Strain and Cultivation

An axenic strain of the marine pelagic diatom T. weissflogii (CCMP 1336) was obtained from the Provasoli-Guillard National Center for Marine Algae and Microbiota (NCMA; formerly CCMP). The diatoms were cultured in F/2 medium plus silicate [27] prepared with filtered (0.45 μm) and autoclaved North Sea seawater (salinity 35). The cultivation temperature was 15°C, the light:dark cycle was 10:14 h, and the light intensity was 160 μmol photons m⁻² s⁻¹. T. weissflogii was frequently checked for possible contaminations with bacteria by careful phase-contrast microscopy and by plating out subsamples of the cultures on nutrient agar plates. Additionally, all T. weissflogii cultures used in the experiments were checked by DAPI staining of cell suspensions immobilized on polycarbonate membrane filters (0.2 μm; Osmonics). A contamination of the diatom strain with prokaryotes was never detected.

Consumption and Production of Inorganic N-compounds in Dark/Anoxic versus Light/Oxic Conditions

The time courses of intracellular NO₃^- concentrations in T. weissflogii, and NO₃^- and NH₄⁺ concentrations in the growth medium under dark/anoxic versus light/oxic conditions were followed in a non-labeling experiment and a ¹⁵N-stable isotope labeling experiment (see below). For the non-labeling experiment, the cells were washed with sterile NaCl (salinity 35) and centrifuged (10 min at 1000g) three times to remove NO₃^- from the medium, and transferred into NO₃^--free artificial seawater. The cell number was determined (see below), and the experiment was started by dividing the culture for (a) the dark/anoxic incubation, and (b) the light/oxic control. For the dark/anoxic incubation, 20 mL of the diatom suspension was transferred into a dark serum bottle (wrapped in aluminum foil), flushed with N₂ for 20 min to remove O₂, sealed with a gas-tight rubber stopper, and incubated at 15°C. For the light/oxic control, the culture was kept under light/oxic culture conditions (see above). At time intervals of 0, 1, 2, 3, 4, 5, 6, and 7 h, 2 mL diatom suspension each was taken and transferred into a sample tube for centrifugation (10 min at 1000g). To assure anoxia, the dark serum bottle was flushed with N₂ after each sampling for 2 min. NO₃^- and (non-labeled) NH₄⁺ were determined in the cell-free supernatant and the diatom pellet was used for measurements of intracellular NO₃^-.

NO₃^- was measured with an NO_x analyzer connected to a reaction chamber (CLD 66s plus a Liquid NO Setup; EcoPhysics). In the reaction chamber, acidified VCl₃ (0.1 M) reduces NO₃^- plus NO₂^- to NO at 90°C, which is then measured by a chemiluminescence detector [28]. If not noted differently, the results of the NO₃^- plus NO₂^- analyses are reported as NO₃^- concentrations throughout, because NO₂^- concentrations were << NO₃^- concentrations. For intracellular NO₃^- measurements, the diatom pellet was directly injected into the reaction chamber where cells burst and release the stored NO₃^-. Intracellular NO₃^- concentrations were calculated from the difference of NO₃^- concentrations in the medium and the cell pellet, the cell numbers in the pellet, and the average cell volume of 1.22 pL [8]. Cell numbers were counted in a Fuchs-Rosenthal counting chamber with phase-contrast microscopy at 400× magnification. The total cell number in the medium was used to calculate the total intracellular NO₃^- concentration per volume of medium from the cell-specific intracellular NO₃^- concentration. Ammonium was measured by photometric absorbance determination at λ = 640 nm with a Genesys 10S spectrophotometer (Thermo Scientific; USA) following the sodium-nitroprusside-catalyzed reaction of NH₄⁺ ions with salicylate and hypochlorite [29].

