Calcium impacts carbon and nitrogen balance in the filamentous cyanobacterium Anabaena sp. PCC 7120.

Calcium is integral to the perception, communication and adjustment of cellular responses to environmental changes. However, the role of Ca(2+) in fine-tuning cellular responses of wild-type cyanobacteria under favourable growth conditions has not been examined. In this study, extracellular Ca(2+) has been altered, and changes in the whole transcriptome of Anabaena sp. PCC 7120 have been evaluated under conditions replete of carbon and combined nitrogen. Ca(2+) induced differential expression of many genes driving primary cellular metabolism, with transcriptional regulation of carbon- and nitrogen-related processes responding with opposing trends. However, physiological effects of these transcriptional responses on biomass accumulation, biomass composition, and photosynthetic activity over the 24h period following Ca(2+) adjustment were found to be minor. It is well known that intracellular carbon:nitrogen balance is integral to optimal cell growth and that Ca(2+) plays an important role in the response of heterocystous cyanobacteria to combined-nitrogen deprivation. This work adds to the current knowledge by demonstrating a signalling role of Ca(2+) for making sensitive transcriptional adjustments required for optimal growth under non-limiting conditions.

Introduction

Cyanobacteria form a diverse clade of highly specialized bacteria able to perform oxygenic photosynthesis and thus are proclaimed to be the evolutionary ancestors of the chloroplasts of photoautotrophic eukaryon class="Chemical">tes (Douglas, 1998; Falcon ). They comprise five different groups based on their morphology and metabolic pathways (Rippka ). For instance, there are unicellular cyanobacteria, such as the model organism Synechocystis sp. PCC 6803 (Kaneko and Tabata, 1997), and multicellular (filamentous) cyanobacteria, of which Anabaena sp. PCC 7120 (hereafter designated as Anabaena) is one of the common models (Kaneko ). Anabaena also represents the heterocystous cyanobacteria that can fix atmospheric nitrogen in conditions lacking combined nitrogen (Wolk, 1996; Flores and Herrero, 2005; Zhang ).

Calcium (n class="Chemical">Ca2+) is well known to play an essential role as second messenger and signalling molecule in all living organisms (Clapham, 1995). In plants, Ca2+ pulses of specific magnitude, duration, frequency, and source are called Ca2+ ‘signatures’ and are triggered in response to many different environmental stimuli in order to activate response mechanisms (Knight M. , 1992; Knight H. , 1997; Knight and Knight, 2000). However, besides the calcium signals, the cytosolic free Ca2+ concentration ([Ca2+]i) is tightly regulated by Ca2+-binding proteins, Ca2+ pumps and other transporters to maintain low [Ca2+]i levels of about 100nM in order to avoid toxic calcium phosphate precipitation within the cell.

Recently, an important role of Ca2+ sequestration and signalling has been highlighted in regulating the activities of chloroplasts (Stael , b; Rocha and Vothknecht, 2012; Nomura and Shiina, 2014). Meanwhile, in cyanobacteria, the chloroplast ancestors, information concerning n class="Chemical">Ca2+-regulated processes is still rather scarce. Cyanobacteria have been shown to use Ca2+ signalling under various environmental stimuli (Torrecilla ; Barrán-Berdón ), but so far only one cyanobacteria-specific Ca2+-binding protein (Ccbp in Anabaena) has been characterized (Zhao ; Shi ; Hu ; Dominguez ). Ccbp is involved in the signalling cascade resulting in the formation of nitrogen-fixing heterocysts (Flores and Herrero, 2005). The concept of Ca2+ being involved in atmospheric nitrogen fixation was reported as early as the 1930s by Burk , and was supported several times over the following decades (Allison ; Allen and Arnon, 1955; Mishra ). It is now known that Ca2+ controls this process in Anabaena through the negative regulation of Ccbp by HetR and the transcription factor NtcA, by which the release of Ca2+ induces the differentiation of vegetative cells into nitrogen-fixing heterocysts. The process of heterocyst formation in Anabaena was shown to be slightly enhanced in nitrogen-depleted conditions under elevated exogenous Ca2+ concentrations ([Ca2+]e) up to 1–2mM, while concentrations above that have inhibitory and toxic effects on cell growth and health (Smith ; Singh and Mishra, 2014). Despite the apparent impact of Ca2+ on cell growth and development, changes in the amplitude and kinetics of Ca2+ signatures evoked upon nitrogen starvation and increased [Ca2+]e were shown to be minor (Torrecilla ).

Ca2+ effects have also been studied with respect to photosynthetic activity in cyanobacteria. Enhanced n class="Chemical">oxygen evolution resulted from increasing [Ca2+]e in different cyanobacteria species, which may be linked to the involvement of Ca2+ in the photosystem II (PSII) oxygen evolving complex (Piccioni and Mauzerall, 1978). The link between Ca2+ and photosynthesis has received much more attention in plants (see Hochmal for a recent review), where it is known that the transition from light to darkness triggers a cytosolic Ca2+ signal that inhibits the enzymes of the Calvin–Benson–Bassham cycle. This observation of a dark-induced calcium signal has also been demonstrated in Anabaena (Torrecilla ). This signal is most likely dependent on the redox state of photosynthetic electron transport chain components in plants as well as in cyanobacteria.

