Strain, growth condition, and treatment

An Anabaena strain that lost its nitrogen fixation ability (Anabaena^ΔN strain) was generated by culturing cells through ten successive sub-culturing steps in nitrogen-rich medium and subsequently testing growth under nitrogen-deficient conditions. Loss of diazotrophic capacity was confirmed by growth arrest under nitrogen-depleted conditions.

For staining and visualization of cyanoglobules, starter cultures were incubated at 32 °C with continuous illumination (150 μmol photons m^–2 s^–1) using Sylvania 15W Grolux cool white bulbs, under shaking at 150 rpm with 2% CO₂ supplementation, inside a Multitron Incubator (Infors HT, Bottmingen, Switzerland). Cultures were grown for two days and harvested at either logarithmic (3–5 µg chl a mL^–1) or stationary phase (8–50 µg chl a mL^–1).

For N-deficiency treatment, triplicate starter cultures at logarithmic phase (OD₇₅₀ 1.2) were centrifuged at 4700 rpm for 3 minutes and the resulting cell pellets were resuspended in nitrogen-limited BG11 medium (0.55 mM N; BG11^–N). For control conditions, resuspension was made in standard BG-11 medium containing 17.6 mM NaNO₃. Modified BG11^–N was prepared by reducing NaNO₃ and replacing the equivalent sodium using NaCl to maintain ionic balance (final Na⁺ concentration = 17.05 mM). Cultures were maintained until they reached stationary phase (ca. 7 days) before sample collection.

Genome isolation, sequencing, and analysis

Genomic DNA from wild-type (Anabaena^WT) and mutant (Anabaena^ΔN) strains was sequenced using Oxford Nanopore long-read sequencing (Plasmidsaurus Inc., South San Francisco, USA). Raw reads were filtered with Filtlong v0.3.1 (Wick, 2018), retaining only reads ≥1,000 bp in length and the top 95% of reads by quality to remove short or low-quality reads that may bias downstream analyses. The filtered reads were aligned to the Anabaena sp. PCC 7120 reference genome [12] using Minimap2 v2.26-r1175 [13] with the “-ax map-ont” preset, which is optimized for long-read sequencing. Alignment files in SAM format were converted to BAM, sorted and indexed using SAMtools v1.19.2 [14]. Variants were then called with Clair3 [15], a deep-learning–based variant caller that has been benchmarked to outperform traditional pipelines on nanopore data [15], using default parameters with the "--include_all_ctgs" flag to enable variant detection across all contigs and the "--no_phasing_for_fa" flag to correctly model the haploid genome of Anabaena. Variant call files (VCFs) were filtered to retain only high-confidence variants with QUAL ≥10, followed by normalization with BCFtools v1.19 [14] to standardize multi-allelic and indel representations across samples. Variants unique to the mutant strains were extracted by removing those shared with wild-type, ensuring the analysis focused specifically on mutant-associated mutations. Functional annotation was performed with SnpEff v5.2c [16] to predict coding and protein-level impacts, and SnpSift v5.2c [16] was used to filter and manipulate annotations for prioritizing variants of interest. All commands and scripts used in this analysis are available on GitHub at https://github.com/mohitmahey/Anabaena_variant_calling_ONT_reads.

Stress treatments

Starter cultures of Anabaena^ΔN were grown in 125-mL flasks of BG11 media at 25 °C with shaking at 150 rpm under 13–15 µmol photons s⁻¹ m⁻² from cool white, fluorescent lights for 2 days. For nutrient stresses, starter culture was harvested, pelleted and aliquots resuspended in BG11^-N (0.55mM; for N stress), low P media BG11^-P (0.0072mM; for P stress), or in standard BG11 media (for control) and grown until stationary phase (8–50 µg Chl a mL⁻¹), at which point samples were collected for CLSM and TEM. For dark stress, starter culture was harvested, pelleted and aliquots resuspended in BG11 medium and wrapped completely in aluminum foil and grown for an additional 7 days, at which point samples were collected for CLSM and TEM. Independent triplicate cultures were sampled and imaged by CLSM and TEM.