Final Products of Dissimilatory Nitrate Reduction

The time courses of intracellular NO₃^- consumption and the possible products of dissimilatory NO₃^- reduction, i.e. NH₄⁺ for DNRA and N₂ for complete denitrification, were investigated with a ¹⁵N-stable isotope labeling experiment. Prior to the experiment, the (non-labeled) intracellular NO₃^- pools of T. weissflogii were depleted by a starvation procedure. The cells were separated from the NO₃^--containing culture medium via gentle centrifugation (10 min at 1000g), transferred into NO₃^--free artificial seawater [30], and exposed to dark/anoxic conditions for six days. After this pre-incubation, (non-labeled) intracellular NO₃^- had been completely consumed. For the subsequent storage of intracellular ¹⁵N-labeled NO₃^- (98 atom%; Cambridge Isotope Laboratories), the NO₃^--starved cells were harvested, re-inoculated into sterile, ¹⁵NO₃^--containing F/2 medium plus silicate in artificial seawater, and cultured under optimal growth conditions for three days (see above). The cells were then washed via gentle centrifugation with sterile NO₃^--free artificial seawater (salinity 35; 10 min at 1000g) to remove ¹⁵NO₃^- from the medium, and transferred into NO₃^--free artificial seawater enriched with 200 μM Na-acetate and 25 μM non-labeled NH₄⁺. Thus, the only NO₃^- source during the ¹⁵N-stable isotope labeling experiment was ¹⁵NO₃^- stored intracellularly by the diatoms. The cell density was obtained and the experiment was started by dividing the culture for (a) the dark/anoxic incubation, and (b) the light/oxic control. For the dark/anoxic incubation, ca. 200 mL of the diatom suspension was transferred into a dark bottle (wrapped in aluminum foil) and flushed with He for 30 min to remove O₂ and then transferred into 24 replicate 6 mL Labco-exetainers® wrapped in aluminum foil. At time intervals of 1, 2, 3, 4, 5, 6, 8, and 10 h, a He headspace of 3 mL was set in three Labco-exetainers® each, and the diatom cells in the remaining 3 mL were killed with 100 µL ZnCl₂ (50%). The Labco-exetainers® were stored upside down at room temperature until measurement of ¹⁵N-labeled N₂ by gas chromatography-isotope ratio mass spectrometry (GC-IRMS, VG Optima; Isotech). The cell suspension collected during setting the headspace was filled into 15-mL tubes and centrifuged (10 min at 1000g). Part of the cell-free supernatant was used for immediately measuring the extracellular NO₃^- concentrations, while the pellet was used for intracellular NO₃^- determination (see above). Further, 1 mL cell-free supernatant was frozen at -20°C until ¹⁵NH₄⁺ analysis using the hypobromite assay [31], followed by N₂-¹⁵N analysis using GC-IRMS. The hypobromite assay actually measures the sum of ¹⁵NH₄⁺ and ¹⁵N-labeled volatile N compounds such as methyl amines [32]. For the light/oxic control, the culture was kept under light/oxic conditions, and at time intervals of 0, 1, 2, 5, and 10 h, 3 mL cell material each was taken and processed exactly like the material that was obtained during setting the headspace in the dark/anoxic treatment. The sample collected at time point zero was used for both, the dark/anoxic incubation and the light/oxic control.

Intermediates and By-Products of Dissimilatory Nitrate Reduction

The time courses of N₂O and NO₂^- as possible intermediates or by-products of dissimilatory NO₃^- reduction were measured during the ¹⁵N-stable isotope labeling experiment. Nitrous oxide was measured in the headspace of the Labco-exetainers® from the dark/anoxic incubation experiment after N₂-¹⁵N analysis had been completed (see above; the gas volume removed for N₂-¹⁵N measurements was taken into account for the subsequent calculation of N₂O concentrations), using a GC 7890 (Agilent Technologies) equipped with a CP-PoraPLOT Q column and a ⁶³Ni electron capture detector. Nitrite was determined in the supernatant of the medium from the dark/anoxic and the light/oxic incubation with an NO_x analyzer as described for the NO₃^- determination, except that the reaction chamber contained acidified NaI (2 M) that reduces NO₂^- to NO at 20°C.

