Genetic inactivation of diatom nitrate reductase abolishes nitrate assimilation but not nitrate uptake and results in transcriptional activation of nitrate storage and triacylglycerol biosynthesis pathways.
Abstract
The ecological prominence of diatoms in the ocean environment largely results from their superior competitive ability for dissolved nitrate (NO3−). To investigate the cellular and genetic basis of diatom NO3− assimilation, we generated a knockout in the nitrate reductase gene (NR-KO) of the model pennate diatom Phaeodactylum tricornutum. In NR-KO cells, N-assimilation was abolished although NO3− transport remained intact. Unassimilated NO3− accumulated in NR-KO cells, resulting in swelling and associated changes in biochemical composition and physiology. Elevated expression of genes encoding putative vacuolar NO3− chloride channel transporters plus electron micrographs indicating enlarged vacuoles suggested vacuolar storage of NO3−. Triacylglycerol concentrations in the NR-KO cells increased immediately following the addition of NO3−, and these increases coincided with elevated gene expression of key triacylglycerol biosynthesis components. Simultaneously, induction of transcripts encoding proteins involved in thylakoid membrane lipid recycling suggested more abrupt repartitioning of carbon resources in NR-KO cells compared with the wild type. Conversely, ribosomal structure and photosystem genes were immediately deactivated in NR-KO cells following NO3− addition, followed within hours by deactivation of genes encoding enzymes for chlorophyll biosynthesis and carbon fixation and metabolism. N-assimilation pathway genes respond uniquely, apparently induced simultaneously by both NO3− replete and deplete conditions.
INTRODUCTION
Diatoms arose in the marine environment early in the Mesozoic Era, appeared in the fossil record around 190 million years ago (Falkowski et al., 2004; Sims et al., 2006), and are among the predominant eukaryotic phytoplankton taxa in oceans worldwide (Malviya et al., 2016). The ability of diatoms to thrive in periodic nutrient-rich conditions resulting from upwelled seawater makes them the basis for one of the world’s shortest and most energy-efficient food webs. Some of the largest fisheries in the world are driven and maintained primarily by diatom-based CO2 fixation that is fueled by nitrate (NO3−)-rich, upwelling seawater. Their capacity for rapid uptake and storage of NO3− enables higher growth rates and provides them with a competitive advantage over other species during seasonal blooms (Cermeño et al., 2011). Under conditions of elevated NO3− availability, such as in areas of coastal upwelling which occur in temperate oceans and in polar regions, diatoms may take up NO3− in excess of their immediate needs for growth (Clark et al., 2011; Dortch, 1982; Lomas and Glibert, 1999, 2000; Villareal et al., 1993). The NO3− may then be stored unreduced in vacuoles (Raven, 1987), utilized as an electron sink during periods of energetic imbalance (Lomas and Glibert, 1999), or reduced to NO2−, which, in turn, may be either released into the environment or reduced to NH4+ for subsequent incorporation into amino and nucleic acids within the cell (Clark et al., 2011).
As in vascular plants, photosynthesis and carbon (C) flux in diatoms are vitally linked to nitrogen (N) metabolism. Inorganic N-assimilation requires both reductants and organic C skeletons produced by photosynthesis and the tricarboxylic acid (TCA) cycle (Falkowski and Stone, 1975). In turn, it provides the reduced amino groups that are necessary for synthesis of proteins (particularly Rubisco), the light-harvesting complexes, and photosynthetic pigments (Nunes-Nesi et al., 2010; Turpin, 1991). In NO3−-limited conditions, diatoms and certain green algae redirect C from storage carbohydrates such as chrysolaminarin and starch to accumulation of neutral lipids, specifically triacylglycerol (TAG) (Rismani-Yazdi et al., 2012; Shifrin and Chisholm, 1981; Yu et al., 2009). Much recent work has focused both on characterizing the genes and metabolic pathways that induce lipid biosynthesis and on identifying routes and regulators that could be harnessed to substantially increase TAG production without negatively affecting growth (Alipanah et al., 2015; Ge et al., 2014; Hildebrand et al., 2012; Levitan et al., 2015; Radakovits et al., 2010; Trentacoste et al., 2013; Yang et al., 2013; Yoon et al., 2012). Deleterious impacts of NO3− depletion include reduction of photosynthetic capacity, C-fixation, and N-assimilation and cessation of growth (Alipanah et al., 2015; Bender et al., 2014; Hockin et al., 2012; Syrett et al., 1986; Turpin, 1991).
The assimilatory nitrate reductase enzyme (NR) is presumed to be essential for N-assimilation and central to the competitive advantage of diatoms in marine environments. Also, the interaction of NR with C-metabolism is thought to underlie partitioning of cellular energy into growth rather than to storage. Diatom NR proteins were first purified and characterized in Thalassiosira pseudonana and Skeletonema costatum (Amy and Garrett, 1974; Gao et al., 1993), and the gene was first isolated and characterized in the diatom Phaeodactylum tricornutum (Allen et al., 2005). Plant NR structure was first described in maize (Zea mays; Lu et al., 1992). The diatom protein is ∼110 kD and is a self-contained electron transport chain with subunits that bind molybdopterin, heme, and FAD (reviewed in Campbell, 1999). NADH, another required cofactor, provides the electrons necessary to reduce NO3− to NO2−.
Here, we focus on the NR gene and protein in the model marine pennate diatom, P. tricornutum. We monitored the cellular localization of native and overexpressed NR protein in relation to N source and availability. Using the TALEN methodology, which enables selectable, target-specific, double-stranded DNA breaks (Weyman et al., 2015), we produced a homozygous gene knockout of NR in P. tricornutum. The striking changes in physiology in the NR-KO cell lines were recorded by confocal fluorescent and electron microscopy, and their biochemical compositions were analyzed by Fourier transform infrared spectroscopy (FTIR). We also characterized diatom nitrate uptake and assimilation, C-metabolism, lipid and fatty acid biosynthesis and recycling, and global gene expression in the NR-KO and wild-type cell lines and report on differences between wild-type and NR-KO cells in response to the addition to NO3−. Based on these studies, we identify vacuolar NO3− transporters in diatoms and propose a scenario for the interaction of NR, vacuolar transport, and NO3− storage. We also suggest a connection between the loss of NR and lipid recycling and compare the differences in transcriptional activation in N-stress conditions between wild-type and NR-KO cells.
RESULTS
NR in P. tricornutum Shows Dynamic Localization in Relation to Extra- and Intracellular NO3− Availability
The nitrate reductase gene (NR) from P. tricornutum was overexpressed as a YFP-NR fusion and introduced into wild-type cells by biolistic transformation. Confocal fluorescent microscopy imaging showed robust YFP-NR protein expression throughout the cytosol in NO3−-replete conditions (Figure 1A). As cells entered stationary phase, when NO3− becomes limiting, the YFP-NR protein relocalized to punctate aggregations of differing sizes (Figure 1B). Imaging by transmission electron microscopy of immunogold-stained cross sections of YFP-NR transgenic cell lines sampled during NO3−-depleted conditions clearly showed electron-dense punctate signal in what appears to be the peroxisome (Figures 1C and 1D).
Figure 1.
Intracellular Localization of NR in P. tricornutum in Overexpression Line and in the Wild Type.
(A) to (D) Localization of YFP-NR fusion protein is affected by growth stage. Bars = 5 µm in (A) and (B) and 500 nm in (C) and (D).
(A) YFP-NR cells during early exponential phase growth. Yellow, YFP-NR fusion protein; red, chloroplast autofluorescence.
(B) YFP-NR fusion protein aggregates into punctate forms during stationary phase.
(C) Gold-labeled YFP-NR is visible in an organelle, possibly in the peroxisome.
(D) Enlargement of (C).
(E) to (H) Native NR protein, immunolabeled with Alexafluor 488, during a short (1–18 h) time-course experiment. Green, labeled NR; red, chloroplast autofluorescence. Bars = 5 µm.
(E) Wild-type cells were grown in F/2 media (with NH4+ in place of NO3−) to mid-exponential phase. Cultures were sampled just prior to harvest. Six green spots are labeled NR protein in aggregated form.
(F) Cells were transferred into N-free (−N) media. Samples collected after 2 h without N.
(G) Wild-type cells respond to addition of NO3−. NO3− (300 μM) was added to cultures; cells were incubated for 2 h and sampled. Labeled NR protein spread throughout the cell.
(H) Samples harvested 18 h after addition of NO3−. Labeled NR has coalesced into small punctate aggregates.
To examine the in vivo distribution of native NR, an immunofluorescence experiment was performed to determine the location of endogenous NR under various N conditions over short time scales (15 min to 18 h) in wild-type cells. In cultures grown to exponential phase on NH4+, a condition known to repress NO3− uptake and NR activity (Glibert et al., 2016), sparse NR protein signals had aggregated into punctate forms (Figure 1E; Supplemental Figures 1A and 1B). When cultures were then transferred to and incubated in N-free (−N) media for 2 h, NR remained present as punctate signals, a few seen in each cell (Figure 1F; Supplemental Figures 1C to 1F). However, 90 min after cultures were spiked with NO3− (150 μM), native NR was abundant and its distribution was diffuse throughout the cell, including the cytosolic spaces surrounding the vacuolar membranes (Figure 1G). After cells were rinsed twice in N-free media and resuspended in NO3− (300 μM), the NR signal was abundant through the 15- and 45-min time points, whereas by 18 h with NO3− depleted from the media (Supplemental Table 1), NR had become fully aggregated with its greatest abundance nearest the chloroplast (Figure 1H). The similarity of localization profiles for native NR in wild-type cells (Figure 1H) and YFP-NR fusion proteins in transgenic cell lines under NO3−-deplete conditions (Figure 1B) indicated that the punctate appearance in those lines was not the result of an artifact associated with recombinant overexpression.
To corroborate the transmission electron microscopy evidence that punctate NR resides in the peroxisome (Figures 1C and 1D), NR was colocalized with a known peroxisomal matrix protein, 3-ketoacyl-CoA thiolase (KAT), which was identified in P. tricornutum by Gonzalez et al. (2011). YFP-NR and CFP-KAT constructs were each inserted into regulatory cassettes controlled by the NR promoter (ProNR) and NR terminator (TermNR) (Poulsen and Kröger, 2005). Both vectors were then transformed simultaneously into P. tricornutum. Clones showing both fluorescent signals in the same cell were analyzed in a time-course experiment similar to the one described above and were monitored using confocal microscopy. Potential colocalization occurrences were analyzed by spot correlations, a method that estimates the correlation of pixels for each wavelength co-occurring at specified region of interest. At 4 h, the signals of both YFP-NR and CFP-KAT were generally indicative of their characteristic localization: cytosolic for NR and punctate for KAT. Nevertheless, some green NR signal overlaps with blue KAT signal, indicating that some NR had moved to peroxisomes. At 18 and 36 h, with extracellular NO3− depleted, the NR signal was found colocalized with KAT (Supplemental Figures 2C, 2E, and 2F). Overlay statistics for the 18-h time point confirmed that the area of the dot was nearly 90% populated by both yellow and blue pixels (Pearson correlation 0.92), indicating that most of the YFP-NR is trafficked to the peroxisome in conjunction with extracellular NO3− depletion. These results indicated that extracellular NO3− availability may be a driver of localization: When extracellular NO3− is present, NR is largely cytosolic, but when it is depleted or absent, NR distribution is punctate or collected in the peroxisome.
Knocking Out Nitrate Reductase Transforms P. tricornutum Physiology
To examine the role of NR in diatom physiology and viability, we used TALEN genome editing to enable biallelic homologous recombination of the Sh ble gene, a zeocin antibiotic resistance marker, into a double-strand break in P. tricornutum NR (Weyman et al., 2015). After biolistic transformation, 16 transgenic lines were selected for further screening. Insertion of the Sh ble cassette within the NR loci was confirmed by PCR. All 16 lines were grown in duplicate in both NH4+ and NO3− amended media. Eight lines were identified by their lack of growth on NO3− as the sole N source. The loss of NR was confirmed in a 24-well plate assay; one of the eight lines (NR-KO14) is shown in Figure 2A. In a second assay, we monitored both wild-type and several NR-KO lines for photosynthetic efficiency and growth in both media (Figures 2B and 2C). As shown in Figure 2, NR-KO cells could not grow on NO3−. We then analyzed wild-type and NR-KO14 protein samples, collected during a subsequent time-series analysis (see below), by immunoblot (Figure 2E) and found that NR protein was not detectable in the NR-KO14 line grown on NO3−.
Figure 2.
Loss of Nitrate Reductase Function in NR-KO Transgenic Lines.
(A) A 24-well plate growth assay comparing wild-type and NR-KO cells. On the left side of plate, cells grown in NH4+; on the right side, cells grown in NO3−. Plate shows cells 5 d after inoculation into the two N media. BL labels indicate blank wells; the wells were filled with N media, but not inoculated with cells, providing quality control for crosstalk between wells.
(B) Time-course experiment comparing the wild type to NR-KO14 line for photosynthetic efficiency over a 7-d period. Fv/Fm was recorded on a PAM fluorometer. Both lines were initially cultured in 880 μM NH4+ medium to mid-exponential phase (Exp on x axis); cells were then washed in N-free media and resuspended in NO3− (“0” on x axis) and sampled daily for 6 d. Wild-type, black square; NR KO14, white square. Means ± sd of two biological replicates are shown for each cell line.
(C) Measurement of chlorophyll a autofluorescence over 7 d provided a proxy for growth rate. Means ± sd of two biological replicates are shown for each cell line.
(D) Cell counts (cells/mL) for the wild type versus NR-KO corresponding to growth measurements. Only one replicate was counted for each line at each time point. Cell counts for day 1 occurred immediately after cell lines were transferred into NO3− media.
(E) NR immunoblot with wild-type and NR-KO14 total protein extract. Protein fractions harvested for 10 d wild type versus NR-KO14 time series were compared. Protein ladder, lane e; wild-type samples, lanes a to d; knockout samples, lanes f to i. The NR monomer has 892 amino acid residues, equaling ∼98 kD. The NR band comigrates with the 100-kD band on the protein ladder.
In keeping with traditional genetic studies, P. tricornutum NR-KO lines 9 and 14 were complemented by a conjugative plasmid containing the native NR gene fused to YFP under the control of the native ProNR and TermNR; the plasmid was delivered by Escherichia coli (Karas et al., 2015). Sixteen transformants from each KO line were screened on NO3− for viability, and all grew on solid and liquid media amended with NO3−. In 10 rescued NR knockouts, YFP expression was also induced when the cells were switched from NH4+ to NO3− media (Supplemental Figure 3).
FTIR spectroscopy, previously calibrated using biochemical measurements (Levering et al., 2016; Mayers et al., 2013; Stehfest et al., 2005), provided evidence of the changes in biochemical composition occurring in the NR-KO lines compared with the wild type (Figure 3A). During a 7-d time course experiment, three NR-KO lines (#9, 14, and 15) and the wild type were initially grown in NH4+ media and then transferred into NO3−-amended media and sampled periodically. Twelve hours after exposure to NO3−, substantial changes were detected in the NR-KO lines: The percentage of biomass for proteins, lipids, and carbohydrates deviated sharply from the distribution in wild-type cells (Figure 3). Carbohydrates, for example, constituted 40% of the biomass in NR-KO but only 13% in wild-type cells, respectively (Figure 3C). At 60 h, the percentage of lipids was 10% higher in the NR-KO lines than the wild type (Figure 3D). The ratios of relative contributions of lipids to carbohydrates were 1.8 times higher in NR-KO cells as well.
Figure 3.
Addition of NO3− Induces Rapid Changes in Biochemical Composition of NR-KO Cells
(A) to (D) Analysis of biochemical composition by Fourier-transform infrared spectroscopy. Three NR-KO lines (KO9, KO14, and KO15) (biological replicates, n = 3) were compared with the wild type (technical replicates, n = 3) over a 7-d time course.
(A) One day after transfer into NO3− media, wild-type and NR-KO9 cells were scanned using an infrared spectrometer. The NR-KO1 scan is shown in orange and the wild-type scan in blue. The following peaks (wave numbers from left to right, highest to lowest) correspond to basic functional groups: 1745 cm−1, esters of lipids or fatty acids; 1655 cm−1, protein (amide I); 1545 cm−1, protein (amide II); 1200 to 900 cm−1, polysaccharides/siloxane.
(B) to (D) Means ± sd of triplicate biological/technical replicates are shown for NR-KO14 and wild-type samples. NR-KO14 data series are shown in white circles; wild-type data series are in black circles. Pairwise t tests compared each NR-KO cell line triplicates to the wild-type triplicates at each time point, for each component. Of the 54 total t tests, only 14 t tests were not significant (P = 0.05). Gray dots below a data point indicate that at least 1 in 3 of the comparisons for the data point was not significantly different.
(B) Protein content expressed as a percentage of total biomass for wild-type and NR-KO lines.