Microscopy of cell morphology

Anabaena^WT and Anabaena^ΔN strains were cultured under two conditions: nitrogen-free BG-11^-N medium and nitrogen-replete BG-11 medium. For microscopy, 1 mL of culture was harvested by centrifugation at 3,000 × g for 5 min, and the pellet was gently resuspended. A 5 μL aliquot of the cell suspension was placed onto a clean glass microscope slide. To immobilize the filaments and maintain hydration during imaging, a thin agarose pad (1.5% [w/v] agarose in BG-11 medium, cut into ~0.5 cm squares) was placed carefully on top of the cell suspension.

Samples were imaged using a Thunder Imager System (Leica Microsystems, USA) equipped with a fluorescence light source and a high-sensitivity sCMOS camera. Images were acquired with a 100 × oil-immersion objective using phase contrast to visualize vegetative cells and heterocysts. Acquisition parameters were kept constant across Anabaena^WT and Anabaena^ΔN strains to allow direct comparison. Images were processed with the Leica Thunder image software to reduce background and improve contrast.

Confocal laser scanning microscopy (CLSM) of cyanoglobule lipid droplets

To visualize intracellular cyanoglobules lipid droplets, cyanobacterial pellets from 2 mL of culture were stained with the lipophilic dye monodansyl pentane (MDH). MDH was diluted to a final concentration of 0.1 mM from a 1 mM DMSO stock, and cells were incubated in this solution for 1–2 minutes. Stained cells were immobilized on 1% agarose gel pads and mounted on glass slides for imaging. Fluorescence signals were acquired using an Olympus FLUOVIEW FV1000 confocal microscope with a 100x UPlanApo oil objective (NA 1.40). MDH was excited at 444 nm and emissions were captured at 485–535 nm. Chlorophyll autofluorescence was visualized by excitation at 647 nm and collection from 662–737 nm. Time-series images were collected every 10 seconds over a 30-minute interval.

Transmission electron microscopy (TEM)

For ultrastructural analysis, 1 mL of culture was pelleted at 14,000 × g for 1 minute and embedded in 8% agarose. Embedded samples were fixed with 2% glutaraldehyde and 2% paraformaldehyde in 0.05 M cacodylate buffer (pH 7.3), followed by three 10-minute washes. Post-fixation was performed with 1% osmium tetroxide for 2 hours. Dehydration was carried out through graded acetone steps (30% to 100%, each for 15 minutes) and repeated with final dehydration in 100% acetone three times.

Samples were embedded sequentially in Spurr’s resin, progressing from 20%, 50%, 75%, and finally 100% resin, with overnight incubation at each step. After three additional changes in fresh 100% resin (one per day), samples were polymerized at 60 °C for 24 hours. Ultrathin sections (60–90 nm) were prepared using a Power Tome XL ultramicrotome (RMC Boeckeler, Tucson, USA), mounted on Formvar-coated 150 mesh copper grids, and stained with 2% uranyl acetate followed by Reynolds’s lead citrate. Imaging was carried out using a JEOL JEM-1400 Flash transmission electron microscope equipped with a Metataki Flash CMOS camera. Measurements of cyanoglobule size and number were quantified using ImageJ.

Isolation of cyanoglobules

Isolation of cyanoglobule fractions was performed based on a previously published protocol [17] with slight modifications. Stationary phase cultures (50 mL, OD750 ~ 2.0) were pelleted and washed twice in Buffer A (25 mM HEPES-KOH pH 7.8, 250 mM sucrose). Cell disruption was achieved by three successive passes through a French Press (AMINCO FA-028) at 1100 psi. Cell lysates were centrifuged at 150,000 × g for 30 minutes at 4 °C after overlaying the homogenate with medium R (50 mM HEPES-KOH pH 8.0, 5 mM MgCl₂). The resulting floating lipid-rich layer was collected as crude cyanoglobules.

To further purify cyanoglobules, the crude fraction was mixed with medium R containing 0.7 M sucrose and layered beneath medium R containing 0.2 M sucrose. This gradient was centrifuged at 150,000 × g for 90 minutes at 4 °C. The resulting cyanoglobule fraction was harvested, lyophilized, and stored at –80 °C. All buffers contained protease and phosphatase inhibitor cocktails (74 μm antipain, 130 μm bestatin, 16.5 μm chymostatin, 56 μm E64, 2.3 μm leupeptin, 37 μm phosphoramidon, 209 μm AEBSF, 0.5 μm aprotinin, 50 mm NaF, 25 mm β-glycerophosphate, 1 mm Na-orthovanadate and 10 mmNa-pyrophosphate) to preserve protein integrity.