Degradation of Photopigments in Response to Dark/Anoxic Conditions

Chlorophyll a and fucoxanthin were determined in cultures of T. weissflogii that were first exposed to favorable growth conditions (i.e. with light and O₂; time 0) and then to dark/anoxic conditions for a time period of 46 weeks. To adjust dark/anoxic conditions, diatom cultures were transferred into gas-tight, dark bottles (wrapped in aluminum foil), flushed with N₂ for 30 min and kept at 15°C until sampling. At each time point, 2 mL of the cell suspension was taken in 3 replicates and cell numbers were counted. The samples were freeze-dried for 2 days and 5 mL ice-cold acetone was added for extraction of photopigments. After vigorous mixing and sonication for 5 min, the samples were left over night at -20°C, mixed again, and centrifuged for 5 min at 3000g. The supernatants were filtered (Acrodisc^® CR 4 mm, 0.45 µm Versapor^®; Gelman Laboratory) and the extracted photopigments were separated by means of HPLC (Waters 2695; U.S.A.) and analyzed by a photodiode array detector (Waters 996; U.S.A.) as described in Stief et al. [26]. In the chromatograms, chlorophyll a and fucoxanthin were identified according to their specific retention time and absorption spectra and the respective peaks were integrated with the Millenium®32 software (Waters, U.S.A.). Calibrations were made with serial dilutions of chlorophyll a and fucoxanthin stock solutions (DHI, Denmark). All procedures were made under dark conditions and using HPLC-grade chemicals.

Dissimilatory Nitrate Reduction to Ammonium by T. weissflogii

Our results strongly indicate that the ubiquitous pelagic diatom T. weissflogii is able to perform DNRA, similar to the benthic diatom Amphora coffeaeformis, which was the first phototrophic eukaryote shown to dissimilatorily reduce NO₃^- under dark/anoxic conditions [8]. Consumption of intracellular NO₃^- and simultaneous production of NH₄⁺ in response to dark/anoxic vs. light/oxic conditions have been followed in two separate experiments: (a) a non-labeling experiment in which NH₄⁺ was measured photometrically (Figure 1) and (b) a ¹⁵N-stable isotope labeling experiment (Figure 2). In both experiments, the rapid consumption of intracellular NO₃^- and ¹⁵NO₃^- by T. weissflogii was accompanied by the production and release of NH₄⁺ and ¹⁵NH₄⁺, respectively, only under dark/anoxic conditions, but not in the presence of light and O₂ (Figures 1,2). In the ¹⁵N-stable isotope labeling experiment, the initial ¹⁵NH₄⁺ concentration was 2 µM because the hypobromite assay actually measures the sum of ¹⁵NH₄⁺ and ¹⁵N-labeled volatile N compounds such as methyl amines [32]. The concentration of NO₃^- in the medium, i.e. extracellular NO₃^-, only decreased under light/oxic conditions, but remained constant after exposure to dark/anoxic conditions (Figure 1). This constant (and not increasing) extracellular NO₃^- concentration indicates that the intracellular NO₃^- (expressed in μmol L^-1 of growth medium) was indeed consumed by T. weissflogii rather than released from the cells into the medium. Intracellular NO₃^- was also consumed under light/oxic conditions, even at a higher rate than under dark/anoxic conditions (Tables 1,2), most probably because NO₃^- was used for assimilation by photosynthetically active diatoms [33-35]. For N-assimilation, NO₃^- is also reduced to NH₄⁺, but NH₄⁺ is not released from the cells.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Figure 1. Non-labeling experiment.

Time courses of total intracellular NO₃^- (expressed in μmol L^-1 of growth medium), and extracellular NO₃^- and NH₄⁺ concentrations in an axenic T. weissflogii culture in response to (A) dark/anoxic conditions, and (B) light/oxic conditions. Dark/anoxic conditions were initiated at time point 0.

https://doi.org/10.1371/journal.pone.0082605.g001

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Figure 2. ¹⁵N-stable isotope labeling experiment.