(C) Carbohydrate as a percentage of biomass for wild-type and NR-KO lines.
(D) Lipid as a percentage of biomass for wild-type and NR-KO lines.
Wild-type and NR-KO lines observed by confocal microscopy during the growth assay had a pronounced phenotype. Images collected 4 and 6 d after wild-type and NR-KO cultures were transferred into NO3−-containing media revealed swollen NR-KO cells with inflated vacuoles and chloroplasts that were reduced in size and pushed toward the ends of the cells (Figures 4A to 4C). To further investigate this phenotype, quick-freeze deep-etch (QFDE) electron micrographs (Heuser, 2011) were prepared from NR-KO14 lines after 4 d of growth on both NH4+ and NO3−. In NR-KO14 cells grown on NH4+ (Figure 4D), vacuoles occupied the two termini and were much smaller than in cells grown on NO3− where the vacuoles extend the full length of the cell and fill most of the cell volume (Figures 4E and 4F). Vacuolar content was also distinctly different in the two samples: In NH4+-grown cells, it had the appearance expected of a salt solution from which water has been removed during the etching process (Figure 4G); however, in NO3−-grown cells, the vacuole was packed with fibrous material (Figure 4H). QFDE cross sections of wild-type and NR-KO14 cells (Supplemental Figures 4A to 4C), as well as the full-length images, showed the bloated vacuoles appressing cytoplasmic organelles, although their ultrastructure appeared largely uncompromised (Figure 4F; Supplemental Figure 4C). The chloroplasts were the exception: Their size was reduced and the thylakoids appeared wavy and somewhat swollen. The vacuolar membranes in NR-KO14 cells appeared robust, fully containing their contents (Figure 4F; Supplemental Figure 4C).
Figure 4.
Morphological Changes to NR-KO Cells When Grown on Nitrate.
(A) to (C) Bars = 5 µm. Chlorophyll autofluorescence (red) indicates locations of chloroplasts.
(A) Confocal micrograph of wild-type cells 4 d after transfer to N-free F/2 media.
(B) NRKO-14 cells 4 d after being transferred from NH4+ to NO3− amended F/2 media.
(C) NRKO-14 cells 6 d after being transferred from NH4+ to NO3−.
(D) QFDE electron micrographs of a NR-KO14 cell in mid-exponential phase grown in NH4+. Vacuoles are found at either end of cell; the nucleus and cytoplasm, in the center. C, chloroplast; G, Golgi; L, lipid body; M, mitochondrion; N, nucleus; V, vacuole; VM vacuolar membrane. Bar = 2 μm.
(E) NR-KO14 cell grown on NO3− F/2 media for 4 d. The vacuolar membrane appears to extend the length of the cell. Bar = 2 μm.
(F) NR-KO14 cell grown on NO3− F/2 media for 4 d. The nucleus and cytoplasm are constricted. Bar = 2 μm.
(G) and (H) Comparison of vacuole contents. Bars = 300 nm.
(G) Cell maintained in NH4+ media for 4 d.
(H) Cell maintained in NO3− media for 4 d.
NR-KO Impacted NO3− Assimilation and Associated Gene Expression
To characterize the impact of NR-KO on the physiology and biochemistry of P. tricornutum, we conducted a 10-d time-course experiment in which both wild-type and NR-KO14 cells were initially grown on NH4+, transferred to N-free media for 2 h, subsequently resuspended in NO3−, and then sampled periodically. Biological duplicates for extra- and intracellular NO3−, total organic carbon/total nitrogen (TOC/TN), and lipid content by fatty acid methyl ester (FAME) analyses were collected at the N-free (–N) condition and nine following time points: t = 1 h, 18 h, 36 h, 60 h, 84 h, 132 h, and 228 h (10 d).
Uptake of NO3− and its assimilation was tracked by monitoring intra- and extracellular NO3− concentrations. In wild-type cells, extracellular NO3− was rapidly depleted from the media and by 18 h was undetectable (Figure 5A); intracellular NO3− stores were fully assimilated by 36 h (Figure 5B) with cell growth arresting at 84 h (Figure 5C). In NR-KO14 cells, assimilation of N derived from NO3− was dramatically impaired. Initially, and similar to the wild type, NR-KO14 cells rapidly took up extracellular NO3−. However, in contrast to the wild type, intracellular NO3− concentration in NR-KO14 cells started to rise at 3 h and continued to increase through 60 h, despite growth arresting at 18 h (Figures 5A to 5C and 6, i). Differences in both extra- and intracellular NO3− concentrations between wild-type and NR-KO14 cells over the entire course of the time series were shown to be significant by a two-way ANOVA (Supplemental File 1). These data document that, whereas wild-type cells exhibited the rapid transport and assimilation of NO3−, typical of diatom growth, NR-KO14 cells, although capable of NO3− transport into the cell, were unable to convert it into cellular biomass.
Figure 5.
Nitrate Uptake and Cell Concentrations in NR-KO14 and Wild-Type Cells
NR-KO14 and wild-type cells were collected on filters; filtrate for each time point was also collected. NO3– (300 μM) was added to N– media 1 h prior to the first sampling. The wild type is shown in black squares; NR-KO14, white squares. Means ± sd of two biological replicates are shown. Asterisks indicate significant differences in concentrations of NO3− between genotypes (paired t tests, P ≤ 0.05).
(A) Extracellular NO3– removed from the media by wild-type and NR-KO14 cells measured by UV spectroscopy. Significant differences (P = 2.9 × 10−6) were observed in extracellular NO3− concentrations between wild-type and NR-KO14 strains over the time course by two-way ANOVA (Supplemental File 1).
(B) Intracellular NO3– extracted from cells measured by UV spectroscopy. Significant differences between strains in intracellular NO3– concentrations (P = 1.0 × 10−10) were also observed in a two-way ANOVA (Supplemental File 1).
(C) Cell counts for the wild type and NR-KO14. Error bars indicate 
sd between biological duplicates, with the exception of wild-type samples 36-h, 84-h, 132-h, and 10-d time points, where only one duplicate for each sample was counted. Significant differences were not determined for cell counts.
Figure 6.
Overview of Physiological and Transcriptomic Changes in NR-KO14 Cells Involving Proposed Vacuolar Storage of NO3−.
(i) NO3− is pumped into vacuole as suggested by confocal results and NO3− UV spectrophotometry data.
(ii) FAME analyses showed that, in NR-KO14 cells, C16 FA fractions of total lipids change and TAG production increases 1 to 3 h after NO3− is added. Lipid bodies (LB) were more abundant in NR-KO14 than in wild-type cells at 36 h (Figure 9D and 9E).
(iii) NRT2 (J26029) is the most highly upregulated of the six MFS NO3− transporters in the NR-KO14 transcriptome.
(iv) Three ClC NO3− vacuolar transporters, EG01952, J28245, and J46097 were upregulated NR-KO14 cells at 1 h (Figure 8B).
(v) Nitrite transporter NAR1 (J13076) was highly expressed at 1 h in NR-KO14 cells (Figure 7B).
(vi) Chloroplast-localized N-assimilation genes NiR-Fd (J12902), GSII (J51092), ACOAT (J50577), and EG02286 (MFS) are upregulated at 1 h in NR-KO14 cells (Supplemental Figures 5A, 5B, and 5D).
(vii) Two genes putatively involved in preliminary steps in sterol biosynthesis: a C4 sterol methyl oxidase (J10852) and a sterol C5 desaturase (J45494) are upregulated at 18 h (Figures 11C and 11D). Sterols make up ∼30% of vacuolar membrane lipids in plants.
(viii) Three genes for sphingolipid biosynthesis catalyze the first three steps in the production of all sphingolipids: a putative serine palmitoyltransferase (J41807), a sphinganine reductase (J48977), and a sphingolipid DS (J22677) (Figures 11A to 11C). A lipid recycling gene, phospholipase (J48391) (Figure 11G), is upregulated at 1 to 42 h. It is putative recycler of nonchloroplast phospholipids for new lipids and/or signaling.
(ix) Patatin-like galactolipase (EG0071) is involved in recycling chloroplast thylakoid lipids in other microalgae and plants (Figure 11F).
(x) Lipid exporter ABCA1 (J23497) (Supplemental Data Set 1, lipid column).
(xi) Two phosphatidylinositol transfer proteins, PIPT (J33651, J33873) (Figures 11H and 11I), are known in higher eukaryotes to be involved in regulation of lipid metabolism and signaling.
(xii) TAG mediators are also upregulated. Chloroplast-targeted pPDH E1 (EG02309), ER-localized FADS (J28797), PAP (J39949), DGAT2 (J43469), and MLDP (J48859) are canonical TAG biosynthesis enzymes (Figure 10). Two TAG lipases (J41624 and EG02610) that hydrolyze TAG for FA recycling are moderately upregulated in NR-KO14 transcripts (Supplemental Data Set 1, lipid column).
(xiii) Genes involved in the β-oxidation of fatty acids are induced in NR-KO14 cells when NO3– is added. LCA dehydrogenase (J11014), SCA dehydrogenase (J25932), and enoyl-CoA hydratase (J35240) are upregulated (Supplemental Data Set 1, lipid column).
(xiv) The photosynthetic gene cluster (PSII, LHC complex, and PET) is sharply downregulated following the addition of NO3− (Supplemental Figures 9A and 9B).
(xv) Chlorophyll biosynthesis transcriptionally inactivated at 1 h (Supplemental Figures 9C to 9E).
(xvi) CB cycle gene transcription is downregulated at 1 to 42 h (Supplemental Figure 10).
(xvii) Mitochondrial glycolysis transcription is also downregulated at 42 h (Supplemental Figure 10).
(xviii) Three TCA cycle genes involved in 2OG production follow NR cohort expression (Supplemental Figure 11).
(xix) Biosynthesis of 99 ribosomal structural proteins, 28S and 60S, are steeply downregulated at 1 h (Supplemental Figure 12).
Black lettering indicates physiological results; red and blue lettering indicate upregulated or downregulated expression, respectively, in response to the addition of NO3−; bold lettering show greater differences in log2 fold change values in NR-KO14 cells compared with the wild type (see Supplemental Data Set 1 for gene expression details).
Subsequently, RNA-seq data were compiled from a second time-course experiment in which samples from triplicate NR-KO14 and wild-type cultures were collected at time points similar to those in the previous experiment. For the RNA-seq analysis, samples were harvested representing both NH4+ and (–N) preconditions, then after transfer into NO3− at 1, 18, 42, 66, 90, 114, 138, 162, and 228 h (10 d). Because the physiological results suggested that N-assimilation does not proceed in NR-KO14 cells, despite the transport of NO3− into the cells, we compared RNA-seq profiles for genes normally induced for NO3− transport, utilization, and storage in both transcriptomes. A cohort of 35 genes that coexpress with NR (J54983) was identified in the wild-type transcriptome by expression template matching using the threshold P value (P < 0.0001) (Supplemental Data Set 1). The cohort included genes encoding the primary proteins in the NO3− assimilation pathway: NO3− transporter (HANT NRT2, J26029), chloroplast-targeted nitrite (NO2−) transporter (NAR1, J13076), and chloroplast-targeted Fd-dependent NO2− reductase (NiR-Fd, J12902) (Figures 7A, 7B, and 7E; Supplemental Table 2A). Six other genes among the cohort were highly expressed and have putative functional links to NR, NO3− assimilation, or the integration of N- and C-metabolism. They include (1) a second putative NADH-NO2− reductase (NiRBD, EG02286) (Bowler et al., 2008) targeted to the chloroplast, which is similar to NirB found in fungi and bacteria; (2) a voltage-gated, vacuolar NO3– (chloride channel [ClC] family) transporter (Phatr3_EG01952); (3) an acetylornithine transaminase (ACOAT, J50577), a chloroplast-targeted protein producing N-acetylornithine and 2-oxoglutarate (2-OG) from glutamic precursors (Fernandez-Murga et al., 2004); (4) isocitrate dehydrogenase (ICDH; J24195), the third enzyme in the TCA cycle and primary producer of 2-OG; (5) carbamoyl phosphate synthetase (unCPS, J24195), a key component of the ornithine-urea pathway (Allen et al., 2011); and (6) a molybdenum cofactor biosynthesis protein (MoaC, J15625), a subunit of molybdopterin synthase, in which molybdopterin is the required cofactor necessary for NO3− binding and electron transfer in the NR active site (Figures 7C and 7D; Supplemental Table 1A). Of the other upregulated NR cohort genes, 10 have lower reads per kilobase million (RPKM) values and 13 are not associated with statistically significant annotations beyond “hypothetical protein.”
Figure 7.
NR Cohort Expression in Wild-Type and NR-KO14 Transcriptomes.
(A) Wild-type transcript abundance of NR (J54983) and three coexpressed cohorts. The cohorts are the nitrate transporter, NRT2 (J26029); nitrite transporter, NAR1 (J13076), and Fd-nitrite reductase, NiR-Fd (J12902). NRT2 (dotted line) is plotted on the secondary axis. Means ± sd of triplicate biological replicates are shown. Significant differential expression (DE) analysis for NRT2, NAR1, and Fd-NiR is shown (C) and (D).
(B) Transcript abundance of the NR cohort genes in the NR-KO14 transcriptome.
(C) NR cohort cluster (n = 21) determined in wild-type transcriptome using a template matching algorithm. The top 11 genes, ranked by highest total RPKMs and presence of an annotation, are shown in the heat map.
(D) Wild-type NR cohort identified in NR-KO14 and aligned with the wild type. Heat map colors: magenta, higher expression values; green, lower expression values (normalized RPKMs) across the time series. Differential gene expression was determined with two paired t tests (edgeR): (1) Gene expression at each time point was compared with values at the N-free (N-) time point for each cell line, which served as the control for all time points. For both gene sets, nearly all N-free-anchored t tests were significant (FDR = 0.05). White dots on the heat map indicate time points where differential expression was not significant. (2) edgeR paired t tests compared NR-KO14 versus the wild type for each gene at each time point; 86% of the pairwise comparisons (76 of 88) through 162 h were significantly different (FDR = 0.05) (Supplemental Data Set 3).
(E) Wild-type and NR-KO14 N-assimilation schemes. NRT2, nitrate transporter; NR, nitrate reductase; NAR1, nitrite transporter; NiR, Fd-nitrite reductase; GSII, glutamine synthetase; GOGAT, glutamate synthase (GltS). The orange line represents the plasma membrane; green curved line is the chloroplast membrane.
In wild-type P. tricornutum, the expression of NR cohort genes peaked sharply at 1 h and again broadly from 66 to 162 h, corresponding to the addition of NO3− and to the onset of N-limitation (Figures 5A and 5B). Elevated expression of NR cohort genes after the addition of NO3− is consistent with earlier studies in Arabidopsis (Scheible et al., 1997; Wang et al., 2000), where the addition of NO3− was linked to increases both in uptake rates for NO3− and in expression of NR and the genes involved in N-assimilation (Figures 7A and 7C). Induced expression of the cohort in response to N-stress is also in agreement with previously described results in Arabidopsis thaliana and P. tricornutum (Alipanah et al., 2015; Glass et al., 2002; Levitan et al., 2015; Scheible et al., 2004; Wang et al., 2000, 2004). In NR-KO14 cells, the patterns of gene expression were atypical. Transcript levels were upregulated at 1 h, as in the wild type, but then remained high through 18 h. Notably, expression levels at those time points were 1.5 to 2 times greater in NR-KO14 cells as compared with the wild type (Figures 7B and 7D; Supplemental Table 2A). Significant differential expression was observed for nearly all the NR cohort genes in both genotypes (Figures 7C and 7D; Supplemental Data Set 2).
Expression of downstream N-assimilation pathway genes including the two NO2− reductases and two glutamine synthetases (GSII, chloroplast [J51092], and GSIII, mitochondria [J22357]) was also impacted by the loss of NR (Figure 7E; Supplemental Figures 5A and 5B). In wild-type cells, expression was induced by both the addition (1 h) and depletion (66–162 h) of NO3−, whereas in NR-KO14 cells, maximal transcript abundance occurred at 1 to 18 h. Again, expression levels were 2 to 4 times greater in NR-KO14 than in wild-type cells at those early time points, except for GSIII (Supplemental Figure 5B and Supplemental File 1). The timing and elevated levels of expression in these genes suggests that transcription initiation was induced by both NO3−-responsive and N-starved activators/derepressors, which in wild-type cells occurred at two temporally distinct points in the experiment. In NR-KO14 cells, however, it appears that such dual transcription may occur almost simultaneously in response to the addition of NO3−. This unusual expression was not observed for genes encoding GOGAT, chloroplast-targeted GltS (J24739), or mitochondrial GltX (J51214) (Supplemental Figure 5C). By contrast, expression of genes encoding ACOAT, unCPS, and glutamate dehydrogenase (GLDH, J13951), which are further downstream of NR in the N-assimilation pathway, also appeared to be driven by dual transcription between 1 and 18 h in NR-KO14 cells (Supplemental Figure 5D to 5F). See Supplemental Data Set 2 for differential expression analysis of the N-assimilation gene set.