Proteomics analysis of purified cyanoglobules from the Anabaena^ΔN strain

Lyophilized cyanoglobule samples were reconstituted in SDS-PAGE buffer, heated at 60 °C for 10 minutes, and loaded on 12.5% BioRad Criterion gels. After brief electrophoresis (50 V for ~20 min), proteins were visualized using the Pierce Silver Stain kit (Thermo Fisher, Waltham, USA). Proteins were excised for in-gel digestion.

Protein digestion followed a modified Shevchenko protocol [18]. Bands were dehydrated in acetonitrile, reduced with DTT (10 mM in 100 mM ammonium bicarbonate) at 56 °C for 45 min, alkylated with 50 mM iodoacetamide, and rehydrated in trypsin (0.005 µg/µL) overnight at 37 °C. Peptides were extracted using 60% acetonitrile with 1% TFA and dried via vacuum centrifugation.

Peptides were then resuspended in 100 mM TEAB and labeled using Thermo Scientific 11-plex TMT reagents per the manufacturer’s instructions. Labeled peptides were cleaned using C18 SepPak cartridges, dried, and resuspended in 2% acetonitrile with 0.1% TFA. Peptides were separated on a Thermo EASY-nLC 1200 system using a PepMap RSLC trap and resolving column, with a 45-minute gradient and 300 nL/min flow rate.

Mass spectrometry was performed using a Q Exactive HF-X mass spectrometer in data-dependent acquisition mode. MS1 resolution was 120,000 (m/z 200), followed by HCD fragmentation and MS2 acquisition at 45,000 resolution. Data were processed with MaxQuant v1.6.15.0 [19], using as isobaric labels the ‘TMT10plex-Lys130C’, TMT10plexL131N’ and ‘TMT11plex-Lys131C’. Methionine oxidation, Ser/Thr/Tyr phosphorylation and N-terminal acetylation were set as variable modifications. Carbamidomethylation was set as fixed modification. Data was searched against a custom-built Anabaena protein database that concatenates protein sequence predictions from our genomic sequencing of the Anabaena^ΔN strain and the Anabaena^WT strain (12287 protein entries) with a reversed-sequence decoy database constructed bin the MaxQuant pipeline to estimate false discovery rate. Peptide and protein-level false-discovery rate was set to 1%. The mass spectrometry proteomics data have been deposited to the ProteomeXchange Consortium via the PRIDE partner repository [20] with the dataset identifier PXD069172.

Lipidomics analysis by LC-MS/MS

Immediately following lyophilization, lipid extraction was carried out using 300 µL of a solvent mixture consisting of methanol, chloroform, and formic acid in a 20:10:1 ratio (v/v/v). To induce phase separation, 150 µL of a buffer containing 0.2 M phosphoric acid and 1 M potassium chloride was added, and samples were centrifuged at 13,000 × g for 1 minute at room temperature. The organic (chloroform) layer, containing the extracted lipids, was carefully collected, evaporated under a nitrogen gas stream, and reconstituted in 100 µL of high-purity isopropanol. The resulting lipid extracts were filtered with 0.45-µm PTFE syringe filter (Restek Corp., Bellefonte, USA) and transferred into amber HPLC vials equipped with glass inserts to prevent adsorption loss.

Lipidomic profiling was performed using a Thermo Scientific Q-Exactive hybrid quadrupole-Orbitrap mass spectrometer integrated with a Vanquish ultra-high-performance liquid chromatography (UHPLC) system operating in positive ion electrospray ionization mode. Chromatographic separation was achieved on a C18 reverse-phase Acquity column (Waters) over a 20-minute run. The mobile phases included Solvent A (acetonitrile:water 60:40, v/v) and Solvent B (isopropanol:acetonitrile 90:10, v/v), both containing 0.1% formic acid and 10 mM ammonium formate. The elution program consisted of the following gradient: 20% B for the first 2 minutes, ramping to 43% B from 2.1 to 11.9 minutes, 54% B at 12 minutes, followed by 70% B from 12.1 to 18 minutes, then 99% B at 18.1 to 19 minutes, and re-equilibration to 20% B at 20 minutes. The injection volume was 10 µL, with a flow rate of 0.4 mL/min, and the column temperature maintained at 55 °C.