Time courses of total intracellular ¹⁵NO₃^- (expressed in μmol L^-1 of growth medium), and extracellular ¹⁵NH₄⁺ and N₂-¹⁵N concentrations in an axenic T. weissflogii culture in response to (A) dark/anoxic conditions, and (B) light/oxic conditions. Dark/anoxic conditions were initiated directly after time point 0. Some error bars, which indicate standard deviation (n=3), are smaller than the symbols.

The NO₃^- was measured with a non-labeling sensitive technique and is consequently not plotted as ¹⁵NO₃^-. However, the only NO₃^- source in the experiment was ¹⁵NO₃^- (see Materials and Methods).

https://doi.org/10.1371/journal.pone.0082605.g002

In the absence of O₂, intracellular NO₃^- can be used for dissimilation by sulfur bacteria [36-39] and only a few unicellular eukaryotes and fungi (e.g. [8,18,20,40,41]). The ubiquitous diatom T. weissflogii can now be added to the short list of eukaryotes that dissimilatorily reduce NO₃^-. Notably, T. weissflogii is the first marine pelagic eukaryote shown to have an anaerobic NO₃^- metabolism, whereas all known eukaryotic NO₃^- reducers thrive in stratified waters, sediments and soils in which anoxic conditions occur in subsurface layers. So far it is not known, whether DNRA in T. weissflogii is respiratory or fermentative. Briefly, in respiratory DNRA, ATP is generated by an electrochemical proton potential across a cell membrane, at which electrons are transferred from the donor to the acceptor NO₃^-, and in fermentative DNRA, ATP is generated by substrate-level phosphorylation [16,42-44]. In prokaryotes, the electron donor and acceptor for respiratory DNRA usually originate from an external source and not from cell metabolism, but may be either organic or inorganic, whereas the electron donor in fermentative DNRA is usually organic [45]. So far, the electron donor used by diatoms for DNRA is not known. In our labeling experiment, acetate was added as a potential electron donor. However, it needs to be further investigated, if T. weissflogii can perform DNRA also with intracellularly stored electron donors, like polysaccharides (e.g. chrysolaminarin), and if the external supply of acetate indeed influences the rate of DNRA.

Our experiments revealed that the rate of NO₃^- consumption after exposure to dark/anoxic conditions depends on the concentration of intracellularly stored NO₃^-. In the non-labeling experiment, the initial intracellular NO₃^- concentration was 20 mM, and in the labeling experiment only 4 mM, resulting in a 6 times lower rate of NO₃^- consumption (Tables 1,2). In the labeling experiment, the production of ¹⁵NH₄⁺ (plus N₂O and NO₂^-) by T. weissflogii balanced the consumption of intracellular ¹⁵NO₃^- within the bounds of accuracy (Figure 2; Table 2), whereas in the non-labeling experiment, the net production of NH₄⁺ did not balance the consumption of intracellular NO₃^- (Figure 1). On average, less than half of the NO₃^- was found back as NH₄⁺ in the culture medium; further, the NH₄⁺ concentration first increased and then decreased slightly with time (Figure 1). This decrease of the NH₄⁺ concentration in the non-labeling experiment is explained by an uptake of NH₄⁺ by T. weissflogii under dark/anoxic conditions that has also been confirmed in other experiments (data not shown). A dark NH₄⁺ uptake and assimilation, respectively, is generally known for phytoplankton [34,46] and was recently also confirmed by gene expression analysis in Thalassiosira pseudonana [47]. This dark NH₄⁺ uptake is not apparent in the labeling experiment, because the addition of non-labeled NH₄⁺ as background concentration (see Materials and Methods) obscures the putative uptake of ¹⁵NH₄⁺. However, under the light/oxic conditions of the labeling experiment, ¹⁵NH₄⁺ also decreased after 5 h because the (non-labeled) background NH₄⁺ was completely taken up (data not shown). The labeling approach did not reveal a production of N₂-¹⁵N by T. weissflogii (Figure 2), which further supports that DNRA and not denitrification is used as a dissimilatory NO₃^- reduction pathway by T. weissflogii.

