Previous Article | Next Article ![]()
Applied and Environmental Microbiology, February 2004, p. 765-770, Vol. 70, No. 2
0099-2240/04/$08.00+0 DOI: 10.1128/AEM.70.2.765-770.2004
Copyright © 2004, American Society for Microbiology. All Rights Reserved.
Marine Sciences Research Center, Stony Brook University, Stony Brook, New York 11794-5000,1 Romberg Tiburon Center for Environmental Studies, San Francisco State University, Tiburon, California 94920,2 Conservation Genetics Laboratory, Biology Department, San Francisco State University, San Francisco, California 94132,3 Botanical Institute, Stockholm University, S-106-91 Stockholm, Sweden,4 Wrigley Institute for Environmental Studies, University of Southern California, Los Angeles, California 900895
Received 11 July 2003/ Accepted 4 November 2003
|
|
|---|
and
) and unicellular cyanobacteria are common in both the tropical North Atlantic and Pacific oceans. In near-surface waters proteobacterial nifH transcripts were present during both night and day while unicellular cyanobacterial nifH transcripts were present during the nighttime only, suggesting separation of N2 fixation and photosynthesis by unicellular cyanobacteria. Phylogenetic relationships among unicellular cyanobacteria from both oceans were determined after sequencing of a conserved region of 16S ribosomal DNA (rDNA) of cyanobacteria, and results showed that they clustered together, regardless of the ocean of origin. However, sequencing of nifH transcripts of unicellular cyanobacteria from both oceans showed that they clustered separately. This suggests that unicellular cyanobacteria from the tropical North Atlantic and subtropical North Pacific share a common ancestry (16S rDNA) and that potential unicellular N2 fixers have diverged (nifH). N2 fixation rates for unicellular bacterioplankton (including small cyanobacteria) from both oceans were determined in situ according to the acetylene reduction and 15N2 protocols. The results showed that rates of fixation by bacterioplankton can be almost as high as those of fixation by the colonial N2-fixing marine cyanobacteria Trichodesmium spp. in the tropical North Atlantic but that rates are much lower in the subtropical North Pacific. |
|
|---|
Previous research has identified the presence of fragments of the unicellular cyanobacterial and proteobacterial nifH genes (which encode one of the peptide molecules that form the dinitrogenase reductase subunit of nitrogenase, the enzyme responsible for N2 fixation [31]) in the subtropical North Pacific and tropical North Atlantic oceans (16, 43, 44). Preliminary data on N2 fixation rates for the 10- to 0.2-µm-size fraction of the bacterioplankton have suggested that they could make an important contribution to the global N cycle (14, 44). These studies used 15N2 to measure 24-h N2 fixation rates in 10-µm-pore-size-prefiltered water from the subtropical North Pacific. Also, N2 fixation rates were estimated based on unicellular phytoplankton cell numbers that came from literature values (6). Cyanobacteria are autotrophs and generate ATP necessary for N2 fixation through photosynthesis; unicellular cyanobacteria that fix N2 need to do so during the night (2) since nitrogenase, the enzyme that catalyzes N2 fixation, is inhibited by O2 (15). Thus, it is important to obtain nitrogenase activity rates for the daytime and the nighttime.
In order to better define the in situ rates of N2 fixation for the unicellular bacterioplankton community in the tropical North Atlantic and subtropical North Pacific oceans, we sampled at different depths throughout diel cycles for different seasons by the acetylene reduction and 15N2 protocols with slight variations (7, 10). Further, we coupled the N2 fixation activity to the phylogenetic characterization of the populations responsible throughout the analysis of nifH transcripts. Here we demonstrate that potential N2-fixing unicellular bacterioplankton differ between the tropical North Atlantic and North Pacific; we also show that nifH transcripts of unicellular cyanobacteria occur during the nighttime only while those for proteobacteria are present during both the day and the night. Major differences are also found in N2 fixation rates between the two oceans; in the tropical North Atlantic, the unicellular bacterioplankton activity is almost as high as that of the major oceanic N2 fixers Trichodesmium spp. but in the tropical North Pacific is much lower. Our findings provide unique and new information essential to understanding N and C cycles in the Atlantic and Pacific oceans.
|
|
|---|
Nitrogenase activity.