MS1 full scan spectra were acquired in the m/z 200–2000 range, and data-dependent MS2 fragmentation was performed for the five most intense precursor ions per scan cycle. Data acquisition and peak detection were optimized for retention time precision and mass accuracy.

Acquired data files were processed using Progenesis QI software version 2.3 (Nonlinear Dynamics, Newcastle-upon-Tyne, UK). The software was used to perform baseline correction, retention time alignment, peak detection, and normalization. Statistical filtering of the identified features was carried out with a threshold of p < 0.01 and q < 0.01 to retain high-confidence lipid species. Lipid annotation was performed by matching accurate mass and fragmentation spectra against the LipidBlast library within Progenesis QI and further validated by comparison with reference spectra from LipidMaps and the Metabolomics Workbench.

To distinguish lipid profile changes between treatment groups, multivariate statistical analysis was conducted using orthogonal partial least squares discriminant analysis (OPLS-DA) in EZinfo version 3.0.3 (Sartorius, Göttingen, Germany). Lipid enrichment and class-level distributions were calculated based on normalized ion intensities and visualized graphically for comparative interpretation.

Development of a non-N2-fixing strain of Anabaena

An Anabaena strain that lost its nitrogen fixation ability (Anabaena^ΔN) was identified following successive sub-culturing of cells at least ten times through nitrogen-replete BG11 medium and subsequently testing growth under nitrogen-deficient conditions (BG11^-N). Genome sequencing and SNP analysis revealed mutations in several genes important for heterocyst formation and function (S1 Table). Among them, there were three genes essential for growth in the absence of nitrogen: alr2817 (hetC), alr2887 (devB), and alr0952 (coxC). These genes are required for differentiation of heterocysts or protection of the nitrogenase from oxidation [21–24].

In addition to these three core genes, four other candidates—alr3956 (ndhF), alr3355 (ndhH), all4019 (zwf), and alr0693—may also contribute to the inability of Anabaena^ΔN strain to fix nitrogen and potentially influence other aspects of cellular physiology. These genes are involved in supporting heterocyst metabolism. A summary of the whole genome sequence results of Anabaena^WT and Anabaena^ΔN is presented in S1 Fig.

Increased size and abundance of Anabaena^ΔN cyanoglobules under various stresses

We first tested cyanoglobule proliferation under a suite of three different stressors, N-deficiency, P-deficiency and darkness (Fig 1, S2 Fig). To examine cyanoglobule accumulation, Anabaena^ΔN cells were stained with monodansyl pentane (MDH), commonly used for lipid droplet detection [25].

Download:

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

Fig 1. Visualization and quantification of cyanoglobules in Anabaena^ΔN cells under nitrogen deficiency.

(A) Representative confocal laser scanning microscopy (CLSM) images showing cyanoglobule accumulation stained with monodansyl pentane (representative cyanoglobules are indicated with white arrowheads) in control (17.6 mM N; BG11,) and nitrogen-deficient conditions (0.55 mM N; BG11^-N). Scale bars indicate 2 µm. (B) Transmission electron micrographs (TEM) illustrating osmiophilic cyanoglobule structures (indicated by white arrowheads) in Anabaena^WT and Anabaena^ΔN. Scale bars indicate 500 nm. (C) Quantitative analysis of cyanoglobule abundance (number of cyanoglobules per cell) obtained from CLSM and TEM data, and diameter measurements (nm) derived from TEM. Data represent the mean ± 1 standard deviation of three biological replicates (n = 3). Statistical significance was determined using a two-tailed unpaired Student’s t-test; asterisks indicate statistically significant differences of p < 0.05.