Genes encoding proteins responsible for NO3− transport fall into two categories: major facilitator superfamily (MFS) genes, which transport NO3− into the cell from the environment (Figure 6, iii; Supplemental Data Set 1 and Supplemental Figure 6) (Forde, 2000; McDonald et al., 2010), and voltage-gated ClC NO3− transporters, which in vascular plants move NO3− from the cytosol into and out of the vacuole as needed (Figure 6, iv) (Bergsdorf et al., 2009; De Angeli et al., 2006; von der Fecht-Bartenbach et al., 2010). Five paralogous MFS NO3− transporter genes in the P. tricornutum genome (Bowler et al., 2008; http://protists.ensembl.org/Phaeodactylum_tricornutum/Info/Index) were identified in both transcriptomes (Supplemental Figures 6A to 6C, Supplemental Table 2B, and Supplemental Data Set 2). The NR cohort gene NRT2 was the most highly expressed of the MFS NO3− transporters; at 138 h, its transcript levels exceeded 12,000 RPKM (Figure 7A). A second MFS transporter, EG02608, is coexpressed with J26029, albeit at much lower transcript abundances. Interestingly, the vacuolar NO3− (ClC) transporter EG01952, initially identified in the NR cohort, is also coexpressed with J26029 and EG02608 in both wild-type and NR-KO14 cells (Figure 8A).
Figure 8.
Outer Membrane and Vacuolar Nitrate Transporters.
(A) Expression of two MFS nitrate transporter genes and a (ClC) NO3− vacuolar transporter (EG01952) in both wild-type and NR-KO14 cells. Means ± sd of triplicate biological replicates are shown.
(B) Normalized RPKMs for six vacuolar transporters. Data are shown for wild-type transcripts (top panel) and NR-KO14) transcripts (bottom panel). Green and orange outlines around gene IDs refer to Figure 7D. All genes were differentially expressed in each transcriptome except where indicated by white dots.
(C) Response of two (ClC) NO3− vacuolar transporters to nitrate flux in wild-type and NR-KO14 cells. Paired t tests were run for both genes. Nearly all (18 of 19) comparisons were significantly different between NR-KO14 and the wild type (FDR < 0.05). Means ± sd of triplicate biological replicates are shown.
(D) Upper panel is a Clustal Omega alignment showing conserved residues for anion binding in Arabidopsis (At) ClC-a and three putative P. tricornutum ClC NO3− transporters, J46097, J28245, and J52412. ClustalX colors represent types of residues: blue, hydrophobic; green, polar; magenta, negative charge; orange, glycine; red, positive charge; yellow, proline; uncolored residues indicate conservation was <60%. Lower panel: Jalview plot of the conservation threshold of the alignment (scale = 1–9, asterisk represents 100% conservation of residues at that position in the sequence).
In addition to EG01952, five other P. tricornutum genes are predicted to encode vacuolar NO3− (ClC) transporters (PF00654) (Figure 8B; Supplemental Table 2B and Supplemental Data Set 2). In wild-type cells, the correspondence between increased expression occurring between 18 and 66 h and the flux of available extra- and intracellular NO3− (Figures 5A and 5B) suggested vacuolar storage of NO3− may be an imperative for sustaining growth. Transcript profiles for NR-KO14 showed a less coherent pattern in response to ambient and intracellular levels of NO3− with more genes upregulated at higher transcript levels from 1 to 42 h. Such expression may be a factor in the expansion of vacuoles observed in the micrographs of NR-KO14 cells. A comparison of wild-type expression profiles for two vacuolar transporter genes, EG01952 and J28245, clearly documents alternatively timed responses to NO3− availability (Figure 8C). In addition to NO3− (ClC) vacuolar transporters, V-type ATPases and H+ PPases provide energy, or generate a proton motive force, to pump NO3− and other ions across the vacuolar membrane (reviewed in Martinoia et al., 2007). Their expression profiles were similar to the NO3− (ClC) vacuolar transporter profiles in both the wild-type and NR-KO14 transcriptomes (Supplemental Data Set 1).
NR-KO Also Impacts Lipid Metabolism, Remodeling, and Gene Expression
As NR-KO14 cells appear to sense to both N-replete and N-starved signals simultaneously following the addition of NO3−, we monitored the effect of NO3− on C:N ratios (Figure 9A) and lipid accumulation in NR-KO14 and wild-type cells (Figures 9B and 9C). Total lipids were extracted and quantified for fatty acid (FA) composition by FAME analysis in both wild-type and NR-KO14 cells. Following an initial decline, lipid content in wild-type cells had doubled by 132 h (Figure 9B). This corresponded to the rise in its C:N ratio; extra- and intracellular NO3− supply had been depleted (Figure 5C), and growth had just arrested (Figures 2C and 2D). By contrast, FAME content in NR-KO14 cells increased rapidly following addition of NO3− and more than doubled, prior to plateauing by 60 h (Figure 9B). Differences in accumulation of FAME were significant for the two genotypes through 60 h as determined by a two-way ANOVA (Supplemental File 1). FAME composition in both wild-type and NR-KO14 cells was dominated by C16:0, C16:1, and C20:5 FAs (Supplemental Figures 7A to 7E), consistent with previous reports for diatom FAME composition (Abida et al., 2015; Shen et al., 2016; Yu et al., 2009). As previous studies have established (Bender et al., 2014; Hockin et al. 2012; Li et al., 2014; Rismani-Yazdi et al., 2012; Shifrin and Chisholm, 1981; Yang et al. 2013; Yoon et al., 2012), N-limited conditions in photosynthetic microalgae induce accumulation of lipids, predominantly TAG, which is stored in lipid droplets. Therefore, it was not surprising to observe very rapid increases in TAG accumulation in NR-KO14 cells (Figure 8C) as N-stress is sensed amid NO3−-replete conditions. In wild-type cells, TAG did not increase until after 18 h (Figure 8C). In a separate experiment, the presence of TAG was visualized in both cell types at 36 h (Figures 8D and 8E), further confirming that TAG was more abundant in NR-KO14 than in wild-type cells.
Figure 9.
Impact of NR-KO on Total Carbon, FAMEs, and TAG.
(A) to (C) Differences in carbon flux in wild-type and NR-KO14 cells. The wild type is shown with black squares; NR-KO14 is shown with white squares. Means ± sd of two biological replicates are shown. Asterisks beneath the x axis show significant difference in pairwise comparison (t tests, P ≤ 0.05) of the wild type versus NR-KO14.
(A) Ratio of total carbon to total nitrogen (C:N). t tests not determined for C:N data.
(B) Accumulation of lipids in wild-type and NR-KO14 cells. Lipid content was determined by FAME analysis.
(C) Ratio of key cellular FA components: C16:0 and C16:1/C20:5 serve as indicators of the changing lipid distribution in microalgae in response to N-stress.
(D) Wild-type cells live-stained with Bodipy 36 h after NO3− addition. Green, Bodipy stain, showing neutral lipid droplets; red, chlorophyll autofluorescence.
(E) NR-KO9 cells stained with Bodipy 36 h after NO3− addition.
Bars = 5 µm in (D) and (E). For larger images, see Supplemental Figure 14.
A comparison of expression profiles in the NR-KO14 and wild-type transcriptomes for key enzymes involved in TAG biosynthesis indicated that NR-KO14 cells more rapidly upregulate TAG-cluster genes at 1 and 18 h than do wild-type cells (Figure 10A). Chloroplast-targeted pyruvate dehydrogenase E1α (EG02309) was 4 times more highly expressed in NR-KO14. In plants and Chlamydomonas reinhardtii, the enzyme produces acetyl-CoA, a precursor to FA and TAG (Shtaida et al., 2014) (Figure 10B). Two desaturases involved in TAG biosynthesis and C20:5 biosynthesis (J9316 and J28797) were ∼10 times more highly upregulated at 18 h in NR-KO14 cells as compared with the wild type (Figures 10C and 10D). J9316, a palmitoyl-acyl-carrier protein desaturase (PAD), observed in other microalgae as highly responsive to N-stress (Liu et al., 2012; Li et al., 2014; Dolch and Maréchal, 2015), is localized to the chloroplast stroma and converts C16 to C16:1. Its elevated expression was correlated to the timing of increased C16:1 FA accumulation in NR-KO14 cells (Supplemental Figures 7C and 7E). In addition, J28797, an ADS Δ9 fatty acid desaturase (or FAD), which may be localized on the endoplasmic reticulum (ER) membrane (Dolch and Maréchal, 2015), converts C18:0 to C18:1, the first step in C20:5, eicosapentaenoic acid, biosynthesis (Domergue et al., 2003). Phosphatidic acid phosphatase (PAP, J39949), the penultimate enzyme in the TAG pathway, showed a spike in transcription at 18 h (Figure 10E). Expression levels of J43469, diacylglycerol acyltransferase 2 (DGAT2), which completes the final step in TAG biosynthesis, although previously reported as strongly upregulated in N-stress conditions (Mus et al., 2013; Levitan et al., 2015), were only slightly higher in NR-KO14 cells (18 h) relative to the wild type (Figure 10F). At 18 h, expression of the gene (J48859), a homolog in P. tricornutum of the major lipid droplet protein (MLDP), was ∼5 times more highly expressed in the NR-KO14 transcriptome compared with the wild type (Figure 10G); MLDP has been reported in other microalgae to regulate lipid droplet size (Yoneda et al., 2016).
Figure 10.
Rapid Expression of FA and TAG Biosynthesis Genes in NR-KO14 Cells.
(A) Schematic diagram of FA and TAG biosynthesis in the diatom cell. The outer green dotted line indicates the fourth membrane that surrounds the diatom chloroplasts. In diatoms, the ER is appressed to the chloroplast such that the two organelles appear to share an outer membrane. Signal peptide data indicate that fatty acid biosynthesis, assembly of TAG, and production of lipid droplets are localized in the chloroplast (green), ER (blue), and cytosol (orange). The diagram shows the location of the cognate proteins involved in these processes. The transcript abundance of genes shown in color is charted in (B) to (G); genes in black are not charted as their expression profiles were similar in both transcriptomes.
(B) to (G) Wild-type and NR-KO14 transcript abundances. NR-KO14 expression is shown by white squares and wild-type expression by black squares. Means ± sd of triplicate biological replicates are shown. Differential expression for these genes was significant (with N-free time point used as the control, FDR = 0.05) except where indicated by a gray dot beneath data point. Expression was significantly different in the paired t tests for five of six genes shown at the 1- and 18-h time points (FDR < 6.7 × 10−4), but the sixth, DGAT2, was not.
(B) Plastidic pyruvate dehydrogenase complex E1α (pPDC, EG02309), produces the fatty acid and TAG precursor, acetyl-CoA.
(C) An acyl-acyl carrier protein desaturase (acyl-ACP DS, J9361) catalyzes a conversion of C16:0 to C16:1.
(D) Δ9-fatty acid desaturase (FADS, J28797) introduces a double bond in C18:0 FAs.
(E) Phosphatidic acid phosphatase (PAP, J39949) cleaves the terminal phosphate from diacylglycerol 3-phosphate.
(F) Diacylglycerol transferase (DGAT2, J43469) catalyzes the last reaction in biosynthesis of TAG.
(G) A putative major lipid droplet protein (MLDP, J48859) has recently been identified (Yoneda et al., 2016). It regulates lipid droplet size. ACC, acetyl-CoA carboxylase; ACS, acetyl-CoA synthetase; LPA, lysophosphatidic acid; PA, phosphatidic acid; DAG, diacylglycerol. (The diagram shown in (A) is adapted from Yu et al. (2011)
To determine whether expression of a subset of genes within the lipid cluster (KOG class, n = 217) correlated with the rise and plateau of FA biosynthesis at 36 h, and the lipid demands of expanding vacuolar membranes (Figures 4E and 4F; Supplemental Figure 4C), the cluster was sorted for elevated expression levels in NR-KO14 cells at 1, 18, or 42 h (n = 71) (Supplemental Figure 8A). Within the most highly expressed genes (18 h), more than half (n = 48) encoded putative proteins involved in long-chain FA synthesis, lipid remodeling, and vacuolar membrane lipid precursor biosynthesis (Supplemental Figure 8C). They included lipases, sterol, and sphingolipid biosynthesis genes and phospholipid transport regulators (Supplemental Data Sets 1 and 2). Sphingolipids and sterols are major components of vacuolar membranes in plants (reviewed in Michaelson et al., 2016; Zhang et al., 2015). Three of the genes, J41807, J48977, and J22677, encoded components of sphingolipid biosynthesis and were 2 to 5 times more highly expressed at 18 h in NR-KO14 cells than in the wild type (Figures 11A to 11C), with BLASTP E-values from 3 × 10−14 to 9 × 10−103 for alignments to their Arabidopsis homologs (http://www.plantcyc.org). Expression levels for two putative sterol biosynthesis genes, J10852 and J45494, were also 10 and 2 times higher, respectively, in NR-KO14 cells than the wild type in response to the addition of NO3− (Figures 11D and 11E). Four genes annotated for lipid recycling, transport, and signaling (Figures 11F to 11I) were all more highly expressed in NR-KO14 than in wild-type cells (18 h). One of the lipases, EG00718, is predicted to be patatin-like galactolipases, which in microalgae and plants release FAs from membrane lipids, such as monogalactosyldiacylglycerol, under N-stress conditions (Li et al., 2012; Li et al., 2014). Other genes encoded putative proteins with roles in phospholipid biosynthesis and phosphatidylinositol signaling (PITPs) (Figures 11G to 11I), choline transport, fatty acid biosynthesis, elongation, and scavenging (Supplemental Figure 8C and Supplemental Data Set 1).
Figure 11.
Vacuolar Membrane Building Blocks and Potential Thylakoid Membrane Remodeling.
(A) to (H) Expression profiles of nine genes showing significant upregulation of transcript abundance in NR-KO14 cells at 18 and 42 h. Means ± sd of triplicate biological replicates are shown; 72% (13 of 18) of the comparisons 18- and 42-h time points were significant (the N-free time points served as the control, P < 0.03) (Supplemental Data Set 2). The pairwise t test comparisons of NR-KO14 versus wild-type expression for 1 to 42 h showed nearly all were significantly different (P < 0.025) (Supplemental Data Set 3). Gray dots indicate lack of significant expression.
(A) Serine-palmitoly-CoA transferase (J41807) catalyzes the initial step in sphingolipid biosynthesis in plants (https://www.arabidopsis.org/biocyc).
(B) 3-Dihydrosphinganine reductase (J48977) is the second enzyme in the sphingolipid pathway in plants (see link above).
(C) A putative Δ4-sphingolipid desaturase (DS) (J22766) catalyzes the final step in glucosyl-N-acylsphingosine (C18:1) synthesis (see link above).
(D) A predicted C4 sterol methyl oxidase (J10852) (Pfam PF04116, InterPro IPR006694, KOG) is required in plants and yeast for production of (phyto)sterol .
(E) A sterol C5 desaturase (J45494) (PF04116, IPR006694, KOG) catalyzes an intermediate step in the biosynthesis of major sterols.
(F) A patatin-like phospholipase (EG00718), identified by (PF1734, IPR002641), releases FAs from membrane lipids and TAG.
(G) A predicted phospholipase (J48391) (KOG, http://genome.jgi.doe.gov/) is involved in phospholipid disassembly and signaling.
(H) A phosphatidylinositol transfer protein (PITP J33651) plays a key role in lipid transport and PI signaling. It is defined by its lipid binding domain in both Pfam and InterPro. PI transport proteins, also including other P. tricornutum PITPs (Supplemental Figure 8C), move PI from the ER to vacuolar and outer membranes. In yeast and mammalian systems, PITPs, in addition to membrane lipid trafficking, function in lipid-mediated intracellular signaling and regulation of lipid metabolism (reviewed in Bankaitis et al., 2010; Cockcroft and Garner, 2013).
(I) J33873 is also annotated as a PITP.
The Photosynthesis Cluster, Chlorophyll Biosynthesis, and Carbon Fixation Are Inactivated in NR-KO Cells by Nitrate Addition
Following maintenance in NO3− for 4 d, morphological changes in the NR-KO14 chloroplasts were apparent (Figure 4F; Supplemental Figure 4B). As early as 1 h and continuing through 42 h, however, addition of NO3− to NR-KO14 cells triggered the transcriptional deactivation of the genes that encode components of PSII, photosynthetic electron transport (PET), and light-harvesting complexes in NR-KO14 cells (Supplemental Figures 9A and 9B and Supplemental Data Set 2; Figure 6, xiv). In contrast, wild-type expression profiles indicated an acute response to the presence of NO3− where induction of photosynthetic genes was upregulated from 1 to 42 h (Supplemental Figure 9B). At the N-free preconditioning time point, expression patterns also diverged markedly. NR-KO14 profiles showed nearly half of the genes were induced, whereas most of the same wild-type genes were not induced. It is worth noting that, in the subset of 38 core photosynthetic genes (Supplemental Figures 9A and 9B), the transition in the expression profile in the wild-type from 42 to 66 h, when cells have exhausted or assimilated available NO3− (42 h) to when they sense the dearth of NO3− (66 h), was nearly identical to the transition in NR-KO14 cells from N-free to 1 h after the addition of NO3−.