Release of Nitrous Oxide and Nitrite during Nitrate Dissimilation

The production of N₂O and NO₂^- in response to dark/anoxic conditions has been followed during the ¹⁵N-stable isotope labeling experiment (Figure 3). Both, N₂O and NO₂^- were produced and released from the cells in the same time pattern that has been observed for ¹⁵NH₄⁺, and their production apparently mirrors the consumption of intracellular NO₃^- (Figures 1,2,3). However, the production rates of N₂O and NO₂^- were about 1000 and 100 times, respectively, lower than the production rate of ¹⁵NH₄⁺ (Table 2). Thus, N₂O and NO₂^- are not final products of dissimilatory NO₃^- reduction, but the congruent time patterns indicate that T. weissflogii releases N₂O and NO₂^- as by-product and intermediate, respectively, of DNRA. In prokaryotes, N₂O is a well-known by-product and intermediate of nitrification and denitrification, respectively, and there are some indications that N₂O is also released as a by-product of DNRA, which might have been overseen in some organisms [16,48]. In higher plants, N₂O is emitted from leaves by plant NO₃^- assimilation, strictly speaking during photoassimilation of NO₂^- in the chloroplast [49]. Recently, N₂O production was also found in axenic, illuminated cultures of the green algae Chlorella vulgaris [50]. A release of N₂O by phototrophic eukaryotes under darkness and anoxia has to our knowledge not been documented so far. Even though the rate of N₂O released by T. weissflogii during DNRA might seem low (see above, Table 2), this finding can be of environmental importance because diatoms are highly abundant in the world’s oceans (e.g. [3,51]), hypoxic and anoxic marine environments are spreading [52], and N₂O is a particularly strong greenhouse gas [53]. The production of N₂O under dark/anoxic conditions has recently also been confirmed for the benthic diatom A. coffeaeformis [8], and it might be worth to screen other benthic and pelagic diatom species for N₂O emission under these conditions.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Figure 3. Time courses of extracellular N₂O and NO₂^- concentrations in an axenic T. weissflogii culture in response to dark/anoxic conditions.

Dark/anoxic conditions were initiated directly after time point 0. Nitrous oxide and NO₂^- concentrations were measured in samples from the ¹⁵N-stable isotope labeling experiment using techniques that do not distinguish between ¹⁴N- and ¹⁵N-labeled forms of N₂O and NO₂^-. Some of the error bars, which indicate standard deviation (n=3), are smaller than the symbols.

https://doi.org/10.1371/journal.pone.0082605.g003

The NO₂^- release during DNRA by T. weissflogii could be due to cell leakage or excretion that is frequently observed in marine phytoplankton, including diatoms [54,55], but has not been linked to a response of phytoplankton to darkness and anoxia so far. The observed NO₂^- release might be supported by a slightly higher rate of NO₃^- reduction than NO₂^- reduction throughout the incubation. Further, there might be a time delay in NO₂^- reduction to NH₄⁺ because of constitutive expression of the NO₃^--reductase gene, whereas the (dissimilatory) NO₂^--reductase gene first needs to be induced by the production of NO₂^-. T. weissflogii is not able to take up the released NO₂^- again under dark/anoxic conditions, which is indicated by the observation that the medium NO₂^- concentration is not decreasing during the incubation (Figure 3). Additionally, intracellular NO₂^- storage does not occur in T. weissflogii (data not shown), probably because of the toxic effects of NO₂^- [54].

