genes and transcripts in the eastern Mediterranean Sea. Environ Microbiol. 2007;9(9):2354–63. Epub 2007/08/10. pmid:17686031.
View Article
PubMed/NCBI
Google Scholar

[112] View Article

[113] PubMed/NCBI

[114] Google Scholar

[ref32] 32. Foster RA, Paytan A, Zehr JP. Seasonality of N2 fixation and nifH gene diversity in the Gulf of Aqaba (Red Sea). Limnology and Oceanography. 2009;54(1):219–33.
View Article
Google Scholar

[116] View Article

[117] Google Scholar

[ref33] 33. Hamersley MR, Turk KA, Leinweber A, Gruber N, Zehr JP, Gunderson T, et al. Nitrogen fixation within the water column associated with two hypoxic basins in the Southern California Bight. Aquatic Microbial Ecology. 2011;63:193–205. Epub 15 April 2011.
View Article
Google Scholar

[119] View Article

[120] Google Scholar

[ref34] 34. Moore CM, Mills MM, Achterberg EP, Geider R, La Roche J, Lucas MI, et al. Large-scale distribution of Atlantic nitrogen fixation controlled by iron availability. Nature Geoscience. 2009;2:867–71.
View Article
Google Scholar

[122] View Article

[123] Google Scholar

[ref35] 35. Jayakumar A, Al-Rshaidat MMD, Ward BB, Mulholland MR. Diversity, distribution, and expression of diazotroph nifH genes in oxygen-deficient waters of the Arabian Sea. FEMS. 2012;82(3):597–606. pmid:22697171
View Article
PubMed/NCBI
Google Scholar

[125] View Article

[126] PubMed/NCBI

[127] Google Scholar

[ref36] 36. Bonnet S, Dekaezemacker J, Turk-Kubo KA, Moutin T, Hamersley RM, Grosso O, et al. Aphotic N₂ Fixation in the Eastern Tropical South Pacific Ocean. PLoS ONE. 2013;8(12):e81265. pmid:24349048
View Article
PubMed/NCBI
Google Scholar

[129] View Article

[130] PubMed/NCBI

[131] Google Scholar

[ref37] 37. Fernandez C, Farías L, Ulloa O. Nitrogen Fixation in Denitrified Marine Waters. PLoS ONE. 2011;6(6):e20539. pmid:21687726
View Article
PubMed/NCBI
Google Scholar

[133] View Article

[134] PubMed/NCBI

[135] Google Scholar

[ref38] 38. Hewson I, Moisander PH, Achilles KM, Carlson CA, Jenkins BD, Mondragon EA, et al. Characteristics of diazotrophs in surface to abyssopelagic waters of the Sargasso Sea. Aquatic Microbial Ecology. 2007;46(1):15–30. ISI:000244194200002.
View Article
Google Scholar

[137] View Article

[138] Google Scholar

[ref39] 39. Kolber ZS, Plumley FG, Lang AS, Beatty JT, Blankenship RE, VanDover CL, et al. Contribution of aerobic photoheterotrophic bacteria to the carbon cycle in the ocean. Science. 2001;292(5526):2492–5. ISI:000169658200042. pmid:11431568
View Article
PubMed/NCBI
Google Scholar

[140] View Article

[141] PubMed/NCBI

[142] Google Scholar

[ref40] 40. Yurkov VV, Beatty JT. Aerobic anoxygenic phototrophic bacteria. Microbiol Mol Biol Rev. 1998;62(3):695–724. Epub 1998/09/08. pmid:9729607; PubMed Central PMCID: PMC98932.
View Article
PubMed/NCBI
Google Scholar

[144] View Article

[145] PubMed/NCBI

[146] Google Scholar

[ref41] 41. Riemann L, Farnelid H, Steward GF. Nitrogenase genes in non-cyanobacterial plankton: prevalence, diversity and regulation in marine waters. Aquatic Microbial Ecology. 2010;61(3):235–47.
View Article
Google Scholar

[148] View Article

[149] Google Scholar

[ref42] 42. Hansell DA, Carlson CA, Repeta DJ, Schlitzer R. Dissolved organic matter in the ocean: A controversy stimulates new insights. Oceanography. 2009;22(4):202–11.
View Article
Google Scholar

[151] View Article

[152] Google Scholar

[ref43] 43. Berman-Frank I, Lundgren P, Chen Y-B, Kupper H, Kolber Z, Bergman B, et al. Segregation of Nitrogen Fixation and Oxygenic Photosynthesis in the Marine Cyanobacterium Trichodesmium. Science. 2001;294(5546):1534–7. pmid:11711677
View Article
PubMed/NCBI
Google Scholar