The 15 genes required for chlorophyll biosynthesis also showed opposing patterns of expression in the two transcriptomes (Supplemental Figures 9C to 9E and Supplemental Data Set 2; Figure 6, xiv). To form one porphyrin ring, eight molecules of 5-aminolevulinic acid are assembled in the chloroplast (Von Wettstein et al., 1995), and in all photosynthetic eukaryotes, 5-aminolevulinic acid is synthesized by the C5 pathway starting with the precursor l-glutamate (Oborník and Green, 2005). However, in NR-KO14 cells, the glutamate required to assemble chlorophyll is presumably not synthesized. Indeed, at 1 and 18 h, all 15 transcripts were coordinately repressed (Supplemental Figure 10C). In this instance, the absence of glutamate may indicate the absence of both a regulatory signal and the N-substrate.
The disruption of N-assimilation in NR-KO14 cells also resulted in repression of transcripts for the Calvin-Benson (CB) cycle and other genes involved in C-metabolism (KEGG Pathways; Figure 6, xvi). Nine CB cycle genes, starting with phosphoribulokinase (PRK, J10208) and including two fructose 1,6-bisphosphatases (FBP) and three fructose bisphosphate aldolases (FBA), were downregulated in NR-KO14 cells beginning at 1 h and continuing to 42 h as compared with the wild type (Supplemental Figures 10A to 10D). Downregulation of expression also occurred at 42 h in transcripts for three components of the diatom mitochondrial glycolysis pathway (Supplemental Figures 10C and 10D; Figure 6, xvii) (Smith et al., 2012). In contrast to the profiles for the CB cycle, gene expression in the TCA cycle was similar to the profile of NR cohort genes (Supplemental Figure 11A; Figure 6, xvii). Specifically, genes encoding TCA cycle enzymes citrate synthase (J30145), aconitase (ACON, J26290), and isocitrate dehydrogenase (ICDH, J14762), which produce α-ketoglutarate (2OG), a required metabolite for GOGAT-mediated N-assimilation (Nunes-Nesi et al., 2010), appeared to be tightly coexpressed with the NR cohort in both wild-type and NR-KO14 transcriptomes (Figures 7B and 7C; Supplemental Figure 11B and Supplemental Data Set 2). These shared expression profiles in response to N-stress set TCA cycle genes apart from the other C-metabolism gene clusters and suggest that TCA cycle gene expression may instead share the same transcriptional signals with N-assimilation genes (Flynn, 1991).
The Nitrate Reductase Knockout Affects Both Transcription and Translation
The loss of NR and its impact on transcription may also adversely affect translation. For example, relative to the response in wild-type cells, mRNA levels for a subset (n = 99) of ribosomal structural genes, encoding 40S and 60S ribosomal proteins (∼70) as well as nuclear-encoded chlorophyll/mitochondrial ribosomal proteins (∼25), were significantly downregulated within 1 h of exposure to NO3− in NR-KO cells (Supplemental Figure 12A and Supplemental Data Set 2). This was in stark contrast with wild-type cells, which maintain high levels of ribosomal structural transcripts until 66 h when the cells were N-depleted. Their expression profiles then resemble those of NR-KO14 (Supplemental Figure 12C). The rapid downregulation of these genes in NR-KO14 cells suggested their transcription was regulated by a mechanism that senses reduced NO3− assimilation prior to the cell being in a truly N-limited state.
In many of the gene expression profiles presented in this study (Figures 7C, 7D, 8B, 10, and 11; Supplemental Figures 5F, 6D, 8A, 9A to 9E, 10A, and 10B), transcript levels at the preconditioning −N time point were markedly different in NR-KO14 and wild-type transcriptomes. Globally, we identified 849 genes for which wild-type and NR-KO14 expression was similar at the preconditioning NH4+ time point but opposite at −N, where the wild type is downregulated and NR-KO14 is upregulated (Supplemental Figure 13 and Supplemental Data Set 4). Genes encoding functions related to NO3− transport and assimilation, photosynthesis, chlorophyll biosynthesis, and carbon fixation and metabolism were predominant in this set. The distinct differences of expression patterns in the two cell lines, prior to addition of NO3−, point to a role for the NR protein and transcript as a probable, direct, or indirect factor in their regulation.
DISCUSSION
A NR Knockout in P. tricornutum Gives a New Perspective on Wild-Type NR Function
The reduction of NO3− by NR in vascular plants and algae is the crucial first step of N-assimilation during periods of growth supported by NO3−. Localization of wild-type NR is dynamic within the cell depending on N source and availability. In wild-type cells grown in NO3− media, NR is an abundant cytosolic enzyme (Figure 1). Upon depletion of extracellular NO3−, NR proteins coalesce, both within the scant cytosolic areas surrounding the vacuoles and engaging the peroxisome. Subsequently, NR is sequestered in the peroxisome, where it may be involved in peroxisomal nitric oxide (NO) biosynthesis. Production of NO in the peroxisome (reviewed in Del Río, 2011) and NR production of NO (Crawford and Guo, 2005) are both well known in vascular plants. NR production of peroxisomal NO remains to be established. In this study, we produced a NR knockout in P. tricornutum, complemented two NR-KO cell lines with endogenous NR, and confirmed both physiologically and biochemically that NO3− is reduced solely by NR encoded by Phatr3_J54983. When NO3− is the sole N source, NR is required for survival.
Striking differences between NR-KO14 and wild-type cells demonstrate that the loss of NR profoundly affects the physiology, biochemical balance, and global transcriptome of P. tricornutum to a degree not observed in previous studies of N-stress (Alipanah et al., 2015; Levitan et al., 2015). Our analysis shows that the addition of NO3− to NR-KO cells has an acute, immediate impact on key metabolic pathways such as N-assimilation and C-fixation, structural pathways for light-harvesting complexes, ribosome production, chlorophyll biosynthesis, and carbon partitioning. In sharp contrast to the wild type, NR-KO14 cells confront an unusual condition: NO3− is abundant and inaccessible simultaneously. In response, NR-KO14 cells show a dramatically altered morphology that is likely triggered by unmodulated import of NO3− into the cell and presumably into the vacuole (Figure 4). As the vacuole expands, organelles are pushed aside and chloroplasts are compressed.
The Loss of NR: The Scope of Its Influence
Comparisons between the coexpressed NR cohort and N-assimilation genes (Figures 7A to 7D; Supplemental Figures 6A and 6B) revealed two different patterns of expression in NR-KO14 and wild-type cells, indicating that transcriptional regulation of these genes is highly sensitive to changes to the N-assimilation pathway. Wild-type expression profiles are characterized by two temporally distinct peaks, whereas their expression in NR-KO14 cells peaks once (1 to 18 h). Our data suggest that the large, single peak in expression is the result of two distinct stimuli: NO3−-replete and N-deplete activators/effectors undefined to date that trigger individual transcription initiation events that occur nearly simultaneously. In NR-KO14 cells, RPKM peak values at the addition of NO3− reflect that dual stimulus.
In contrast, GOGAT genes, GltS and GltX, enzymes at the heart of N and C metabolic integration, show only slight differences in expression profiles and RPKM abundances between the two transcriptomes, and greater transcription in NR-KO14 cells does not occur (Supplemental Figure 6C). Since their expression is not dependent on supply of de novo N, or a lack thereof, it may be driven instead by the availability of key substrates, including 2OG and NH3+; the former originating from possible mitochondrial or chloroplast-derived sources of plastidic ACOAT, or from alanine/aspartate transaminase activity (Nunes-Nesi et al., 2010), and the latter arising from GLDH-mediated deamination of scavenged glutamate (Alipanah et al., 2015). Indeed, expression profiles of genes encoding ACOAT, unCPS, GLDH, and ICDH (Supplemental Figures 5D to 5F and 11B) are more similar to the upstream NR cohort genes and support their substantive links to N-assimilation and C-metabolism. Plastidic ACOAT, a component of the recently proposed in silico ornithine-glutamine shuttle, transfers reducing equivalents generated by photosynthesis to the mitochondria (Levering et al., 2016). During N-stress, its secondary product, chloroplast-produced 2OG, may supplement the 2OG produced by ICDH in the mitochondria. UnCPS, suggested to facilitate recovery from N-limited conditions (Allen et al., 2011), is highly upregulated in N-depleted conditions in the wild type (138–162 h), whereas in the mutant, after an elevated response to the addition of NO3−, its pattern of expression suggests the possibility that its mRNA transcripts are stabilized, ready for a return to normal activity. GLDH, the least similar to the NR cohort, has cumulative RPKM values after 66 h that are among the highest in the transcriptome, pointing to GLDH in wild-type cells as an essential recycler of glutamate, carbon skeletons, and NH3+ in response to prolonged N starvation (Supplemental Figure 5F).
Shortly after NO3− is introduced to NR-KO14 cells, expression levels of key functional and metabolic gene clusters indirectly related to N-assimilation are sharply downregulated. Despite ample extra- and intracellular NO3−, transcript profiles for these clusters respond similarly to those observed in wild-type cells under N-stress (Alipanah et al. 2015; Bender et al., 2014; Juergens et al., 2015; Levitan et al., 2015; Schmollinger at al., 2014). However, in contrast to wild-type cells which gradually acclimate to a dwindling supply of N-product, the response in NR-KO14 cells is quite rapid. De novo N-production ceases abruptly, reductions in glutamine drive down the canonical Gln/2OG ratio, and whole metabolic pathways are in effect switched off. Genes encoding proteins for PSII, PET, and light-harvesting complex components are immediately impacted (Supplemental Figure 9A). Transcriptional deactivation of chlorophyll biosynthesis is complete at 3 h in the mutant (Supplemental Figure 9D). This pattern of deactivated expression is similar for CB cycle genes, mitochondrial glycolysis, and other carbon metabolism genes as well (Supplemental Figure 10). Expression of genes for ribosome structure and translation initiation is also strongly repressed (Supplemental Figure 12).
Analyses of two clusters of genes (6–8% of the transcriptome), from NH4+ through −N to NO3− time points, whose expression diverges sharply in −N conditions, suggest that the absence of NR may directly or indirectly affect transcription of the genes identified in the clusters (Supplemental Figure 13). Interestingly, many of these genes are involved in key metabolic pathways, including NO3− assimilation, photosynthesis, carbon metabolism, TCA cycle, and regulation of transcription, and many are discussed in this article (Supplemental Data Set 4). Wild-type and NR-KO14 expression patterns are quite similar in growth on NH4+, a well-established inhibitor of NO3− uptake and assimilation, but they diverge sharply in the transition from NH4+ to −N: NR-KO14 expression is upregulated and that of the wild type is downregulated. The wild-type response to N-stress in this cluster is apparent in the elevated expression that occurs both at 18 h when extra- and intracellular NO3− is nearly exhausted and at 66 h when biochemical N-stress is pronounced. In the NR-KO14 cluster, elevated expression after 2 h in −N conditions suggests enhanced sensitivity to perception of N-stress, perhaps mediated by a signaling cascade that results in absence of the NR protein.
Vacuolar NO3− Transport and Storage
The rapid induction and robust gene expression of the MFS transporter J26029 and several vacuolar NO3− transporters, including EG01952 (Supplemental Figures 6A and 6B; Figure 8), correlate with the overall pattern of NO3− uptake data (Figures 5A and 5B). This underscores the remarkable ability of diatom cells to rapidly transport, assimilate, and, when necessary, store NO3−, presumably in the vacuole (Figure 4D). Coupled with the results of the uptake assays showing intracellular NO3− accumulating in NR-KO14 cells, QFDE images (Figures 4E to 4H) provide a solid foundation for our hypothesis. Vacuoles are enlarged in NR-KO14 cells, maintained in NO3− but not NH4+, and their content has a different appearance in response to exposure to NO3− versus NH4+. Furthermore, the content appears to be confined by the vacuolar membrane and, as such, does not compromise the cytosolic organelles (Figure 4F; Supplemental Figure 4C). Future efforts to isolate diatom vacuoles, identify their content, and functionally characterize NO3− transport proteins associated with the vacuole will provide additional support for these results.
Diatoms, including the model organism P. tricornutum, are among the more vacuolate of the dominant lineages of the eukaryotic phytoplankton (Raven, 1987). Our transcriptomic identification of a putative family of vacuolar transporters reinforces data from marine environments that indicate diatoms have high maximum nutrient uptake rates and storage capabilities (i.e., luxury uptake) relative to rates of immediate assimilation (Dortch et al., 1984; Lomas and Glibert, 2000; Villareal et al., 1993). This ecologically advantageous physiological trait of diatoms under nonequilibrium conditions in the marine environment (Cermeño et al., 2011) may rely, in part, on the capacity of vacuolar transporters to respond quickly to elevated NO3−, as we have observed here in the 1-h response by NR-KO14 cells (Figures 8A and 8B). Additional factors suggest that EG01952 and three other NO3− ClC cohorts, J28245, J46097, and J52412, encode proteins that are involved in transport NO3− into and out of the vacuole: (1) Their expression profiles in both transcriptomes correspond to the need for NO3− to be pumped either into or out of the vacuole. (2) In vascular plants, intracellular NO3− is stored in the vacuole by ClC family proteins (Granstedt and Huffaker, 1982; Martinoia et al., 1981). (3) In Arabidopsis, the ClC-a and ClC-b proteins have been shown to be vacuolar NO3−/H+ antiporters (Bergsdorf et al., 2009; De Angeli et al., 2006; Vidmar et al., 2000). (4) Three of the four P. tricornutum vacuolar NO3− ClC homologs have BLASTP E-values between 4 × 10−80 and 7 × 10−10 and share conserved anion (NO3−) binding residues with Arabidopsis ClCa (Nguitragool and Miller, 2006) (Figure 8D).
The presence of swollen vacuoles in NR-KO14 cells suggests that the they can transport NO3− into the vacuole but are unable to pump it out (Figure 6, v), perhaps because of low concentrations of ATP/H+, the driving forces of vacuolar pumps, and/or the lack of functional NR. Given the closely juxtaposed organelles, four plastid membranes and confined cytosolic spaces of P. tricornutum (Apt et al., 2002; Prihoda et al., 2012), an association between NR, plasma membrane transporters, and vacuolar membrane transporters may form as the proteins coalesce under specific conditions to facilitate the flux of NO3− from the environment into and out of the vacuole. It remains to be determined whether free NO3− is ever resident in the cytosol and whether, upon transport into the cell, it is either immediately reduced and transported into the chloroplasts or pumped into the vacuole.
Nitrate-Replete Conditions Trigger Biosynthesis of Lipids in NR-KO14 Cells
The collective data gathered for NR-KO14 and wild-type cells provide a unique view of two N-stressed, lipid phenotypes. With the addition of NO3−, wild-type cells sense N-replete conditions and downregulate gene expression for ∼48% of the lipid cluster genes (n = 217) (Supplemental Figure 7H). By 36 h, with available NO3− assimilated, wild-type cells redirect carbon resources to TAG biosynthesis (Figures 9B and 9C), a well described phenomenon for various microalgae. Conversely, in NR-KO14, the NO3− addition elicits the signals for N-stress, which results in dramatic increases in transcript abundance for ∼35% of the cluster (Supplemental Figure 7I) and drives initiation of biosynthesis of FAs C16:0, C16:1, the two major components of TAG in P. tricornutum (Supplemental Figures 7A, 7C, and 7E) (Abida et al., 2015; Shen et al., 2016). In the wild type, TAG accumulates through the end of the experiment, but in NR-KO14 cells, C16 FA accumulation and total lipid content subsides by 36 h. We surmise that the end of TAG biosynthesis in NR-KO14 cells is directly related to rapid downregulation of transcription at 1 h of the genes for the major photosynthesis complexes, specifically PET (Supplemental Figure 9A). N deprivation has been reported to affect photosynthetic output in P. tricornutum and other microalgae through a process that results in a gradual, managed deconstruction of the photosynthetic apparatus (Juergens et al., 2015; Levitan, et al., 2015; Simionato et al., 2013). Here, the immediate transcriptomic response of NR-KO14 cells presumably interrupts production of ATP and NAPDH, causing FA biosynthesis and FA desaturation to shut down. Differences in the timing of TAG (neutral lipid) accumulation in the two cell lines are captured qualitatively in the confocal images of the cells at 36 h stained with Bodipy to visualize lipid bodies (Figures 9D and 9E; Supplemental Figure 14).