https://doi.org/10.1371/journal.pone.0343220.g001

CLSM observations demonstrated that all three stresses promote greater numbers of cyanoglobules per cell, particularly for P-deficiency and N-deficiency (Fig 1A, S2 Fig, S2 Table). Transmission electron microscopy (TEM) confirmed this trend, revealing darkly stained osmiophilic lipid globules consistent with the previously described cyanoglobule morphology in Synechocystis [7] (Fig 1B & 1C, S2 Fig, S2 Table). Notably, the average diameter of cyanoglobules also increased substantially under each of the stresses. Again, this was most striking for P-deficiency and N-deficiency (Fig 1B 1C, S2 Fig, S2 Table). Thus, both microscopic methods revealed consistent trends in cyanoglobule abundance and size, reinforcing the conclusion that cyanoglobule formation and accumulation are markedly enhanced under stress.

Growth and pigmentation phenotypes of Anabaena^ΔN

Because nitrogen availability strongly influences photosynthetic metabolism and lipid storage, we used nitrogen deprivation as a physiological model to investigate how cyanoglobules respond to nutrient stress. In Arabidopsis, nitrogen limitation leads to an increase in both the size and number of plastoglobules, reflecting enhanced lipid remodeling [26]. Thus, we continued our investigation by focusing on the composition and behavior of cyanoglobules under N-deficiency. After transferring Anabaena^ΔN into nitrogen-deficient media (0.55 mM nitrogen; BG11), compared to control conditions (17.6 mM nitrogen; BG11^-N), distinct physiological changes were evident. By day four of nitrogen deprivation, cells exhibited lighter pigmentation compared to control growth conditions (Fig 2A), indicative of chlorosis. This visual phenotype aligned well with optical density (OD₇₅₀) and chlorophyll a levels (S3 Fig), and were consistent with the previous reports of other cyanobacterial species under nutrient deficiency [27–29].

Download:

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

Fig 2. Isolation and purification of cyanoglobules from Anabaena^ΔN grown in control and N-deficient media.

(A) Growth and pigmentation phenotypes of Anabaena^ΔN under control (BG11) and varying levels of N-deficiency. (B) Separated cell lysate (left panel, sucrose gradient) and floated cyanoglobules (right panel, sucrose gradient) from cultures grown in BG11 and BG11^-N media. Note that yellowish lipid pads were obtained from control BG11 cultures, while bright-orange pads were isolated from nitrogen-deficient BG11^-N conditions. The final purified cyanoglobule fraction appeared as a single lipid pad atop the sucrose gradient. (C) Silver staining of SDS-PAGE demonstrates a common banding pattern of isolated cyanoglobules from cells grown under both control and N-deficient conditions.

https://doi.org/10.1371/journal.pone.0343220.g002

To examine whether Anabaena^ΔN exhibits altered cell morphology, we compared Anabaena^WT and Anabaena^ΔN under nitrogen-depleted conditions. In the wild-type, growth in BG11^-N medium without nitrogen supplementation led to the expected formation of heterocysts, which were clearly distinguishable along the filaments (S4A Fig). In contrast, the ΔN strain displayed heterogeneous morphologies, including filaments with normal-appearing cells as well as abnormally enlarged and elongated cells (S4B Fig). Additionally, while wild-type cultures remained as a uniform green suspension, the ΔN strain consistently developed clumps in liquid culture, indicating an altered physiological response to nitrogen starvation. Although we cannot definitively determine whether the observed morphological changes arise directly from the loss of the targeted genes or represent a secondary response to the inability to fix atmospheric N₂, the results highlight that Anabaena^ΔN has a clear defect in N metabolism that manifests in both filament organization and culture-level behavior.

The increased size and number of cyanoglobules observed in Anabaena^ΔN cells subjected to nitrogen deficiency align closely with previous observations in Synechocystis under phosphorus deprivation, and parallel the stress-induced plastoglobule formation commonly reported in land plants [30]. These findings collectively suggest a conserved and dynamic role of cyanoglobules as lipid reservoirs or stress-responsive organelles, modulating cellular responses under nutrient-limited conditions.