The filtrate from each light depth (4 liters) was concentrated onto 0.2-µm-pore-size membranes (47-mm diameter; polycarbonate PCTE; Millipore). Cell concentrates were diluted by washing the filters with 0.2-µm-pore-size-filtered seawater, to the volume necessary to assay nitrogenase activity by the 15N2 and acetylene reduction protocols (7, 10). Each 0.2-µm-pore-size membrane was screened under the epifluorescence microscope to verify that the cells had been washed off. Final enrichments were about 130- to 160-fold. 15N2 experiments were carried out during the summer 2001 cruise with 14 ml of cell concentrate (no headspace left in 14-ml vials) during 24 h and incubated with 9 µl of 15N2 (data from three stations, n = 3 per depth per station). Experiments were ended by filtering the contents of each vial onto a combusted glass fiber filter that was folded and left to dry in an oven at 60°C before being analyzed with a mass spectrometer. During the spring 2002 (Atlantic) and fall 2002 (Pacific) cruises, the diel and depth variations of N2 fixation were examined by the acetylene reduction procedure (n = 18 per depth [Atlantic]; n = 24 per depth [Pacific]). Experiments were started with injection of acetylene into serum vials (14 ml) which contained 10 ml of cell concentrate; time zero values were registered immediately after. Samples were run for 24 h, and ethylene readings were carried out approximately every 4 to 6 h, preceded by ethylene standard quantifications. Twenty-four-hour N2 fixation rates were integrated over the water column depth for the times when nitrogenase activity was observed. Each depth experimental sample (n = 3) was left inside a bag that mimicked its corresponding light level in a flowing seawater on-deck incubator. Control assays were carried out in parallel with 0.2-µm-pore-size-filtered seawater.
Cloning and sequencing.
After completion of the 24-h nitrogenase activity experiments, daytime and nighttime samples were filtered onto 0.2-µm-pore-size Durapore membrane filters (25-mm diameter; Millipore) for RNA extraction (41) with an RNeasy kit; residual amounts of DNA were removed using an RNase-Free DNase set (Qiagen, Valencia, Calif.). An Access reverse transcriptase PCR kit (Promega, Madison, Wis.) was used to amplify cDNA for nifH genes (357-bp fragment) (42). The reverse transcription reaction was carried out for 30 min using 1 µM primer nifH 3, 28 µl of H2O, 10 µl of 5x avian myeloblastosis virus buffer, 1 µl deoxynucleoside triphosphates (10 mM each), 1 µl of avian myeloblastosis virus reverse transcriptase, and 1 µl of DNA-free RNA (41). Four liters of prefiltered water was collected onto 0.2-µm-pore-size Durapore membrane filters (25-mm diameter; Millipore), and DNA was extracted according to standard protocols (26, 33). Primers (CYA 359F and 781b) that amplify a conserved 16S ribosomal DNA (rDNA) cyanobacterial region (approximately 422-bp fragment) (29) were used in a typical PCR. Purified nifH and 16S rDNA products were ligated with the pGEM-T Easy vector system (Promega) (33). Reaction mixtures (50-µl final volume) contained the following: 33.5 µl of RNA-DNA-free water, 5 µl of PCR buffer, 5 µl (16S rDNA) or 8 µl (nifH) of MgCl2 (25 mM), 1 µl of deoxynucleoside triphosphate mix, 2 µl of each primer (12.5 µM) (29, 42), 1 µl of template (cDNA for nifH), and 0.5 µl of Taq DNA polymerase (Promega). The thermal cycle for amplification of 16S rDNA was as follows: 94°C for 1 min, 60°C for 1 min, and 72°C for 1 min for 35 cycles. The thermal cycle for amplification of nifH was as follows: 94°C for 1 min, 55°C for 1 min, and 72°C for 1 min for 30 cycles. Purified amplified clone products were cycle sequenced with an ABI Prism BigDye kit (Perkin-Elmer, Foster City, Calif.).
Phylogenetic reconstruction analysis using PAUP*4.0b8 (Sinauer Associates Inc., Sunderland, Mass.) included parsimony bootstrapping (250 replicates, random addition initial trees). Distance- and maximum-likelihood-based analysis (data not shown) gave topologies nearly identical to that of the parsimony analysis. The resulting 72-taxon nifH data set included 217 parsimony-informative characters. The resulting 32-taxon 16S rDNA summarized data set (1) included 110 parsimony-informative characters.