The burst of C16:1 FA accumulation in NR-KO14 correlates with the robust expression at (1–18 h) of two genes whose cognate proteins play predominant roles in C16:1 FA biosynthesis: pPDE1α, the plastid-targeted pyruvate dehydrogenase E1α, produces the starting material acetyl-CoA; and PAD, the palmitoyl-ACP desaturase, introduces the first double bond in C16:0 forming C16:1 (Figures 10A to 10C). Elevated expression for three other genes in the TAG pathway, PAP, DGAT2, and MLDP (Figures 10E to 10G), support the immediate accumulation of TAG (3 h) observed in lipid/FAME results (Figures 9B and 9C). Long-chain C18 FAs are also more abundant in NR-KO14 cells, albeit at lower concentrations (Supplemental Figure 7F), and those increases correlate with higher expression of C18 FA desaturase, ADS J28797 (Figure 10D). Increases in TAG inversely track with the decline in C20:5 FA concentrations in both cell types (Supplemental Figures 7C and 7D). In P. tricornutum and other microalgae, the decrease in C20:5 FA during N-stress is believed to reflect the efforts by alga to pursue a second source for TAG by recycling C20:5 and C16 FA moieties from chloroplast thylakoid membrane glycerolipids (Abida et al., 2015; Li et al., 2012; Martin et al., 2014; Simionato et al., 2013). Specific galactoglycerolipid lipases, known in plants as patatin lipases (reviewed in Scherer et al., 2010), have been functionally characterized in C. reinhardtii and Nannochloropsis oceanica (Li et al., 2012; Li et al., 2014) as thylakoid glycerolipid recyclers. Ten transcriptionally active, patatin-like lipases candidate genes are expressed in both NR-KO14 and wild-type transcriptomes, and one, EG00718, is more highly expressed in NR-KO14 cells through 42 h (Figure 11F). The fact that within the lipid cluster a subset of genes (n = 71) are upregulated in the mutant in response to addition of NO3− and of those, at least half correlate to lipid accumulation and recycling suggests there may be a direct transcriptional link between N-sensing and lipid induction that remains to be described in wild-type cells.
Nitrate Uptake into the Vacuole, a Competitive Advantage in Marine Environments
Considering the accumulation and decline of TAG and FA in the context of the physiological changes to the chloroplasts in NR-KO14 cells (Figures 4B, 4C, and 4F; Supplemental Figures 4B and 4C), we propose that the glycerolipids of thylakoid membranes are the likely substrates of the P. tricornutum galactoglycerolipid lipases and phospholipases and that free FAs of dismantled glycerolipids are recycled into new membrane lipids for the enlarging vacuoles. Vacuolar membrane lipids are poorly characterized in diatoms, but those that make up the vacuolar membranes in plants that store NO3− provide a starting point for consideration of the composition of P. tricornutum vacuolar membranes. In vascular plants, the majority of vacuolar membrane lipids are phospholipids PE and PC, sterols, and sphingolipids, with sterols and sphingolipids comprising 30% and 10 to 20%, respectively (Ozolina et al., 2013; Yoshida and Uemura, 1986; reviewed in Zhang et al., 2015). Within the lipid cluster subset, we highlighted five genes encoding proteins with functions that are putatively well suited to vacuolar membrane lipid biosynthesis in P. tricornutum. Two, J10852 and J45494, are involved in sterol biosynthesis and three potentially play key roles in sphingolipid biosynthesis (Figures 11A to 11D). The serine-palmitoyltransferase gene J41807 encodes the protein that produces the C18-NH3+ base for all sphingolipids in Arabidopsis (reviewed in Michaelson et al., 2016), and the other two genes, J48977 and J22677, a reductase and a desaturase, complete the second and last steps in the synthesis of the sphingolipid. Recently, similar sphingolipid desaturases have been characterized in the diatom T. pseudonana (Michaelson et al., 2013; Tonon et al., 2005). In yeast, and in Arabidopsis, regions within the vacuolar bilayer, enriched in sphingolipids and sterols (Chung et al., 2003; Finnigan et al., 2011), have been identified as essential to the function of vacuolar-ATPases (Ozolina et al., 2013; Yoshida et al., 2013). Vacuolar-ATPases are the proton pumps in plants (reviewed in Martinoia et al., 2007) that provide energy for NO3− transport into the vacuole via NO3− ClC transporters identified in this study. The presence of putative NO3− vacuolar transporters in P. tricornutum, combined with results supporting the likelihood that diatom vacuolar membrane lipids can be constructed from recycled thylakoid membrane lipids, prompts consideration of whether marine NO3− upwelling events produce a similar response in P. tricornutum and other vacuolate diatoms. In efforts to capture excess NO3− for future assimilation, we propose that diatoms quickly express new vacuolar NO3− transporters and pumps and assemble new vacuolar membranes by recycling some portion of their thylakoid membrane lipids and/or by diverting new carbon from metabolism to de novo lipid biosynthesis.
The NR knockout and the resulting analysis of the NR-KO14 transcriptome have provided, and will continue to provide, an invaluable resource for new investigations of diatom gene expression, pathway interactions, and regulatory networks. Unraveling the complex interaction between NR and intracellular NO3− trafficking and identifying specific drivers, genetic components, and lipids required for vacuolar storage and efflux over the life cycle of the cell represent a challenge for additional research.
METHODS
NR Overexpression Vectors
Phaeodactylum tricornutum cells were initially transformed with a two-vector system: first, with the overexpression vector with the gene of interest joined to a fluorescent protein gene and controlled by an inducible promoter and terminator pair, and second, with a vector with an antibiotic resistance gene, also controlled by an inducible promoter to provide selection for screening successful transformants (Falciatore et al., 1999; Zaslavskaia et al., 2000). For NR overexpression, the gene (Phatr3_J54983, Pt2_54983) was fused to the YFP gene at its 5′ end. Expression of the gene fusion was controlled by the fucoxanthin-chlorophyll binding protein-B promoter (PromFcpB) and the FcpA terminator (TermFcpA) and induced in the presence of light. To confer resistance to Phleomycin, a second vector with the PromFcpB-Sh ble-TermFcpA cassette was cotransformed with the YFP-NR expression vector. Subsequently, one-shot vectors were designed with a native promoter and terminator driving gene-of-interest expression. Downstream of the gene-of-interest-fluorescent protein fusion, each vector incorporates antibiotic resistance cassettes for screening in both Escherichia coli and P. tricornutum; the backbone is a modified pUC19 vector with E. coli origin of replication. One-shot vectors, using the nitrate reductase promoter (ProNR) and terminator (TermNR) for inducible expression of the appropriate fusion protein, were constructed for both nitrate reductase and 3-ketoacyl-CoA thiolase (Pt2_41969).
Transformation of P. tricornutum
For all but one of the transformations of P. tricornutum cells described in this work, we used previously described biolistic methods (Falciatore et al., 1999). Briefly, for the initial, two-vector transformations, tungsten beads (Bio-Rad) were coated with 6 μg of ProFcpB-YFP-NR-TermFcpA vector and 6 μg of the ProFcpB-Sh ble-TermFcpA cassette plasmid DNA and were shot into ∼3 × 108 cells plated on F/2 medium containing 880 μM NO3− as the sole nitrogen source. For the ProNR-YFP-NR-TermNR vector, 10 μg of plasmid DNA was coated on the beads and shot into 3 × 108 cells plated on seawater (SW) agar medium with F/2 nutrients and containing 880 μM NH4+ as the sole nitrogen source. For the NR knockout transformation vectors pNRKO and pTH-NR (see below), 6 μg of pNR-KO and 6 μg of pTH-NR were loaded on to the beads and bombarded into 3 × 108 cells on F/2, 880 μM NH4+ SW agar plates. After growing for 2 d on nonselective media, plated cells were transferred to seawater agar plates, consisting of 51% filtered seawater, 49% Milli-Q water, and LB agar supplemented with 55 μM NaH2PO4, 100 μM Na2SiO3, 880 μM of the appropriate N source, and 100 μg mL−1 of phleomycin for selection. Colonies, positive for phleomycin resistance, appear between 10 and 30 d.
Gold Tagging for Transmission Electron Microscopy
Three hundred milliliters of an FBAC5-EYFP clone batch culture were grown to late exponential phase and harvested by gentle centrifugation (2000g, 10 m). Cells were fixed in 2% glutaraldehyde and embedded in LR white medium grade resin (London Resin Company). Ultrathin sections were treated with rabbit anti-GFP antiserum (AbCam) followed by goat anti-rabbit conjugated to electron-dense 1-nm gold particles as described previously (Lichtle et al., 1992). Sections were observed with a Jeol CX2 electron microscope at 80 kV.
Culturing and Harvesting of Wild-Type P. tricornutum Cells
The variable N-source, short, time-course experiment was adapted from earlier investigations of NO3− and NH4+ uptake in P. tricornutum as described by Cresswell and Syrett (1979). In that work, it was observed that 300 μM NO3− was depleted from media by P. tricornutum cells grown to exponential phase (∼2 × 107 cells) ∼3 h after its addition. In our study, duplicate 2-liter cultures of wild-type P. tricornutum (CCAP-1055) cultures were grown in artificial seawater (ASW) media with F/2 nutrients (55 μM NaH2PO4, 100 μM Na2SiO3, and 880 μM of the appropriate N source), trace metals, and vitamins. In place of NO3−, 880 μM NH4+ was added as the sole nitrogen source both to repress NR induction and to provide biomass for the experiment. Cultures were stirred and bubbled with air in cyclical diel light (14-h:10-h, light:dark, 150 μE m−2 s−1) at 18°C. At mid-exponential phase growth (Fv/Fm ≥ 0.6, ∼3 × 106 cells mL−1) cells were collected by centrifugation (10 min, 700g, 18°C), washed three times in N-free media, resuspended in N-free media in duplicate 2-liter flasks, and then incubated in constant light of the same intensity, with bubbled air at 18°C for 2 h. Cultures were then spiked with NO3− to a final concentration of 150 μM and returned to incubating conditions. After 90 min in NO3− media, cells were collected by centrifugation and washed three times (see above). Pellets were then resuspended in 20 mL N-free media, combined, and aliquoted at ∼7.5 × 105 cells mL−1 into duplicate, 800 mL of standard media containing different nitrogen sources, (1) no nitrogen, (2) 300 μM NH4+, (3) 300 μM NO3−, and (4) 300 μM NO−2, in 1-liter Roux culture bottles (Corning). All cultures were incubated in constant light at 18°C with bubbled air for the duration of the time course. Culture flasks were racked on an angle enabling the bubbled air to thoroughly circulate the cells. Samples for in vivo immunofluorescent microscopy were collected in duplicate at six time points: (1) mid-exponential phase growth on NH4+, (2) 2 h in N-free media, (3) 90 min after NO3− addition, (4) 15 min, (5) 45 min, and (6) 18 h after being aliquoted into the four N-conditions. At 18 h, duplicate biological samples were collected for nutrient analysis to provide final concentrations of NH4+, NO3−, and NO2−.
Colorimetric nutrients-continuous flow analysis was provided by Virginia Institute of Marine Sciences Analytical Service Center for Nutrients (http://web.vims.edu/admin/asc/link4.html). Averages of the duplicates were reported with se that was calculated by the following equation: se = sd/√n, with n = sample size; here, n = 2. Two centrifuges and helpful lab colleagues made it possible to collect and process samples within short time intervals. Final antibiotic concentrations were: phleomycin, 100 μg/mL in plates; zeocin, 50 μg/mL in liquid media.
Immunofluorescent Labeling of Wild-Type NR
Cells were fixed for 30 min at room temperature in fresh 10% paraformaldehyde at a final concentration of 0.01% and stored at −20°C (Dyhrman and Palenik, 2001). The method to prepare fixed cells for labeling was adapted from Tesson and Hildebrand (2010). Cells (1.8 mL; ∼5.0 × 106 mL−1) were pelleted by centrifugation (10 min, 700g); paraformaldehyde was removed by pipette. Cells were then rinsed three times in 1 mL, 1× PBS buffer, salted with 1% NaCl, and pelleted between washes by centrifuging (5 min, 700g). The supernatants were aspirated with a pipette tip attached to a vacuum pump. Tubes were rotated on a rack for 10 min between washes 2 and 3. Cells were then blocked for 60 min in filter-sterilized 1% gamma globulins in salted 1× PBS buffer (GG-PBS buffer), spun down, and aspirated as above. Samples were divided at this point to provide for +/− antibody comparisons. Pellets, resuspended in 1% GG-PBS buffer amended with a 1:1000 dilution of anti-rabbit antibody to P. tricornutum’s NR protein (SDIX), were incubated overnight at room temperature on a rotating rack. To remove excess antibody, cells were rinsed three times in salted PBS (as above). A 1:500 dilution of the secondary antibody, Alexa Fluor 488 (green) or 532 (yellow) (Life Technologies) in 1% GG-PBS buffer, was used to resuspended the pellets. Labeled cells were again incubated overnight in the dark at room temperature on a rotating rack and then rinsed three times in salted PBS to remove excess label (as above) and pelleted. Cells were finally resuspended in ≤50 μL of salted 1× PBS for microscopy.
Colocalization of Peroxisomal Matrix Protein and Nitrate Reductase
DNA encoding 3-ketoacyl-CoA thiolase (Phatr2_41969, Phatr3_J41969) protein, shown previously to localize in P. tricornutum’s peroxisomal matrix (Gonzales et al., 2011), was amplified from genomic DNA and integrated into the vector backbone (described above) by Gibson assembly cloning (Gibson et al., 2009), resulting in the N’CFP-41969 fusion. An N’YFP-NR vector had already been assembled. Sequences encoding the fluorescent proteins were added to the 5′ end of NR and KAT to protect any peroxisomal targeting peptides found at the C terminus of the proteins (Neuberger et al., 2003). Both constructs were under the control of native NR promoters and terminators in order to regulate expression of the fusion protein, as shown successfully in the diatom Cylindrotheca fusiformis (Poulsen and Kröger, 2005), in order to more closely monitor the cellular response to nitrate addition. Wild-type cells were simultaneously cotransformed, and transgenic lines positive for both CFP and YFP expression in the same cell were analyzed by confocal microscopy. Duplicate clones, showing both YFP and CFP signals in the same cells, were selected for a time-course experiment. Samples were collected at four time points: from exponential growth on NH4+, then at 2, 4, and 18 h after resuspension in NO3−-amended media.
Confocal Microscopy
Images of native, and overexpressed, NO3− reductase proteins, as well as those of NR colocalized with a peroxisomal matrix protein were recorded on a Leica TCS SP5 II confocal laser-scanning microscope. All images were collected using the HCX PL APO 100×/1.4-0.07 Oil CS objective. CFP, YFP, and chlorophyll autofluorescence (Chl A-F) was detected through the double dichroic filter 458/514, using separate photomultipliers with bandwidths at 463 to 499 nm, 523 to 600 nm, and 660 to 750 nm for CFP, YFP, and chlorophyll, respectively. For colocalization imaging, with both CFP and YFP spectra emitted from one cell, sequential scans were accumulated and overlaid, enabling CFP, YFP, Chl A-F, and bright-field exposures to be seen in one image.
NR Knockout Vector Construction and Cloning
Construction of the nitrate reductase (Phatr2_54983) knockout followed the TALEN protocol recently described for the P. tricornutum urease gene knockout (Weyman et al., 2015). Two vectors were assembled. (1) The pTH2-NR vector has two TALEN (target-recognition, DNA binding) sequences each linked to FOK-1 endonucleases, which, together, make a double-stranded DNA cut at a specific target site. In each, a malleable DNA binding motif region is designed to specifically recognize a 20-bp sequence flanking the target site locus within NR. (2) pKO-NR1 is a modified pUC19 vector contains two ∼1-kb regions within NR, upstream and downstream of the cut site, which are arranged to flank the Sh ble antibiotic-resistance cassette and enable the homologous recombination of the selectable marker into the double-strand break in the NR gene (Supplemental Figure 15). Both vectors are transformed into E. coli; subsequent restriction enzyme digest and sequencing analysis provided verification of correct construction. After transformation by the biolistic method, and 15 to 24 d after growing on selective plates in diel light (14:10, l:d, 150 μE m32 s31) at 18°C, 16 colonies had grown (10 additional, more slowly growing cells appeared over the following 3 weeks but were not assayed for activity). Sixteen colonies were patch plated on [880 μM] NH4+ amended, SW agar plates with phleomycin and also transferred into liquid media (ASW F/2 media, supplemented with 50 μg mL−1 of zeocin and 880 μM NH4+) in 24-well culture plates.
Verification of NR Knockout
The loss of NR function was initially confirmed visually with a 24-well plate assay. One hundred microliters of each of 16 transgenic cultures growing in F/2 NH4+ media was used to inoculate fresh media in two plates: In the first, 880 μM NH4+ F/2 media; in the second, 880 μM NO3−; zeocin selection was present in both. Wild-type cultures were also grown side-by-side in both N-sources. Plates were monitored on a daily basis for 3 weeks. The presence of successfully recombined Sh ble cassettes within the NR locus was confirmed by PCR. Genomic DNA of transgenic lines was harvested using Plant DNAzol (Thermo Fisher Scientific), and primers were designed to capture the entire locus from within the NR promoter into the NR terminator and from within the Sh ble cassette itself to the promoter and terminator regions.