Isolation and purification of cyanoglobules from Anabaena^ΔN

To examine the biochemical composition of cyanoglobules under nitrogen-limiting conditions, cyanoglobules were isolated from Anabaena^ΔN. Cultures (50 mL) were grown under nitrogen-deficient BG11^-N medium to impose a N-deficient stress until they reached an optical density (OD₇₅₀) of 2.0 per mL, the optimal stage for harvesting cyanoglobule-rich cells (S3 Fig).

Cells subjected to nitrogen deficiency displayed visibly lighter pigmentation (Fig 2A), indicative of chlorosis, and had reduced chlorophyll a content (2.6 ± 0.4 µg chl a mL^-1) compared to control cultures (3.5 ± 0.6 µg chl a mL^-1). These differences occurred despite the cultures having comparable cell densities based on optical density measurements. This visual and biochemical shift in pigmentation provided an indication of physiological stress and cyanoglobule enrichment.

Following cell harvest, a crude cyanoglobule fraction was obtained via two-step differential centrifugation. As illustrated in Fig 2B, the initial step produced two distinct lipidic layers, visible as yellowish (control) or bright orange (nitrogen deficient) zones, suggesting compositional differences in pigment or lipid content. The two lipid pads from each growth condition were pooled and subjected to further purification by sucrose gradient centrifugation. This process resulted in a single, dense lipid pad floating at the top of the gradient, corresponding to enriched cyanoglobules.

To validate the co-isolation of proteins with the isolated cyanoglobule fraction, samples were resolved using SDS-PAGE and visualized by silver staining. As shown in Fig 2C, multiple protein bands were observed in both control and N-deficient cultures. The purified cyanoglobules were subsequently subjected to proteomic and lipidomic analyses for in-depth characterization of the molecular composition.

Proteomic profiling of cyanoglobules in Anabaena^ΔN

To investigate the proteomic composition of cyanoglobules in Anabaena^ΔN, we conducted TMT-based quantitative proteomic analysis of purified cyanoglobules isolated from cells grown under both control and nitrogen-deficient conditions. A total of 97 proteins were confidently annotated and quantified at a peptide and protein level false discovery rate of 1% (S3 Table). Thirteen proteins identified in the isolated cyanoglobules have orthologs in the Synechocystis sp. PCC 6803 cyanoglobule proteome [7] and/or the A. thaliana plastoglobule proteome [5], pointing to the commonality in the composition across disparate cyanobacterial species [7,31–33]. We considered these thirteen proteins to represent the putative cyanoglobule proteome of Anabaena.

The most abundant and differentially enriched proteins under N-deficiency were homologs of NAD(P)H dehydrogenase C1 (NDC1, alr4094) and Plastoglobular protein 18 (PG18, alr4514), with log₂ fold-changes of 4.28 and 4.80, respectively (Table 1). These two proteins are well-established plastoglobule components in A. thaliana [5,33] and Synechocystis [7]. Interestingly, ndbB is also the most abundant protein of the cyanoglobule core proteome in Synechocystis sp. PCC 6803 [4]. NDC1 is essential for electron transport and plastoquinone redox homeostasis. A. thaliana ndc1 mutants exhibit elevated oxidative stress, altered plastid lipids and early bolting, reflecting its importance in plastidic metabolic regulation [34–36]. Similarly, functional studies have shown that PG18 mutants display thylakoid disorganization, reduced photosystem efficiency, and increased photodamage due to impaired chloroplast structure and photoprotective capacity [37].

Download:

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

Table 1. Dynamic changes to the Anabaena^ΔN cyanoglobule proteome under N-deficiency.

https://doi.org/10.1371/journal.pone.0343220.t001

Alongside these core proteins, other conserved plastoglobule markers were also enriched in cyanoglobules. The identification of Fibrillin 2 (FBNβ, all4697) and ABC1K3 (all4960) supports the conserved nature of these structures [33]. While two FBN isoforms are encoded in Anabaena, only FBNβ was detected in the cyanoglobules, paralleling findings from Synechocystis where only one FBN isoform (FBNβ) is cyanoglobule-localized. Likewise, ABC1K3 was the sole member of the four encoded ABC1 kinases present in the cyanoglobules, suggesting selective localization and a conserved role in plastoglobule-related signaling or lipid metabolism.