In this work, the impact of [Ca2+]e on the induction of intracellular n class="Chemical">Ca2+ signalling in Anabaena has been explored through the analysis of gene expression, cell fitness and development. Here, it was found that [Ca2+]e affects the C:N balance through short-term and long-term reprogramming of carbon- and nitrogen-specific metabolism, including up- and down-regulation of transporters, photosynthetic genes and transcription regulators. Changes in [Ca2+]e impacted the accumulation of biomass and proteins, but had no significant effect on heterocyst differentiation in growth medium supplied with nitrate as a nitrogen source.

Materials and methods

Cultures and growth conditions

Anabaena sp. PCC 7120 cells were grown either in regular BG11 medium (Rippka ) (low n class="Chemical">Ca2+; 0.25mM CaCl2) or in BG11 medium plus additional CaCl2 (high Ca2+; 1mM CaCl2). Both media were supplemented with 10mM TES-KOH (pH 8.0). The cultures were cultivated under continuous low light (25–35 μmol photons m−2 s−1) at 30 °C and in 3% (vol/vol) CO2-containing air with mild shaking.

Calcium shift experiment

Anabaena cells were grown under normal growth conditions in low and high n class="Chemical">Ca2+ media. After reaching OD750=1.0, the cells were centrifuged and the cell pellets resuspended to OD750=0.8 in either low or high Ca2+; for the Ca2+ treatments, pellets from high Ca2+ were resuspended in low Ca2+ medium (shifted from high to low) and pellets from low Ca2+ were resuspended in high Ca2+ medium (shifted from low to high). For the controls, pellets were resuspended in fresh medium with the same [Ca2+] (shifted from low to low or high to high). Afterwards, the cultures were again incubated under normal growth conditions. Over a time range of 24h (0, 1, 2, 4, 8, 16, 24h post-shift), samples were taken for the determination of total proteins, carbohydrates, and biomass. Data were collected from three independent Ca2+ shift experiments (n=3).

Determination of the biomass

A 20ml culture volume was passed through a pre-washed, pre-dried and pre-weighed glass-fibre filter in a glass vacuum filtration apparatus and then washed with distilled water. Filters were dried at 60 °C for 24h, stored in a desiccator and then weighed.

Determination of the protein content

A 0.2ml volume of culture was taken in triplicate, centrifuged and washed with distilled water, and the pellets were frozen. The total protein content was determined according to a modified Lowry procedure (Markwell ). OD750 was measured against a n class="Species">bovine serum albumin (BSA) standard curve with a Lambda 25 UV/VIS spectrometer (Perkin Elmer).

Determination of total sugars content

A 1.0ml volume of culture was taken in triplicate, spun down and washed with distilled water. The total n class="Chemical">sugar content was analysed by the colorimetric method described by DuBois after diluting the samples 1:1 with milliQ water.

RNA isolation and next-generation sequencing

Anabaena cultures were shifted between low and high n class="Chemical">Ca2+ media as described above. Each of the four samples (low Ca2+ to high Ca2+, high Ca2+ to low Ca2+, and their controls) was represented by three independent biological replicates (n=3). One hour after the shift, the cultures were centrifuged (6000×g, 6min, RT) and the pellets immediately resuspended in 0.2ml RNA resuspension buffer (0.3M saccharose+10mM sodium acetate, pH 4.5) and 60 μl 0.25M ethylenediaminetetraacetic acid (EDTA) was added. A 0.3ml volume of lysis buffer (2% sodium dodecyl sulfate+10mM sodium acetate, pH 4.5) was added, followed by 0.5ml of phenol:chloroform:iso-amyl alcohol (P:C:IAA=25:24:1). Then the samples were vortexed thoroughly, incubated for 5min at 95 °C and centrifuged for 15min at 15000×g, 4 °C. The upper phase was collected, twice re-extracted with an equal volume of P:C:IAA and then extracted with C:IAA (24:1). The aqueous phase was supplemented with 10M lithium chloride to a concentration of 2.5M, and incubated overnight at –20 °C. The next day, the samples were centrifuged as before and the pellets washed with 1ml 70% ethanol by gentle mixing. After centrifugation, the RNA pellets were air-dried and resuspended in small volumes of milliQ water at 65 °C for 15min. RNA isolates were treated with DNase. RNA samples were submitted to the Turku Centre for Biotechnology (Turku, Finland) for RNA sequencing using a HiSeq 2000 (Illumina).

RNAseq data analysis

RNAseq reads were aligned using the reference genome and annotations of Nostoc sp. PCC 7120, downloaded from Ensembl (EBI). Alignment was using the Tophat algorithm, and ORF calling and read-depth quantification were carried out using the open source analysis software Chipster (CSC, Finland). Significantly differentially expressed genes were identified using a false discovery rate (FDR: Benjamini–Hochberg) cutoff of 0.05. Gene descriptions were collected from Cyanobase (genome.microbedb.jp/cyanobase/) and KEGG (www.genome.jp/kegg/).