Slow Degradation of Photopigments in Darkness and Anoxia

To estimate how long T. weissflogii cells retain the ability to operate photosynthesis after exposure dark/anoxic conditions, the fate of the photopigments chlorophyll a and fucoxanthin was followed. Notably, the degradation of the photopigments did not temporally coincide with DNRA by T. weissflogii in response to dark/anoxic conditions. While DNRA activity peaked during the first few hours of dark/anoxic conditions, the major decrease in cellular photopigment contents occurred during the first 3 days (Figures 2,4). After one week of dark/anoxic incubation, the cellular pigment contents had reached a low, but constant level that was maintained for at least 7.5 weeks (Figure 4). These observations are in good agreement with the hypothesis that diatoms use DNRA to enter a resting stage with low metabolic activity, and that T. weissflogii was found to survive at least for 6 weeks after adjusting them to dark/anoxic conditions [8]. Diatoms are known to start photosynthesis and growth very fast after (re)adjusting them to favorable growth conditions, i.e. light and fresh growth medium, even after extended periods of darkness [24,56,57]. To maintain at least low cellular contents of photopigments must be a prerequisite for this. Our experimental design, i.e. that no O₂ and prokaryotes were present in the T. weissflogii culture, further led to a decreased rate of degradation, as O₂-dependent pigment alteration and grazing-induced cell disruption could not occur [58]. Interestingly, the chloroplasts of T. weissflogii cells showed an autofluorescence even after more than 1 year under dark/anoxic conditions (pictures not shown), which might originate from photopigment degradation products that are still poorly understood [59].

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Figure 4. Time courses of the intracellular chlorophyll a and fucoxanthin concentrations in an axenic T. weissflogii culture in response to dark/anoxic conditions.

Dark/anoxic conditions were initiated directly after time point 0. Some of the error bars, which indicate standard deviation (n=3), are smaller than the symbols.

https://doi.org/10.1371/journal.pone.0082605.g004

Ecological and Evolutionary Perspectives

After the benthic diatom A. coffeaeformis was discovered as the first photothrophic eukaryote that dissimilatorily reduces NO₃^-, it was interesting to ask whether this metabolism also occurs in pelagic diatoms: and indeed, we found T. weissflogii as the so far only marine pelagic eukaryote showing this metabolic trait. The respiration of NO₃^- by diatoms might be widespread in marine ecosystems and could have so far overseen implications on the marine N-cycle. For benthic foraminifera, Piña-Ochoa et al. [20] calculated a contribution for the removal of fixed N from marine ecosystems that may be equally important to bacterial denitrification in the seafloor. DNRA will not directly remove fixed N, but in anoxic or hypoxic environments, the produced NH₄⁺ can serve as electron donor for anammox that might be especially important in OMZs with high abundances of anammox bacteria [60]. Further research on NO₃^- respiration by diatoms might also reveal that certain species are capable of other pathways than DNRA, like denitrification as shown for foraminifera [18,20]. Additionally, the exact ambient O₂ concentration in the (micro)environment of the diatoms may trigger different dissimilatory NO₃^- reduction pathways as known from fungi [45].

To date, genes involved in dissimilatory NO₃^- reduction have not been identified in NO₃^--respiring diatoms, foraminifera or gromidii. In contrast, several functional genes have been identified in the denitrifying fungus Fusarium oxysporum: a copper-containing NO₂^- reductase (nirK) and a nitric oxide reductase (P450nor) have been sequenced and characterized [41,61]. Intriguingly, NO₃^--respiring fungi may use enzymes that are normally involved in assimilatory NO₃^- reduction in a dissimilatory mode instead [62]. This could also hold true for diatoms. Assimilatory NO₃^- reductases, multiple transporters for NO₃^-, and components of a NO₃^--sensing system have only recently been discovered in diatom genomes [63,64]. First insights into diatom genomes and the ensuing ecophysiological studies revealed a fascinating evolutionary history of diatoms. An unexpected combination of genes by endosymbiotic gene transfer from two secondary endosymbionts to the exosymbiont nucleus, and also horizontal gene transfer led to several additional inclusions from Bacteria and Archaea genomes [63-66]. The diverse assortment of genes results in novel biochemical pathways like the urea cycle [63,65,67-71] that formerly was not known for photosynthetic organisms and congruously makes diatoms for Armbrust et al. [51] to be neither plants nor animals. Further work on diatom genomes could lead to the identification of functional genes involved in dissimilatory NO₃^- reduction. This would not only convey genetic evidence of dissimilatory NO₃^- reduction by eukaryotes, but would also provide genetic markers for the cultivation-independent detection of so far unrecognized dissimilatorily NO₃^- reducing diatoms directly in the environment.