Cell counts.
Unicellular cyanobacteria cell counts were done on board within an hour of collection, using an epifluorescence compound microscope (Zeiss Axioscope) at 1,000x. Seawater filtrates were collected onto 0.2-µm-pore-size PCTE membranes (47-mm diameter; Millipore) and placed in microscope slides for observation. Blue and green excitations were used to observe phycoerythrin and chlorophyll a epifluorescence signals, respectively (25).
Nucleotide sequence accession number.
The sequences generated in this study have been deposited in the GenBank database under the indicated accession numbers: tropical North Atlantic nifH, AF536983 to AF536986 and AY191972 to AY191976; tropical North Atlantic 16S rDNA, AY191920, AY191921, and AY191923 to AY191942; subtropical North Pacific nifH, AY191943 to AY191966, AY191969, and AY191971; subtropical North Pacific 16S rDNA, AY191919 and AY191904 to AY191922.
|
|
|---|
-proteobacterium-like sequences. Five unique sequences from the subtropical North Pacific (10 clones out of a 34-clone library) clustered with unicellular cyanobacteria, which had been seen only in the North Pacific in May (44). Most of the Pacific nifH cDNA clone sequences (including all of those obtained from 100-m samples from incubations from 1200 to 1800 h for activity) clustered with
-proteobacteria, and one unique sequence clustered with
-proteobacteria. We cannot assign the observed nitrogenase activity with certainty to one group of bacteria; nevertheless, we obtained only
-proteobacterium-type nifH transcripts from daytime incubations in the subtropical North Pacific;
- and
-proteobacteria and unicellular cyanobacterium-like nifH transcripts were obtained during the nighttime incubations (both oceans). The 16S rDNA cyanobacterial sequences from both oceans clustered together and with other proposed N2-fixing unicellular cyanobacteria (Fig. 2) (Synechocystis sp. strain WH8501). This field observation is consistent with previous studies that have proposed that N2-fixing unicellular cyanobacteria obtain ATP necessary for N2 fixation during the day through C fixation and temporally separate O2 production from nitrogenase activity (17, 23, 27, 28, 32, 35, 36, 37). Also, since proteobacteria are heterotrophic, their N2 fixation activity is not controlled by the day-night cycles of C fixation. Microanaerobic environments can occur in the surroundings of cells that are respiring, which could provide the environments adequate to carry out N2 fixation during the day (19, 30). In any event, the in situ N2 fixation rates measured were greater during the nighttime, and in fact, only the subtropical North Pacific showed a small daytime N2 fixation activity. Nevertheless, even though the obtained cyanobacterial nifH transcripts and 16S rDNA sequences clustered together with those from known N2-fixing unicellular cyanobacteria (Synechocystis sp. strain WH8501 and Cyanothece strain ATCC 51142), we cannot assume that unicellular cyanobacteria analyzed in this study were N2 fixers. In fact, since the 16S rDNA region that we sequenced is only approximately 422 bp in length, it is not yet clear how closely related populations from the Atlantic and Pacific really are.
![]() View larger version (35K): [in a new window] |
FIG. 1. Phylogenetic reconstruction (maximum parsimony) for nifH gene sequences from bacterioplankton in the tropical North Atlantic and subtropical North Pacific oceans. Asterisks indicate sequences obtained as part of this study. Daytime and nighttime transcripts are indicated by open and solid circles, respectively.
|
![]() View larger version (31K): [in a new window] |
FIG. 2. Summary of phylogenetic reconstruction (maximum parsimony) for 16S rDNA sequences from unicellular cyanobacteria in the tropical North Atlantic and subtropical North Pacific oceans. Asterisks indicate sequences obtained as part of this study.
|
In situ rates of N2 fixation.