In addition to these canonical components, several proteins involved in redox regulation and stress response were also identified. Notably, an aldo/keto reductase-like protein (AKR-like, all1225) and multiple short-chain dehydrogenase/reductase (SDR) family members such as all5305 and alr2751 exhibited high enrichment, with log₂ fold-changes of 3.65 and 2.74, respectively. These findings support the role of cyanoglobules in managing oxidative stress, a well-characterized function of plastoglobules [2,34]. Additional NAD(P)-binding oxidoreductases, including flavin reductase-related 1 (alr1788) and flavin reductase-related 2 (all0772/alr2751), were also abundant, indicating active redox metabolism within the compartment.

Interestingly, several proteins associated with pigment and lipid metabolism were highly enriched under nitrogen deficiency. The Neoxanthin-deficient 1 (NXD1, all0736) homolog showed a strong increase under N-deficiency (log₂ FC ~ 4.06), suggesting a role for cyanoglobules in carotenoid biosynthesis or regulation during nitrogen stress. Similarly, α/β-hydrolase (alr1709) and β-carotene ketolase (all3744) were enriched (log₂ FC 3.64 and 3.73, respectively), further implicating cyanoglobules in pigment remodeling and lipid turnover processes.

The dataset also included proteins with less-defined or potentially novel functions. Among them, UPF0426 protein (asl4034) was highly enriched (log₂ FC ≈ 4.8). Although UPF0426 is not typically associated with plastoglobules in higher plants, its possible localization to cyanoglobules in Anabaena suggests a specialized adaptation in filamentous cyanobacteria. Together, these observations highlight a conserved yet functionally diversified proteome architecture, where core lipid-metabolic and redox proteins are supplemented by lineage-specific factors to accommodate the physiological needs of Anabaena [2,28,30,36,38].

Taken together, these results reveal a cyanoglobule proteome in Anabaena that shares several conserved features with land plant plastoglobules, including lipid metabolism, redox regulation, and stress response functions. At the same time, the presence of unique or previously unreported proteins suggests that cyanoglobules have evolved species-specific adaptations, potentially reflecting differences in environmental responses and plastidial function in cyanobacteria compared to the modern plants.

Lipidomic profiling of cyanoglobules in Anabaena^ΔN

Plastoglobules in land plants and their cyanobacterial analogs, cyanoglobules, are known to host specialized lipid classes such as prenyl-lipids, redox-active quinones, and stress-associated carotenoids [3–5,7,38]. To investigate how nitrogen deprivation affects the cyanoglobule lipidome in Anabaena^ΔN, we performed LC-MS/MS-based lipidomic analysis on cyanoglobules isolated from cells grown under control and nitrogen-deficient conditions. The results reveal that nitrogen stress triggers extensive lipid remodeling within cyanoglobules, marked by the accumulation of plastoquinone derivatives and specific carotenoids, and a concurrent depletion of thylakoid-associated lipids such as MGDG, TAGs, and chlorophyll breakdown products (Fig 3, Table 2, S4 Table).

Download:

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

Table 2. LC-MS-based quantification of cyanoglobule lipids under N-deficiency.

https://doi.org/10.1371/journal.pone.0343220.t002

Download:

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

Fig 3. Lipidomic analysis of cyanoglobules from Anabaena^ΔN under nitrogen-deficient conditions.

(A) Principal Components Analysis of cyanoglobule lipidomes under control (BG11) and nitrogen-deficient (BG11^-N) conditions. Each colored dot represents a biological replicate (n = 4), each gray dot represents an individual lipid species. The clear separation along Principal Component 1 reflects the highly distinct lipid profiles between growth media. (B) Total lipid class abundance in cyanoglobules of Anabaena^ΔN measured as % of total MS1 ion intensity of all annotated lipids. Inset: TAG content shown on a separate axis due to lower relative abundance. (C) Fold-changes of specific lipid species in cyanoglobules of Anabaena^ΔN grown in BG11^-N relative to BG11 media. Selected lipids that change with statistical significance are presented (p < 0.05, Student’s t-test, n = 4 biological replicates, each of which is an independently grown culture). Several plastoquinone (PQ) and plastoquinol (PQH₂) species were significantly upregulated, particularly PQ-B (C16:0, C18:1, C18:2) and PQH₂-E (18:2). In contrast, chlorophyll breakdown products (pheophorbide a, pheophytin), MGDG species, TAGs, and α-tocopherol were markedly reduced.