[154] View Article

[155] PubMed/NCBI

[156] Google Scholar

[ref44] 44. Foster RA, Subramaniam A, Mahaffey C, Carpenter EJ, Capone D, Zehr JP. Influence of the Amazon River plume on distributions of free-living and symbiotic cyanobacteria in the western tropical north Atlantic Ocean. Limnology and Oceanography. 2007;52(2):517–32.
View Article
Google Scholar

[158] View Article

[159] Google Scholar

[ref45] 45. Thompson AW, Foster RA, Krupke A, Carter BJ, Musat N, Vaulot D, et al. Unicellular Cyanobacterium Symbiotic with a Single-Celled Eukaryotic Alga. Science. 2012;337(6101):1546–50. pmid:22997339
View Article
PubMed/NCBI
Google Scholar

[161] View Article

[162] PubMed/NCBI

[163] Google Scholar

[ref46] 46. Paerl HW. Microzone Formation: Its Role in the Enhancement of Aquatic N₂ fixation. Limnology and Oceanography. 1985;30(6):1246–52.
View Article
Google Scholar

[165] View Article

[166] Google Scholar

[ref47] 47. Paerl HW, Prufert LE. Oxygen-Poor Microzones as Potential Sites of Microbial N₂ Fixation in Nitrogen-Depleted Aerobic Marine Waters. AEM. 1987;53(5):1078–87.
View Article
Google Scholar

[168] View Article

[169] Google Scholar

[ref48] 48. Duarte CM, Regaudie-de-Gioux A, Arrieta JM, Delgado-Huertas A, Agustí S. The Oligotrophic Ocean Is Heterotrophic. Annual Review of Marine Science. 2013;5(1):551–69. pmid:22809189.
View Article
PubMed/NCBI
Google Scholar

[171] View Article

[172] PubMed/NCBI

[173] Google Scholar

[ref49] 49. Poretsky RS, Hewson I, Sun S, Allen AE, Zehr JP, Moran MA. Comparative day/night metatranscriptomic analysis of microbial communities in the North Pacific subtropical gyre. Environmental Microbiology. 2009;11(6):1358–75. pmid:19207571
View Article
PubMed/NCBI
Google Scholar

[175] View Article

[176] PubMed/NCBI

[177] Google Scholar

[ref50] 50. Rijkenberg MJA, Langlois RJ, Mills MM, Patey MD, Hill PG, Nielsdóttir MC, et al. Environmental Forcing of Nitrogen Fixation in the Eastern Tropical and Sub-Tropical North Atlantic Ocean. PLoS ONE. 2011;6(12):e28989.
View Article
Google Scholar

[179] View Article

[180] Google Scholar

Figures

Abstract

Introduction

Materials and Methods

Sample collection, genetic material isolation and cDNA synthesis

Abundance estimations using qPCR

Sequence analysis

Statistical analysis

Results and Discussion

Phylogenetic analysis of γ-proteobacterial nifH

Distribution of Gamma A nifH phylotypes and transcripts

Environmental Niche for Gamma A

Conclusions

Supporting Information

S1 Alignment. Alignment of nifH sequences used in Fig 1.

S2 Alignment. Alignment of nifH sequences used in Fig 2.

S3 Alignment. Alignment of nifH sequences used in Fig 3.

S1 Fig. Phylogenetic tree of nifH from cultured organisms in relation with nifH OTU’s from environmental sequences used as qPCR target phylotypes and nifH contaminants of PCR Reagents.

S2 Fig. Dot plot alignment of γ–proteobacterial nifH amplicons used as targets in qPCR studies.

S3 Fig. Plot of Gamma A cDNA:DNA versus NO3 (μM, a), PO4 (μM, b), and SiO2 (μM, c) concentrations.

S4 Fig. Linear maximum likelihood tree of γ-proteobacterial nifH OTUs.

S1 Table. List of accession numbers for sequences identified as γ-proteobacteria.

S2 Table. Descriptive statistics summary of PCA clusters.

S3 Table. Detection Frequency of Gamma A in qPCR Analysis.

S4 Table. Sampling frequency arranged by latitudinal bands.

S5 Table. Designation of sequences in the Marine 1 clade into OTUs at 100% similarity.

S6 Table. Designation of sequences in the Marine 1 clade into OTUs at 99% similarity.

S7 Table. Designation of sequences in the Marine 1 clade into OTUs at 97% similarity.

Acknowledgments

Author Contributions

References

S3 Fig. Plot of Gamma A cDNA:DNA versus NO₃ (μM, a), PO₄ (μM, b), and SiO₂ (μM, c) concentrations.