We measured unicellular bacterioplankton N2 fixation in the mixed surface water layer of the tropical North Atlantic in the summer of 2001 (when we observed a strong thermocline and nutricline [http://biology.usc.edu/bc/]) and during spring 2002 (which again showed a stable surface water mixed layer [http://biology.usc.edu/bc/]). Nitrogenase activity was observed only during the night both in the summer of 2001 and in the spring of 2002 (Fig. 3). Integrated water column rates for 10 h of nighttime nitrogenase activity in the tropical North Atlantic averaged 47 µmol of N m-2 day-1 in summer 2001 and 37 µmol of N m-2 day-1 in spring 2002. Rates of N2 fixation by Trichodesmium spp. were measured in parallel with our experiments in the tropical North Atlantic during the summer 2001 cruise (D. G. Capone, personal communication). For that cruise, the average rate for 30 stations was 62 ± 21 (standard error) µmol of N m-2 day-1. A cruise in the same vicinity in February 2001 found average rates of 167 ± 49 (standard error) µmol of N m-2 day-1 (n = 23) (D. G. Capone, personal communication). Thus, unicellular bacterioplankton N2 fixation activity at several stations in the tropical North Atlantic was about equivalent to that of Trichodesmium spp. in the summer of 2001 and about 20% of Trichodesmium spp. N2 fixation activity in the spring of 2002.
![]() View larger version (13K): [in a new window] |
FIG. 3. Distribution of average rates of in situ N2 fixation (picomoles of N liter-1 hour-1) by bacterioplankton by depth for the tropical North Atlantic (solid symbols) and subtropical North Pacific (open symbols). 15N2 24-h incubations in summer 2001 are shown by . Symbols indicate periods of incubations and reading time in gas chromatography: , 1800 to 2200 h; , 2200 to 0200 h; and , 0200 to 0600 h for spring 2002, tropical North Atlantic; , 1200 to 1800 h; , 1800 to 2400 h; and , 2400 to 0600 h for fall 2002, subtropical North Pacific. Error bars represent standard errors. Only readings where nitrogenase activity was detected are shown.
|
Unicellular cyanobacterial cell abundances.
Unicellular cyanobacterial abundances were an order of magnitude higher in the Atlantic than in the Pacific, peaking from surface to 25 m in both oceans (Fig. 4); cells were larger in the Pacific (3 to 7 µm) than in the Atlantic (2.5 µm). The cell abundances reported here included all unicellular cyanobacteria (2.5 µm, Atlantic; 3 to 7 µm, Pacific), both the potential N2 fixers and the non-N2 fixers. The fluorochrome system used (24) does not work on unicellular cyanobacteria since these have a similar fluorescent signal with or without a positive in situ immunolocalization of cells expressing nitrogenase.
![]() View larger version (12K): [in a new window] |
FIG. 4. Distribution of average cell counts of unicellular cyanobacteria by depth, in the tropical North Atlantic (solid symbols; 2.5 µm) and subtropical North Pacific (open symbols; 3 and 7 µm) oceans. Error bars represent standard errors.
|
Our data suggest a trend of increased unicellular cyanobacterial abundance and N2 fixation rates in the tropical North Atlantic compared to those in the subtropical North Pacific (Fig. 3 and 4); this is an interesting observation, since it has been suggested that unicellular diazotrophic cyanobacteria might be major diazotrophs in oligotrophic marine ecosystems (44).
We hypothesize that differences in N2 fixation rates of unicellular bacterioplankton and cell abundances of coccoid cyanobacteria (>2.0 µm) observed between oceanic basins are in part due to the larger eolian Fe flux to the North Atlantic (0.2 to 0.8 µmol of Fe m-2 day-1) (40) and to the high Fe/P ratio present in this area, which have been suggested (4, 5, 38) to favor N2 fixation in comparison to the North Pacific Ocean (20) (0.08 to 0.16 µmol of Fe m-2 day-1) (40). Nevertheless, to date, Fe cell quotas for proteobacterial and unicellular cyanobacterial diazotrophs are needed, as are field measurements of the Fe/C ratio of proteobacterial and unicellular cyanobacterial diazotrophs (22). The tropical North Pacific (20, 21) has low light levels and deeper mixed layers that show a pattern of diminished N2 fixation rates (14). In the summer, the thermocline reaches its maximum in tropical oceans, bringing stability and increasing surface water temperatures, which could have a direct effect on the diazotrophic activity of unicellular bacterioplankton.
The magnitude of N2 fixation by small unicellular bacterioplankton, the stability of their populations over time, and the differences found between the tropical North Atlantic and the subtropical North Pacific (3, 5, 8, 13, 18, 20, 21, 34, 38, 40) suggest that the marine N cycle of these spatially extensive ecosystems has more components, and is more complex, than previously thought.
|
|
|---|
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»