https://doi.org/10.1371/journal.pone.0343220.g003

Principal Component Analysis (PCA) of the cyanoglobule lipidome revealed a clear separation between control and nitrogen-deficient samples (Fig 3A), indicating robust and reproducible differences in lipid composition. Broad class-level remodeling was observed (Fig 3B), with quinones becoming the dominant lipid group under nitrogen deficiency, comprising over 40% of the total annotated lipid pool. This shift was driven by significant enrichment of plastoquinone derivatives, particularly PQ-B; C16:0 (log2 FC 1.45), PQ-B; C18:1 (1.23–1.27), and PQH₂-E; 18:2 (1.04). The accumulation of these redox-active lipids suggests an adaptive response to oxidative stress, consistent with the identification of Glutaredoxin and NDC1 (ndbB) in cyanoglobules [39–42].

Among the plastoquinones, PQH₂-E; 16:0 and 18:1 were among the most abundant species under both conditions, while plastochromanol-8 (PCH₂–8) − a reduced cyclized derivative of PQ-9 − showed a modest increase (log2 FC 0.16). Canonical PQ-9 (i.e., PQ-A) also rose by ca. 50%. Several minor variants, including PQ-C, increased moderately, supporting the idea that cyanoglobules function as dynamic reservoirs for redox lipids during nitrogen stress [39].

In contrast, chlorophyll-related metabolites were strongly depleted. Pheophorbide a, a major catabolite of chlorophyll breakdown, decreased by over 4-fold (log2 FC –2.16), and pheophytin by nearly 2-fold (log2 FC –0.81). These trends support the hypothesis that cyanoglobules play a role in chlorophyll turnover and degradation during stress, potentially aiding detoxification or recycling processes. This pattern also aligns with the visual chlorosis observed under nitrogen deprivation [29]. Carotenoid content showed selective changes. Echinenone, the most abundant carotenoid of the cyanoglobules, exhibited divergent regulation, with one isomer modestly increasing (log2 FC 0.78) while another decreased (log2 FC –0.16). Notably, canthaxanthin (9- or 13-cis) showed a nearly 2-fold increase (log2 FC 0.94), whereas α-tocopherol declined significantly (log2 FC –0.75), possibly reflecting increased oxidative stress consumption. These patterns echo changes at the plastoglobules in land plants, where carotenoids contribute to both structure and photoprotection [43,44].

Galactolipids, especially monogalactosyl diacylglycerol (MGDG), were broadly depleted. MGDG; 32:2 and MGDG; 34:1 declined by log2 fold-changes of –0.92 and –0.41, respectively. This likely reflects membrane lipid turnover and thylakoid remodeling. Interestingly, MGDG; 36:1 increased (log2 FC 0.80), suggesting differential regulation of specific lipid species. In contrast to Synechocystis, where galactolipids increase to support cyanoglobule proliferation under phosphate stress, their depletion in Anabaena may indicate thylakoid breakdown or lipid mobilization under nitrogen limitation.

Neutral lipids such as triacylglycerols (TAGs) showed a consistent and substantial decrease. TAG; 46:0, TAG; 48:0, and TAG; 52:0 all declined by log2 FC values between –1.7 and –1.9. This contrasts with findings in Synechocystis, where TAGs accumulate in cyanoglobules under phosphate deprivation, suggesting that TAG regulation in cyanoglobules is stress-specific and/or species-dependent.

Taken together, these findings define the nitrogen-responsive lipid architecture of Anabaena cyanoglobules. The pronounced enrichment of plastoquinone derivatives and specific carotenoids, combined with a reduction in chlorophyll catabolites, TAGs, and thylakoid-derived lipids, supports the view that cyanoglobules serve as specialized compartments for redox buffering and lipid remodeling during nutrient stress [5]. These results reinforce the emerging model of cyanoglobules as dynamic organelle-like structures, analogous to land plant plastoglobules, capable of reorganizing their lipid and protein content in response to environmental cues, promoting stress adaptation of the organism.