Protein Extraction for Immunoblots
Twenty-five to 150 mL (time point dependent) of P. tricornutum culture (7.6 × 105 to 3.5 × 106 cells mL−1) was pelleted by centrifugation (6000g, 10 min, 4°C), flash-frozen in liquid nitrogen, and stored at −80°C. Pellets were thawed in lysis buffer (125 mM Tris-HCl, pH 6.8, 200 mM NaCl, and 1 mM PMSF). Cells were lysed at 4°C by sonication (Bioruptor; Diagenode), and lysate clarified by centrifugation (10,000g, 30 min). Protein concentrations were determined by BCA assay. Ten micrograms of protein in 29 μL of loading buffer (125 mM Tris-HCl, pH 6.8, 4% SDS, 20% glycerol, and 10% β-mercaptoethanol) was separated on 4 to 12% precast polyacrylamide Bis-Tris gels and transferred to PVDF membranes. Membranes were probed using 1:5000 dilution of an antibody to P. tricornutum nitrate reductase (SDIX).
Growth Curves
Transgenic NR-knockout lines were grown on NH4+-amended media to mid-exponential phase and then transferred to 24-well culture plates containing either NH4+- or NO3−-amended media and measured for growth for 6 d, in diel light (14:10, l:d, 150 μE m32 s31) at 18°C. At the start of the experiment, 100 μL from each NR-KO line was used to inoculate each well of F/2 media (1.8 mL of media per well) supplemented with either NH4+ or NO3−. Each line was inoculated into four wells/N-source; inoculated wells were separated from each other by a row of N-free media and a row of empty wells to avoid crosstalk. As a control, wild-type cells were grown under similar conditions. In vivo chlorophyll a fluorescence, a proxy for growth, was measured on a fluorescence plate reader (FlexStation 3; Molecular Devices). All cells were gently resuspended prior to chlorophyll a reading. Photosynthetic efficiency (Fv/Fm) was measured on the PAM fluorometer (Water-PAM; Walz).
Quick-Freeze Deep-Etch Electron Microscopy
A small aliquot of pelleted cells was transferred with a toothpick to a cushioning material and dropped onto a liquid helium-cooled copper block (4K); the frozen material was transferred to liquid nitrogen and fractured, etched at −80°C for 2 min, and Pt/C rotary replicated as described previously (Heuser, 2011). The replicates were examined with a JEOL electron microscope (model JEM 1400) equipped with an AMTV601 digital camera. The images are photographic negatives; hence, protuberant elements of the fractured/etched surface are most heavily coated with platinum and appear white.
FTIR Analyses
FTIR spectrophotometry provides for the simultaneous measurement of functional groups and chemical bonds in cells to allow identification of their biochemical composition (D’Souza et al., 2008; Mayers et al., 2013). We monitored macromolecular groups, lipids, carbohydrates, and proteins, over a 7-d time series using an IR21 spectrophotometer (Shimadzu). Wild-type and NR-KO cell lines were grown in NH4+-amended [880 μM] F/2 media to mid-exponential phase (2.0E+06 cells/mL). Cultures were washed in N-free media and resuspended in NO3−-amended [880 μM] media, returned to the growth chamber, and bubbled, in diel light at 18°C. Cells were sampled 12 h after the addition of NO3− and then once per day for 7 d. Each scan requires ∼4 mg, dry weight, of cells, equivalent to ∼2.0 – 4.0 × 108 cells. Samples for each FTIR scan were collected by centrifugation (5000g, 10 min), washed with an equal volume of 0.5 M NH4CO−3 to remove salts, centrifuged again to remove all liquid, and flash-frozen in liquid nitrogen or dry ice. Prior to scanning, samples were dried with a lyophilizer, maintained in a desiccator, crushed to a powder, and placed on the sample holder. Percentages of protein, carbohydrate, and lipid per biomass, derived from peak values in scans, were compared by paired t tests for significant differences over the 7-d time course (Levering et al., 2016). Each NR-KO line (1 and 14) was individually compared with the wild type.
Complementation by Conjugation
Two P. tricornutum NR-KO lines, 9 and 14, were complemented by a conjugative plasmid delivered into P. tricornutum by E. coli as described by Karas et al. (2015). The plasmid was constructed to contain yeast replication origins and endogenous P. tricornutum NR fused to YFP under the control of the native NRprom and NRterm, as previously described (Karas et al., 2013). The conjugation produced numerous colonies that grew on NH4+-amended SW-agar plates under Zeocin selection. Sixteen transformants from each KO line were screened on nitrate for viability; all grew on solid and liquid media amended with NO3−.
Harvesting of NR-KO14 and Wild-Type Cells: 10-d Physiological and Biochemical Experiments
For a 10-d time-course experiment, NR knockout transgenic line #14 (KO-14) and wild-type cultures were grown under standard conditions: bubbled air, stirred, in diel light (14:10 l:d, 150 μE m2 s1) at 18°C, in ASW media with F/2 nutrients (55 μM NaH2PO4, 100 μM Na2SiO3, and 880 μM of the appropriate N source), trace metals, and vitamins, supplemented with 880 μM NH4+ in place of NO3− as the sole nitrogen source. Six liters of the NR-KO14 culture and three liters of the wild-type culture were grown to mid-exponential phase (Fv/Fm = 0.62, 2.8 × 106 cells mL−1). The additional volume of KO-14 culture was grown, given the decrease in growth rate of KO-14 when NO3− is added to the media, to ensure that there would be adequate cells/milliliter for analyses at the later time points of the experiment. Cells were collected by centrifugation (10 min, 700g, 18°C), washed three times in nitrogen-free media, resuspended in N-free media, and returned to incubating conditions. After 2 h, cells were again pelleted, resuspended in a small volume of N-free media, and aliquoted into one-liter Roux culture flasks. Wild-type cells were grown in duplicate in 700 mL ASW F/2 media, supplemented with 300 μM NO3−. NR-KO14 cells were grown in triplicate in identical media. Starting cell densities were balanced by chlorophyll a fluorescence measurements. All cultures were incubated in diel light (see above) at 18°C with bubbled air for the duration of the time course. Samples were collected in duplicate for TOC/FAME, nutrients, cellular NO3−, and microscopy at 11 time points: (1) mid-exponential phase growth on NH4+, (2) after 2 h of incubation in N-free media, (3) 1 h, (4) 3 h, (5) 18 h, (6) 36 h, (7) 60 h, (8) 84 h, (9) 108 h, (10) 132 h, and (11) 10 d after the addition of NO3− to the media.
Cellular NO3− Analysis
To analyze extra- and intracellular NO3− pools, duplicate samples of 5 to 15 mL of cells (time point, cell count dependent) were harvested under low pressure onto GF/F filters. Prior to filtration, each filter was washed with 25 mL of N-free ASW (Cresswell and Syrett, 1979; Dortch, 1982). Filters were saved in a microcentrifuge tube, covered with 1 mL of N-free ASW medium, and quick-frozen in liquid nitrogen. The resulting 5 to 15 mL of filtrate were collected in 15-mL tubes and stored in the dark at 4°C. Cell contents were extracted by heating the microcentrifuge tubes in a 100°C heat block for ≤10 min. Tubes were then cooled on ice for 5 min, vortexed briefly, then centrifuged for 10 min at 10,000g at 4°C. Filters were washed with supernatant and discarded. Cell extracts were centrifuged again briefly, and the clarified extract was transferred to a clean tube and stored at 4°C. NO3− in the extract and in the filtrate was measured by UV spectrophotometry (Beckman Coulter DU 800 UV/VIS, single-beam spectrophotometer with a 1.8-nm bandwidth and 10-mm light path) at a wavelength of 220 nm (Collos et al., 1999). Two 200-μL technical replicates for each biological duplicate were loaded into a 12-well micro-cell cuvette with blanks and measured. Outputs were correlated to a NO3− standard curve (NO3− concentrations from 5 to 800 μM in ASW N-free media) established at the beginning of the experiment and rerun prior to each use of the DU 800. To determine whether there were significant differences in the extracellular and intracellular NO3− concentration between the two genotypes for the time course, a two-way ANOVA was performed in R, and statistics for time, strain, and interaction were recorded. Paired comparisons (t tests) by time point were also determined.
Total Organic Carbon, Total Nitrogen, and FAME Analysis
NRKO and wild-type cells were collected for TOC/TN and FAME analyses on 47-mm, 8.0 μM Durapore SVLP filters (Millipore), flash-frozen in liquid nitrogen, and stored at −80°C. Volumes (25–125 mL) were cell count and time point dependent. Prior to analyses, filters were thawed in 1.4 mL of ice-cold MQ water, and cells were resuspended and divided in half: 0.7 mL of cell suspensions for TOC/TN was diluted in MQ water to a total volume of 20 mL; those for FAME were transferred to vials and stored at −80°C. TOC/TN concentrations were measured on a Shimadzu TOC-L total organic carbon analyzer and TNM-L total nitrogen measuring unit. Lipids of wild-type and NRKO cells were transesterified and their FAME compositions and concentrations determined using gas chromatography with flame ionization detection. In short, FAMEs were prepared and analyzed using methods derived from AOCS Ce 1b-89 and AOCS Ce 1-62, respectively. Dried samples in 4-mL glass vials were resuspended in 0.5 mL of 0.5 M KOH in methanol, 0.5 mL Ottawa sand, and 80 μL of an I.S. mix containing C11:0 FFA/C13:0 TAG/C23:0 FAME. Samples, capped with PTFE-lined caps, were homogenized in a Geno/Grinder at 1200 rpm for 10 min, heated at 80°C for 30 min, the homogenized for an additional 5 min. Then, 0.5 mL of 14% BF3 in methanol was added to the samples, and after heating and homogenizing, 2 mL of n-heptane followed by 0.5 mL of saturated NaCl was added. The samples were homogenized for a final time for 1.5 min, centrifuged at 1000 rpm for 3 min, and the upper phase sampled directly for gas chromatography with flame ionization detection analysis. Pairwise t tests were run on the data for C:N ratios, accumulation of lipids (FAME), and fatty acid ratios to track differences between wild-type and NR-KO14 cells by time point. To determine if there were significant differences in FAME between the two genotypes over the time course, a two-way ANOVA for both data sets was performed in R, and statistics for time, strain, and interaction were recorded.
RNA-Seq and Harvesting NR-KO14 and Wild-Type Cells for a 10-d Time Course
To link RNA-seq transcriptomic results to the physiological and biochemical experiments, a similar 10-d time-course experiment was performed. NR-KO14 transgenic cells and wild-type cultures were grown under the same conditions (as described above) with the exception of the following details. After preconditioning (growth on NH4+ to provide ample biomass for NR-KO14 cells that do not continue to grow when introduced to NO3−) and incubation in −N media, both NR-KO14 and wild-type cells were aliquoted in triplicate 1L Roux bottles (800 mL/bottle). To ensure there would be adequate NR-KO14 biomass at the later time points in the experiment, each NR-KO14 triplicate consisted of two one-liter Roux bottles. Sampling of NR-KO14 cultures at each time point drew from both triplicate bottles equally throughout the time course. Samples were collected in triplicate at eleven time points: (1) mid-exponential phase growth on NH4+, (2) after 2 h of incubation in N-free media, (3) 1 h after transfer of cultures into NO3− media, at (4) 18 h, and subsequently at 24-h intervals thereafter: (5) 42 h, (6) 66 h, (7) 90 h, (8) 114 h, (9) 138 h, (10) 162 h, and (11) 228 h. Depending on cell density, samples collected for RNA were either filtered on 5.0 μM SVPP Durapore filters (Millipore Sigma) or pelleted by centrifugation (20,000g, 10 min) at 4°C, then flash-frozen in liquid nitrogen and stored at −80°C until use. RNA was extracted using Trizol reagent (Thermo Fisher Scientific), and genomic DNA was removed with DNase I (TURBO DNA-free kit; ThermoFisher Scientific), followed by RNA purification with the Agencourt RNAClean XP beads (Beckman Coulter). RNA samples were enriched for mRNA using the NEBNext Poly(A) mRNA magnetic isolation module (New England Biolabs). Libraries were constructed using ScriptSeq v2 RNA-seq kit (Illumina), library quality verified on the Agilent 2200 TapeStation System, and sequencing of 66 libraries was run on the Illumina Single-Read 50 (SR50) platform.
Sequence-Read Mapping
Paired-end Illumina HiSeq reads were quality trimmed to Phred score 33 and at least 30 bp in length. Filtered reads were mapped to contigs of P. tricornutum (http://genome.jgi.doe.gov/Phatr2/Phatr2.download.html) using BWA MEM (Li, 2013). Raw read counts and RPKMs for genes were based on Phatr3 gene models (http://protists.ensembl.org/Phaeodactylum_tricornutum/Info/Index). Previous Phatr2 gene models can be found at http://genome.jgi.doe.gov/cgi-bin/browserLoad?db=Phatr2andposition=chr_1:1-100000.
Transcriptome Analysis in NR-KO14 and Wild-Type Cells in Response to NO3−
RNA-seq reads for each NR-KO14 and wild-type time point were mapped to the P. tricornutum Phatr2 assembly (ASM15095v2), which includes 34 finished chromosomes and 55 unassembled scaffolds (Bowler et al., 2008). To confirm the insertion of the Sh ble cassette into the NR loci, we mapped the cassette to the NR-KO14 assembly (Supplemental Figure 16). Based on reannotation of the P. tricornutum genome (Phatr3) (http://protists.ensembl.org/Phaeodactylum_tricornutum/), augmented by 90 new RNA-seq libraries, expression levels were computed for 12,177 genes. A working set of 10,710 genes was used for all transcriptomic comparisons (Supplemental Data Set 1). Differential expression analysis was performed using edgeR (Robinson et al., 2010) on raw read counts to obtain normalized fold change and Benjamini-Hochberg adjusted P values for each gene. Normal distributions of RPKMs for visualization and expression pattern analysis were derived from the method developed by Tavazoie et al. (1999). Expression profiles of individual genes and gene clusters were developed using hierarchical and K-means clustering and Pavlidis template matching tools within the MultiExperiment Viewer application (Saeed et al., 2003, and references therein). ANOVA was performed in R using the aov function and statistics provided in Supplemental File 1 (Waterhouse et al., 2009). In silico targeting of cognate proteins to the chloroplast or mitochondria was determined by SignalP 4.1, ASAFind, and TargetP 1.1 (Petersen et al., 2011; Gruber et al., 2015; Emanuelsson et al., 2000). Protein sequence alignments were analyzed by ClustalOmega (Sievers et al., 2011) and visualized by JalView (R Core Team, 2016).
For transcriptomic analysis of the impact of −N on gene expression, log2 fold change values (edgeR, FDR = 0.01) for all genes in the transcriptomes were aligned for three time points (NH4+, −N, of NO3−). In the alignment of the 10,711 transcripts, those sharing significant expression for both the NH4+ and −N time points were eliminated (∼2500 IDs). Of the remaining genes, those where FDRs were above P = 0.01 were also eliminated, resulting in ∼5000 gene IDs. Those with fold change values between 0.5 and −0.5 were eliminated. The subsequent ∼2000 IDs were filtered for genes where RPKM expression values in NH4+ were equal or very similar in both transcriptomes, and where those genes had higher RPKMs in NR-KO14 than in the wild type. The remaining ∼1500 genes were filtered for a final time for annotations. The resulting 849 genes were considered in context of the analyses undertaken throughout these experiments.
Accession Numbers
Sequence data reported in this article have been deposited in the NCBI sequence read archive (BioProject accession number PRJNA382762; BioSample accession numbers SAMN06718377–SAMN06718442). Additional duplicate transcriptome data have also been deposited in the NCBI sequence read archive (BioProject accession number PRJNA311568; Biosample accession numbers SAMN04489008–SAMN04489051). Both sets of RNA-seq data were mapped to existing genomic data sets. This study did not generate new genomic data. Accession numbers for genes described in this work include P. tricornutum NR, J54983; KCT1, J41969; NRT2, J26029; NAR1, J13076; NiR-Fd, J12902; NiR-NADH, EG02286; MFS NO3− transporter, EG02608; V-type Cl− channels: EG01952; J28245, J46097, and J52412; Uroporphyrin-III methyltransferase, J12925; ACOAT, J50577; PK, J5286; ICDH, J14762; unCPS, J24195; ABC transporter, J21548; MoaC, J15625; pPDC E1α, EG02309; Acyl-ACP DS, J9316; ∆9 FADS, J28797; PAP, J39949; DGAT, J43469; MLDP, J48859; PAD, J41807; KDSR, J48977; ∆4 SLDS, J22677; C-4 sterol methyl oxidase, J10852; Sterol C5 desaturase, J45494; Patatin-like lipase, EG00718; Phospholipase, J48391; PITPs, J33651, J33873 (see Supplemental Data Set 5 for gene annotations).
Supplemental Data
Supplemental Figure 1. Production of Wild-Type NR Protein Is Reduced in NH4+ and N-Free Conditions.
Supplemental Figure 2. NR and KAT Colocalize to the Peroxisome 18 h after Addition of NO3−.
Supplemental Figure 3. Complementation of NR-KO14.
Supplemental Figure 4. Morphological Changes to NR-KO14 Cells When Grown on NO3−.
Supplemental Figure 5. NO3− Assimilation and Related Downstream Genes in Wild-Type and NR-KO Transcriptomes.
Supplemental Figure 6. NO3− MFS Transporter Gene Expression Profiles.
Supplemental Figure 7. Profile of Fatty Acid Distributions.
Supplemental Figure 8. Response of Lipid Gene Expression to Addition of NO3−.
Supplemental Figure 9. Photosystem II, Electron Transport Chain, Light-Harvesting Complex, and Chlorophyll Biosynthesis Gene Expression Are Inactivated in NR-KO Cells.
Supplemental Figure 10. Carbon Fixation and Metabolism Are Downregulated in NR-KO14 Cells after the Addition of NO3−.
Supplemental Figure 11. TCA Cycle Gene Expression, Similar to NR Cohort, Is Induced by Addition of NO3− in NR-KO14 Cells.
Supplemental Figure 12. Addition of NO3− Inhibits Protein the Translation in NR-KO14 Cells.
Supplemental Figure 13. Lack of NR Protein Affects the Transition from NH4+ to −N in ∼5% of the Transcriptome.
Supplemental Figure 14. Live Stain for Lipids in Wild-Type and NR-KO Cells.
Supplemental Figure 15. NR-KO Vectors for TALEN-Mediated Homologous Recombination.
Supplemental Figure 16. Reads Mapped to the ΔNR-Sh ble Fusion Gene.
Supplemental Table 1. Concentration of NO3− in Spent Media.
Supplemental Table 2. Total RPKMs, Significant Fold Changes for NR Cohort, NO3− Transport, and N-Assimilation Genes in the Wild Type versus NR-KO Cells.
Supplemental Data Set 1. Wild-Type and NR-KO14 Transcriptome Data for 10,710 Genes: Annotations for Functional, Phylogenetic, and Metabolic Assignments
Supplemental Data Set 2. Differential Expression Analysis (edgeR log2fc) for Carbon Fixation and Metabolism Cluster.
Supplemental Data Set 3. Differential Expression Analysis (edgeR log2fc) of Pairwise Comparisons of NR-KO14 versus the Wild Type per Time Point.
Supplemental Data Set 4. Two Clusters of Genes That Respond Oppositely to the Transition from NH+4 to N-Free.
Supplemental Data Set 5. Phaeodactylum tricornutum 3 (Pt_3) Gene Identification Numbers and Annotations for Genes Described in the Manuscript.
Supplemental File 1. ANOVA Tables for Figures 5A, 5B, and 8B.
Acknowledgments
We thank R. Diner and B. Karas for support with conjugation. This work was supported by the National Science Foundation (MCB-1024913 to A.E.A. and C.L.D.), by the U.S. Department of Energy Genomics Science program (DE-SC00006719 and DE-SC0008593 to A.E.A. and C.L.D.), and by Gordon and Betty Moore Foundation Grant GBMF3828 (A.E.A.).
AUTHOR CONTRIBUTIONS
J.K.M., C.P.B., C.L.D., and A.E.A. designed research. J.K.M., H.Z., M.T., K.B., R.R., and A.E.A. performed research. J.K.M., U.G., C.L.D., C.P.B., and A.E.A. contributed new reagents/analytic tools. J.K.M., S.R.S., J.P.M., C.L.D., and A.E.A. analyzed data. J.K.M., S.R.S., U.W.G., J.P.M., C.L.D, and A.E.A. wrote the article.
Glossary
- TAG
triacylglycerol
- NR
nitrate reductase
- FTIR
Fourier transform infrared spectroscopy
- QFDE
quick-freeze deep-etch
- TOC/TN
total organic carbon/total nitrogen
- FAME
fatty acid methyl ester
- FA
fatty acid
- ER
endoplasmic reticulum
- MLDP
major lipid droplet protein
- PET
photosynthetic electron transport
- CB
Calvin-Benson
- SW
seawater
- ASW
artificial seawater
Footnotes
Articles can be viewed without a subscription.
References
- Abida H., et al. (2015). Membrane glycerolipid remodeling triggered by nitrogen and phosphorus starvation in Phaeodactylum tricornutum. Plant Physiol. 167: 118–136. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Alipanah L., Rohloff J., Winge P., Bones A.M., Brembu T. (2015). Whole-cell response to nitrogen deprivation in the diatom Phaeodactylum tricornutum. J. Exp. Bot. 66: 6281–6296. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Allen A.E., Dupont C.L., Obornik M., Horak A., Nunes-Nesi A., McCrow J.P., Zheng H., Johnson D.A., Hu H., Fernie A.R., Bowler C. (2011). Evolution and metabolic significance of the urea cycle in photosynthetic diatoms. Nature 473: 203–207. [DOI] [PubMed] [Google Scholar]
- Allen A.E., Ward B.B., Song B. (2005). Characterization of diatom (Bacillariophyceae) nitrate reductase genes and their detection in marine phytoplankton communities. J. Phycol. 41: 95–104. [Google Scholar]
- Amy N.K., Garrett R.H. (1974). Purification and characterization of the nitrate reductase from the diatom Thalassiosira pseudonana. Plant Physiol. 54: 629–637. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Apt K.E., Zaslavkaia L., Lippmeier J.C., Lang M., Kilian O., Wetherbee R., Grossman A.R., Kroth P.G. (2002). In vivo characterization of diatom multipartite plastid targeting signals. J. Cell Sci. 115: 4061–4069. [DOI] [PubMed] [Google Scholar]
- Bankaitis V.A., Mousley C.J., Schaaf G. (2010). The Sec14 superfamily and mechanisms for crosstalk between lipid metabolism and lipid signaling. Trends Biochem. Sci. 35: 150–160. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Bender S.J., Durkin C.A., Berthiaume C.T., Morales R.L., Armbrust E.V. (2014). Transcriptional responses of three model diatoms to nitrate limitation of growth. Front. Mater. Sci. 1: 1–15. [Google Scholar]
- Bergsdorf E.Y., Zdebik A.A., Jentsch T.J. (2009). Residues important for nitrate/proton coupling in plant and mammalian CLC transporters. J. Biol. Chem. 284: 11184–11193. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Bowler C., et al. (2008). The Phaeodactylum genome reveals the evolutionary history of diatom genomes. Nature 456: 239–244. [DOI] [PubMed] [Google Scholar]
- Campbell W.H. (1999). Nitrate reductase structure, function, and regulation: bridging the gap between biochemistry and physiology. Annu. Rev. Plant Physiol. Plant Mol. Biol. 50: 277–303. [DOI] [PubMed] [Google Scholar]
- Cermeño P., Lee J.B., Wyman K., Schofield O., Falkowski P.G. (2011). Competitive dynamics in two species of marine phytoplankton under non-equilibrium conditions. Mar. Ecol. Prog. Ser. 429: 19–28. [Google Scholar]
- Chung J.H., Lester R.L., Dickson R.C. (2003). Sphingolipid requirement for generation of a functional v1 component of the vacuolar ATPase. J. Biol. Chem. 278: 28872–28881. [DOI] [PubMed] [Google Scholar]
- Clark D.R., Miller P.I., Woodward M.S., Rees A.P. (2011). Inorganic nitrogen assimilation and regeneration in the coastal upwelling region of the Iberian peninsula. Limnol. Oceanogr. 56: 1689–1702. [Google Scholar]
- Cockcroft S., Garner K. (2013). Potential role for phosphatidylinositol transfer protein (PITP) family in lipid transfer during phospholipase C signalling. Adv. Biol. Regul. 53: 280–291. [DOI] [PubMed] [Google Scholar]
- Collos Y., Mornet F., Sciandra A., Waser N., Larson A., Harrison P.J. (1999). An optical method for the rapid measurement of micromolar concentrations of nitrate in marine phytoplankton cultures. J. Appl. Phycol. 11: 179–184. [Google Scholar]
- Crawford N.M., Guo F.Q. (2005). New insights into nitric oxide metabolism and regulatory functions. Trends Plant Sci. 10: 195–200. [DOI] [PubMed] [Google Scholar]
- Cresswell R.C., Syrett P.J. (1979). Ammonium inhibition of nitrate uptake by the diatom, Phaeodactylum tricornutum. Plant Sci. Lett. 14: 321–325. [Google Scholar]
- De Angeli A., Monachello D., Ephritikhine G., Frachisse J.M., Thomine S., Gambale F., Barbier-Brygoo H. (2006). The nitrate/proton antiporter AtCLCa mediates nitrate accumulation in plant vacuoles. Nature 442: 939–942. [DOI] [PubMed] [Google Scholar]
- Del Río L.A. (2011). Peroxisomes as a cellular source of reactive nitrogen species signal molecules. Arch. Biochem. Biophys. 506: 1–11. [DOI] [PubMed] [Google Scholar]
- Dolch L.J., Maréchal E. (2015). Inventory of fatty acid desaturases in the pennate diatom Phaeodactylum tricornutum. Mar. Drugs 13: 1317–1339. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Domergue F., Spiekermann P., Lerchl J., Beckmann C., Kilian O., Kroth P.G., Boland W., Zahringer U., Heinz E. (2003). New insight into Phaeodactylum tricornutum fatty acid metabolism. Cloning and functional characterization of plastidial and microsomal delta12-fatty acid desaturases. Plant Physiol. 131: 1648–1660. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Dortch Q. (1982). Effect of growth conditions on accumulation of internal nitrate, ammonium, amino acids and protein in three marine diatoms. J. Exp. Mar. Biol. Ecol. 61: 243–264. [Google Scholar]
- Dortch Q., Clayton J.R., Thoresen S.S., Ahmed S.I. (1984). Species differences in accumulation of nitrogen pools in phytoplankton. Mar. Biol. 81: 237–250. [Google Scholar]
- D’Souza L., Devi P., Shridhar D.M., Naik C.G. (2008). Use of Fourier transform infrared (FTIR) spectroscopy to study cadmium-induced changes in Padina tetrastromatica (Hauck). Anal. Chem. Insights 3: 135–143. [PMC free article] [PubMed] [Google Scholar]
- Dyhrman S.T., Palenik B. (2001). A single-cell immunoassay for phosphate stress in the dinoflagellate Prorocentrum minimum (Dinophycaea). J. Phycol. 37: 400–410. [Google Scholar]
- Emanuelsson O., Nielsen H., Brunak S., von Heijne G. (2000). Predicting subcellular localization of proteins based on their N-terminal amino acid sequence. J. Mol. Biol. 300: 1005–1016. [DOI] [PubMed] [Google Scholar]
- Falciatore A., Casotti R., Leblanc C., Abrescia C., Bowler C. (1999). Transformation of nonselectable reporter genes in marine diatoms. Mar. Biotechnol. (NY) 1: 239–251. [DOI] [PubMed] [Google Scholar]
- Falkowski P.G., Katz M.E., Knoll A.H., Quigg A., Raven J.A., Schofield O., Taylor F.J. (2004). The evolution of modern eukaryotic phytoplankton. Science 305: 354–360. [DOI] [PubMed] [Google Scholar]
- Falkowski P.G., Stone D.P. (1975). Nitrate uptake in marine phytoplankton: Energy sources and the interaction with carbon fixation. Mar. Biol. 32: 77–84. [Google Scholar]
- Fernandez-Murga M.L., Gil-Ortiz F., Llacer J.L., Rubio V. (2004). Arginine biosynthesis in Thermotoga maritima: characterization of the arginine-sensitive N-acetyl-L-glutamate kinase. J. Bacteriol. 186: 6142–6149. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Finnigan G.C., Ryan M., Stevens T.H. (2011). A genome-wide enhancer screen implicates sphingolipid composition in vacuolar ATPase function in Saccharomyces cerevisiae. Genetics 187: 771–783. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Flynn K.J. (1991). Algal carbon–nitrogen metabolism: a biochemical basis for modelling the interactions between nitrate and ammonium uptake. J. Plankton Res. 13: 373–387. [Google Scholar]
- Forde B.G. (2000). Nitrate transporters in plants: structure, function and regulation. Biochim. Biophys. Acta 1465: 219–235. [DOI] [PubMed] [Google Scholar]
- Gao Y., Smith G.J., Alberte R.S. (1993). Nitrate reductase from the Marine diatom Skeletonema costatum (biochemical and immunological characterization). Plant Physiol. 103: 1437–1445. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Ge F., Huang W., Chen Z., Zhang C., Xiong Q., Bowler C., Yang J., Xu J., Hu H. (2014). Methylcrotonyl-CoA carboxylase regulates triacylglycerol accumulation in the model diatom Phaeodactylum tricornutum. Plant Cell 26: 1681–1697. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Gibson D.G., Young L., Chuang R.Y., Venter J.C., Hutchison C.A. III, Smith H.O. (2009). Enzymatic assembly of DNA molecules up to several hundred kilobases. Nat. Methods 6: 343–345. [DOI] [PubMed] [Google Scholar]
- Glibert P.M., Wilkerson , F.P., Dugdale R.C., Raven J.A., Dupont C.L., Leavitt P.R., Parker A.E., Burkholder J.M., Kana T.M. (2016). Pluses and minuses of ammonium and nitrate uptake and assimilation by phytoplankton and implications for productivity and community composition, with emphasis on nitrogen-enriched conditions. Limnol. Oceanogr. 61: 165–197. [Google Scholar]
- Glass A.D., Britto D.T., Kaiser B.N., Kinghorn J.R., Kronzucker H.J., Kumar A., Okamoto M., Rawat S., Siddiqi M.Y., Unkles S.E., Vidmar J.J. (2002). The regulation of nitrate and ammonium transport systems in plants. J. Exp. Bot. 53: 855–864. [DOI] [PubMed] [Google Scholar]
- Gonzalez N.H., Felsner G., Schramm F.D., Klingl A., Maier U.G., Bolte K. (2011). A single peroxisomal targeting signal mediates matrix protein import in diatoms. PLoS One 6: e25316. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Granstedt R.C., Huffaker R.C. (1982). Identification of the leaf vacuole as a major nitrate storage pool. Plant Physiol. 70: 410–413. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Gruber A., Rocap G., Kroth P.G., Armbrust E.V., Mock T. (2015). Plastid proteome prediction for diatoms and other algae with secondary plastids of the red lineage. Plant J. 81: 519–528. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Heuser J.E. (2011). The origins and evolution of freeze-etch electron microscopy. J. Electron Microsc. 60: S3–S29. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Hildebrand M., Davis A.K., Smith S.R., Traller J.C., Abbriano R. (2012). The place of diatoms in the biofuels industry. Biofuels 3: 221–240. [Google Scholar]
- Hockin N.L., Mock T., Mulholland F., Kopriva S., Malin G. (2012). The response of diatom central carbon metabolism to nitrogen starvation is different from that of green algae and higher plants. Plant Physiol. 158: 299–312. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Juergens M.T., et al. (2015). The regulation of photosynthetic structure and function during nitrogen deprivation in Chlamydomonas reinhardtii. Plant Physiol. 167: 558–573. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Karas B.J., et al. (2015). Designer diatom episomes delivered by bacterial conjugation. Nat. Commun. 6: 6925. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Karas B.J., et al. (2013). Assembly of eukaryotic algal chromosomes in yeast. J. Biol. Eng. 7: 30. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Levering J., Broddrick J., Dupont C.L., Peers G., Beeri K., Mayers J., Gallina A.A., Allen A.E., Palsson B.O., Zengler K. (2016). Genome-scale model reveals metabolic basis of biomass partitioning in a model diatom. PLoS One 11: e0155038. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Levitan O., Dinamarca J., Zelzion E., Lun D.S., Guerra L.T., Kim M.K., Kim J., Van Mooy B.A., Bhattacharya D., Falkowski P.G. (2015). Remodeling of intermediate metabolism in the diatom Phaeodactylum tricornutum under nitrogen stress. Proc. Natl. Acad. Sci. USA 112: 412–417. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Li H. (2013). Aligning sequence reads, clone sequences and assembly contigs with BWA-MEM. arXiv doi/10.1101/13033997.
- Li J., Han D., Wang D., Ning K., Jia J., Wei L., Jing X., Huang S., Chen J., Li Y., Hu Q., Xu J. (2014). Choreography of transcriptomes and lipidomes of Nannochloropsis reveals the mechanisms of oil synthesis in microalgae. Plant Cell 26: 1645–1665. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Li X., Moellering E.R., Liu B., Johnny C., Fedewa M., Sears B.B., Kuo M.-H., Benning C. (2012). A galactoglycerolipid lipase is required for triacylglycerol accumulation and survival following nitrogen deprivation in Chlamydomonas reinhardtii. Plant Cell 24: 4670–4686. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Lichtle C., McKay R.M.L., Gibbs S.P. (1992). Immunogold localization of photosystem I and photosystem II light-harvesting complexes in cryptomonad thylakoids. Biol. Cell 74: 187–194. [Google Scholar]
- Liu J., Sun Z., Zhong Y., Huang J., Hu Q., Chen F. (2012). Stearoyl-acyl carrier protein desaturase gene from the oleaginous microalga Chlorella zofingiensis: cloning, characterization and transcriptional analysis. Planta 236: 1665–1676. [DOI] [PubMed] [Google Scholar]
- Lomas M.W., Glibert P.M. (1999). Temperature regulation of nitrate uptake: A novel hypothesis about nitrate uptake and reduction in cool-water diatoms. Limnol. Oceanogr. 44: 556–572. [Google Scholar]
- Lomas M.W., Glibert P.M. (2000). Comparisons of nitrate uptake, storage, and reduction in marine diatoms and flagellates. J. Phycol. 36: 903–913. [Google Scholar]
- Lu G., Campbell W., Lindqvist Y., Schneider G. (1992). Crystallization and preliminary crystallographic studies of the FAD domain of corn NADH: nitrate reductase. J. Mol. Biol. 224: 277–279. [DOI] [PubMed] [Google Scholar]
- Malviya S., Scalco E., Audic S., Vincent F., Veluchamy A., Poulain J., Wincker P., Iudicone D., de Vargas C., Bittner L., Zingone A., Bowler C. (2016). Insights into global diatom distribution and diversity in the world’s ocean. Proc. Natl. Acad. Sci. USA 113: E1516–E1525. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Martin G.J.O., Hill D.R.A., Olmstead I.L.D., Bergamin A., Shears M.J., Dias D.A., Kentish S.E., Scales P.J., Botté C.Y., Callahan D.L. (2014). Lipid profile remodeling in response to nitrogen deprivation in the microalgae Chlorella sp. (Trebouxiophyceae) and Nannochloropsis sp. (Eustigmatophyceae). PLoS One 9: e103389. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Martinoia E., Heck U., Wiemken A. (1981). Vacuoles as storage compartments for nitrate in barley leaves. Nature 289: 292–294. [Google Scholar]
- Martinoia E., Maeshima M., Neuhaus H.E. (2007). Vacuolar transporters and their essential role in plant metabolism. J. Exp. Bot. 58: 83–102. [DOI] [PubMed] [Google Scholar]
- Mayers J.J., Flynn K.J., Shields R.J. (2013). Rapid determination of bulk microalgal biochemical composition by Fourier-transform infrared spectroscopy. Bioresour. Technol. 148: 215–220. [DOI] [PubMed] [Google Scholar]
- McDonald S.M., Plant J.N., Worden A.Z. (2010). The mixed lineage nature of nitrogen transport and assimilation in marine eukaryotic phytoplankton: a case study of micromonas. Mol. Biol. Evol. 27: 2268–2283. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Michaelson L.V., Markham J.E., Zäeuner S., Matsumoto M., Chen M., Cahoon E.B., Napier J.A. (2013). Identification of a cytochrome b5-fusion desaturase responsible for the synthesis of triunsaturated sphingolipid long chain bases in the marine diatom Thalassiosira pseudonana. Phytochemistry 90: 50–55. [DOI] [PubMed] [Google Scholar]
- Michaelson L.V., Napier J.A., Molino D., Faure J.-D. (2016). Plant sphingolipids: Their importance in cellular organization and adaption. Biochim. Biophys. Acta 1861: 1329–1335. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Mus F., Toussaint J.P., Cooksey K.E., Fields M.W., Gerlach R., Peyton B.M., Carlson R.P. (2013). Physiological and molecular analysis of carbon source supplementation and pH stress-induced lipid accumulation in the marine diatom Phaeodactylum tricornutum. Appl Microbiol. Biotechnol. 97: 3625–3642. [DOI] [PubMed] [Google Scholar]
- Neuberger G., Maurer-Stroh S., Eisenhaber B., Hartig A., Eisenhaber F. (2003). Motif refinement of the Peroxisomal Targeting Signal 1 and evaluation of taxon-specific differences. J. Mol. Biol. 328: 567–579. [DOI] [PubMed] [Google Scholar]
- Nguitragool W., Miller C. (2006). Uncoupling of a CLC Cl-/H+ exchange transporter by polyatomic anions. J. Mol. Biol. 362: 682–690. [DOI] [PubMed] [Google Scholar]
- Nunes-Nesi A., Fernie A.R., Stitt M. (2010). Metabolic and signaling aspects underpinning the regulation of plant carbon nitrogen interactions. Mol. Plant 3: 973–996. [DOI] [PubMed] [Google Scholar]
- Oborník M., Green B.R. (2005). Mosaic origin of the heme biosynthesis pathway in photosynthetic eukaryotes. Mol. Biol. Evol. 22: 2343–2353. [DOI] [PubMed] [Google Scholar]
- Ozolina N.V., Nesterkina I.S., Kolesnikova E.V., Salyaev R.K., Nurminsky V.N., Rakevich A.L., Martynovich E.F., Chernyshov M.Y. (2013). Tonoplast of Beta vulgaris L. contains detergent-resistant membrane microdomains. Planta 237: 859–871. [DOI] [PubMed] [Google Scholar]
- Petersen T.N., Brunak S., von Heijne G., Nielsen H. (2011). SignalP 4.0: discriminating signal peptides from transmembrane regions. Nat. Methods 8: 785–786. [DOI] [PubMed] [Google Scholar]
- Poulsen N., Kröger N. (2005). A new molecular tool for transgenic diatoms: control of mRNA and protein biosynthesis by an inducible promoter-terminator cassette. FEBS J. 272: 3413–3423. [DOI] [PubMed] [Google Scholar]
- Prihoda J., Tanaka A., de Paula W.B., Allen J.F., Tirichine L., Bowler C. (2012). Chloroplast-mitochondria cross-talk in diatoms. J. Exp. Bot. 63: 1543–1557. [DOI] [PubMed] [Google Scholar]
- Radakovits R., Jinkerson R.E., Darzins A., Posewitz M.C. (2010). Genetic engineering of algae for enhanced biofuel production. Eukaryot. Cell 9: 486–501. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Raven J.A. (1987). The role of vacuoles. New Phytol. 106: 357–422. [Google Scholar]
- R Core Team (2016). R: A Language and Environment for Statistical Computing. (Vienna, Austria: R Foundation for Statistical Computing).
- Rismani-Yazdi H., Haznedaroglu B.Z., Hsin C., Peccia J. (2012). Transcriptomic analysis of the oleaginous microalga Neochloris oleoabundans reveals metabolic insights into triacylglyceride accumulation. Biotechnol. Biofuels 5: 74. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Robinson M.D., McCarthy D.J., Smyth G.K. (2010). edgeR: a Bioconductor package for differential expression analysis of digital gene expression data. Bioinformatics 26: 139–140. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Saeed A.I., et al. (2003). TM4: a free, open-source system for microarray data management and analysis. Biotechniques 34: 374–378. [DOI] [PubMed] [Google Scholar]
- Scheible W.R., Gonzalez-Fontes A., Lauerer M., Muller-Rober B., Caboche M., Stitt M. (1997). Nitrate acts as a signal to induce organic acid metabolism and repress starch metabolism in tobacco. Plant Cell 9: 783–798. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Scheible W.R., Morcuende R., Czechowski T., Fritz C., Osuna D., Palacios-Rojas N., Schindelasch D., Thimm O., Udvardi M.K., Stitt M. (2004). Genome-wide reprogramming of primary and secondary metabolism, protein synthesis, cellular growth processes, and the regulatory infrastructure of Arabidopsis in response to nitrogen. Plant Physiol. 136: 2483–2499. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Scherer G.F., Ryu S.B., Wang X., Matos A.R., Heitz T. (2010). Patatin-related phospholipase A: nomenclature, subfamilies and functions in plants. Trends Plant Sci. 15: 693–700. [DOI] [PubMed] [Google Scholar]
- Schmollinger S., et al. (2014). Nitrogen-sparing mechanisms in Chlamydomonas affect the transcriptome, the proteome, and photosynthetic metabolism. Plant Cell 26: 1410–1435. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Shen P.-L., Wang H.-T., Pan Y.-F., Meng Y.-Y., Wu P.-C., Xue S. (2016). Identification of characteristic fatty acids to quantify triacylglycerols in microalgae. Front. Plant Sci. 7: 162. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Shifrin N.S., Chisholm S.W. (1981). Phytoplankton lipids: interspecific differences and effects of nitrate, silicate and light-dark cycles. J. Phycol. 17: 374–384. [Google Scholar]
- Shtaida N., Khozin-Goldberg I., Solovchenko A., Chekanov K., Didi-Cohen S., Leu S., Cohen Z., Boussiba S. (2014). Downregulation of a putative plastid PDC E1α subunit impairs photosynthetic activity and triacylglycerol accumulation in nitrogen-starved photoautotrophic Chlamydomonas reinhardtii. J. Exp. Bot. 65: 6563–6576. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Sievers F., Wilm A., Dineen D., Gibson T.J., Karplus K., Li W., Lopez R., McWilliam H., Remmert M., Söding J., Thompson J.D., Higgins D.G. (2011). Fast, scalable generation of high-quality protein multiple sequence alignments using Clustal Omega. Mol. Syst. Biol. 7: 539. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Simionato D., Block M.A., La Rocca N., Jouhet J., Maréchal E., Finazzi G., Morosinotto T. (2013). The response of Nannochloropsis gaditana to nitrogen starvation includes de novo biosynthesis of triacylglycerols, a decrease of chloroplast galactolipids, and reorganization of the photosynthetic apparatus. Eukaryot. Cell 12: 665–676. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Sims P.A., Mann D.G., Medlin L.K. (2006). Evolution of the diatoms: insights from fossil, biological and molecular data. Phycologia 45: 361–402. [Google Scholar]
- Smith S.R., Abbriano R.M., Hildebrand M. (2012). Comparative analysis of diatom genomes reveals substantial differences in the organization of carbon partitioning pathways. Algal Research 1: 2–16. [Google Scholar]
- Stehfest K., Toepel J., Wilhelm C. (2005). The application of micro-FTIR spectroscopy to analyze nutrient stress-related changes in biomass composition of phytoplankton algae. Plant Physiol. Biochem. 43: 717–726. [DOI] [PubMed]
- Syrett P.J., Flynn K.J., Molloy C.J., Dixon G.K., Peplinska A.M., Cresswell R.C. (1986). Effects of nitrogen deprivation on rates of uptake of nitrogenous compounds by the diatom, Phaeodactylum tricornutum Bohlin. New Phytol. 102: 39–44. [DOI] [PubMed] [Google Scholar]
- Tavazoie S., Hughes J.D., Campbell M.J., Cho R.J., Church G.M. (1999). Systematic determination of genetic network architecture. Nat. Genet. 22: 281–285. [DOI] [PubMed] [Google Scholar]
- Tesson B., Hildebrand M. (2010). Extensive and intimate association of the cytoskeleton with forming silica in diatoms: control over patterning on the meso- and micro-scale. PLoS One 5: e14300. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Tonon T., Sayanova O., Michaelson L.V., Qing R., Harvey D., Larson T.R., Li Y., Napier J.A., Graham I.A. (2005). Fatty acid desaturases from the microalga Thalassiosira pseudonana. FEBS J. 272: 3401–3412. [DOI] [PubMed] [Google Scholar]
- Trentacoste E.M., Shrestha R.P., Smith S.R., Glé C., Hartmann A.C., Hildebrand M., Gerwick W.H. (2013). Metabolic engineering of lipid catabolism increases microalgal lipid accumulation without compromising growth. Proc. Natl. Acad. Sci. USA 110: 19748–19753. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Turpin D.H. (1991). Effects of inorganic N availability on algal photosynthesis and carbon metabolism. J. Phycol. 27: 14–20. [Google Scholar]
- Vidmar J.J., Zhuo D., Siddiqi M.Y., Schjoerring J.K., Touraine B., Glass A.D.M. (2000). Regulation of high-affinity nitrate transporter genes and high-affinity nitrate influx by nitrogen pools in roots of barley. Plant Physiol. 123: 307–318. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Villareal T.A., Altabet M.A., Culver-Rymsza K. (1993). Nitrogen transport by vertically migrating diatom mats in the North Pacific Ocean. Nature 363: 709–712. [Google Scholar]
- von der Fecht-Bartenbach J., Bogner M., Dynowski M., Ludewig U. (2010). CLC-b-mediated NO3−/H+ exchange across the tonoplast of Arabidopsis vacuoles. Plant Cell Physiol. 51: 960–968. [DOI] [PubMed] [Google Scholar]
- Von Wettstein D., Gough S., Kannangara C.G. (1995). Chlorophyll biosynthesis. Plant Cell 7: 1039–1057. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Wang R., Guegler K., LaBrie S.T., Crawford N.M. (2000). Genomic analysis of a nutrient response in Arabidopsis reveals diverse expression patterns and novel metabolic and potential regulatory genes induced by nitrate. Plant Cell 12: 1491–1509. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Wang R., Tischner R., Gutiérrez R.A., Hoffman M., Xing X., Chen M., Coruzzi G., Crawford N.M. (2004). Genomic analysis of the nitrate response using a nitrate reductase-null mutant of Arabidopsis. Plant Physiol. 136: 2512–2522. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Waterhouse A.M., Procter J.B., Martin D.M.A., Clamp M., Barton G.J. (2009). Jalview Version 2--a multiple sequence alignment editor and analysis workbench. Bioinformatics 25: 1189–1191. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Weyman P.D., Beeri K., Lefebvre S.C., Rivera J., McCarthy J.K., Heuberger A.L., Peers G., Allen A.E., Dupont C.L. (2015). Inactivation of Phaeodactylum tricornutum urease gene using transcription activator-like effector nuclease-based targeted mutagenesis. Plant Biotechnol. J. 13: 460–470. [DOI] [PubMed] [Google Scholar]
- Yang Z.K., Niu Y.F., Ma Y.H., Xue J., Zhang M.H., Yang W.D., Liu J.S., Lu S.H., Guan Y., Li H.Y. (2013). Molecular and cellular mechanisms of neutral lipid accumulation in diatom following nitrogen deprivation. Biotechnol. Biofuels 6: 67. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Yoneda K., Yoshida M., Suzuki I., Watanabe M.M. (2016). Identification of a major lipid droplet protein in a marine diatom Phaeodactylum tricornutum. Plant Cell Physiol. 57: 397–406. [DOI] [PubMed] [Google Scholar]
- Yoon K., Han D., Li Y., Sommerfeld M., Hu Q. (2012). Phospholipid:diacylglycerol acyltransferase is a multifunctional enzyme involved in membrane lipid turnover and degradation while synthesizing triacylglycerol in the unicellular green microalga Chlamydomonas reinhardtii. Plant Cell 24: 3708–3724. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Yoshida K., et al. (2013). Studies on vacuolar membrane microdomains isolated from Arabidopsis suspension-cultured cells: local distribution of vacuolar membrane proteins. Plant Cell Physiol. 54: 1571–1584. [DOI] [PubMed] [Google Scholar]
- Yoshida S., Uemura M. (1986). Lipid cmposition of plasma membranes and tonoplasts isolated from etiolated seedlings of mung bean (Vigna radiata L.). Plant Physiol. 82: 807–812. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Yu E.T., Zendejas F.J., Lane P.D., Gaucher S., Simmons B.A., Lane T.W. (2009). Triacylglycerol accumulation and profiling in the model diatoms Thalassiosira pseudonana and Phaeodactylum tricornutum (Baccilariophyceae) during starvation. J. Appl. Phycol. 21: 669–681. [Google Scholar]
- Yu W.-L., Ansari W., Schoepp N.G., Hannon M.J., Mayfield S.P., Burkart M.D. (2011). Modifications of the metabolic pathways of lipid and triacylglycerol production in microalgae. Microb. Cell Fact. 10: 91. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Zaslavskaia L.A., Lippmeier J.C., Kroth P.G., Grossman A.R., Apt K.E. (2000). Transformation of the diatom Phaeodactylum tricornutum (Bacillariophyceae) with a variety of selectable marker and reporter genes. J. Phycol. 36: 379–386. [Google Scholar]
- Zhang C., Hicks G.R., Raikhel N.V. (2015). Molecular composition of plant vacuoles: important but less understood regulations and roles of tonoplast lipids. Plants (Basel) 4: 320–333. [DOI] [PMC free article] [PubMed] [Google Scholar]