Photosynthetic activity measurements

Oxygen evolution rate was measured using a Clark-type n class="Chemical">oxygen electrode (Hansatech Oxytherm) and quantum yield of PSII (Y(II)) was measured using a Dual-PAM-100 (Walz). For both, samples were dark-adapted for 5min. For the gross oxygen evolution rate in light, the rates of oxygen consumption obtained in darkness were added to oxygen evolution rates obtained in saturating light (intensity: 400 μmol photons m−2 s−1), and values were normalized to the chlorophyll content (chl a estimated in 90% methanol at OD665). For the determination of Y(II), samples were diluted to a chlorophyll concentration of 5 μg chl a ml–1 and the fluorescence parameters F v and F m determined according to Genty .

Light and fluorescence microscopy

Bright-field and fluorescence images were taken using an AxioVert 200M fluorescence microscope (Zeiss) and a Wetzlar light microscope (Leitz) on ×40 magnification. From active Anabaena filaments, 1000–2000 cells were counted for each treatment and the heterocyst frequency calculated as a percentage of total cells counted.

Determination of nitrogen concentration

Nitrate concentration of filtered media was determined spectrophotometrically using a Spectroquant n class="Chemical">Nitrate Test kit (Merck).

Results

Calcium induces both short-term and long-term changes in gene expression

The response of Anabaena sp. PCC 7120 to changes in external n class="Chemical">Ca2+ conditions was studied by pre-growing cells to OD750=1.0 (over approximately 3 days) in either low or high Ca2+ medium (0.25 and 1mM CaCl2, respectively), and then replacing the medium with one of either the same Ca2+ concentration (acting as controls), or with an alternative Ca2+ concentration to create an up-shift (henceforth referred to as ‘high Ca2+’) or a down-shift (‘low Ca2+’) in [Ca2+]. The transcriptomic reaction of Anabaena cells to the change in [Ca2+] was analysed 1h post-shift by transcriptome sequencing. Gene expression in high Ca2+ and low Ca2+ were compared with the corresponding controls to identify genes that had a fold change (FC) ≥2 (log2 FC≥1) between treatment and control (FDR<0.05). Genes that were significantly differentially expressed after 1h in altered [Ca2+] were designated ‘short-term Ca2+-responsive genes’ (see Table 1). Genes were also identified that showed no significant short-term response to Ca2+ (i.e. were not differentially expressed between the shifts and their corresponding controls), but showed strong differential regulation in response to the Ca2+ conditions of the 3-day pre-culture. These genes were designated ‘long-term Ca2+-responsive genes’ (see Table 2). The absolute expression levels of genes in all four samples were hierarchically clustered to identify common expression profiles (Fig. 1).

Table 1.

Differential expression of selected short-term responsive Ca2+-regulated genes

Gene name	Accession	Description	log 2 fold change in high Ca 2+a	log 2 fold change in low Ca 2+	FDR b	Gene Cluster (Fig. 1)
Bicarbonate import/metabolism
cmpA	alr2877	Bicarbonate-binding protein	2.6	–2.2	0.000	1
cmpB	alr2878	Bicarbonate transport permease	2.5	–2.3	0.000	1
cmcC	alr2879	Bicarbonate transport, ATP-binding protein	2.0	–2.0	0.000	1
cmpD	alr2880	Bicarbonate transport, ATP-binding protein	2.3	–2.4	0.000	1
bicA	all1304	Low affinity bicarbonate Na+ symporter	1.8	–1.5	0.000	1
NhaS3-like	all1303	Na+:H+ antiporter	1.6	–1.4	0.000	1
sbtB	all2133	Bicarbonate Na+ symporter	1.1	–1.8	0.000	1
sbtA	all2134	Bicarbonate Na+ symporter	1.6	–2.1	0.000	1
mrpB	all1837	Na+:H+ antiporter subunit B	1.2	–0.4	0.000	1
mrpA	all1838	Na+:H+ antiporter subunit A	1.2	–0.5	0.001	1
mnhG	asl1839	Na+:H+ antiporter subunit G	1.3	–1.1	0.028	1
mnhF	asl1840	Na+:H+ antiporter subunit F	1.3	–1.0	0.025	1
mnhE	all1841	Na+:H+ antiporter subunit E	1.4	–0.7	0.785b	1
mrpD	all1842	Na+:H+ antiporter subunit D	1.3	–0.5	0.000	1
Nitrate import/metabolism
nirA	alr0607	Ferredoxin:nitrite reductase	–1.5	0.5	0.010	4
nrtA	alr0608	Nitrate/nitrite transport substrate- binding protein	–1.6	0.5	0.011	4
nrtB	alr0609	Nitrate/nitrite transport permease	–1.3	0.5	0.032	4
nrtC	alr0610	Nitrate/nitrite transport ATP-binding protein	–1.2	0.7	0.000	4
nrtD	alr0611	Nitrate/nitrite transport ATP-binding protein	–1.0	0.9	0.005	4
narB	alr0612	Ferredoxin:nitrite reductase	–0.8	0.7	0.000	4
cphB2	all0571	Cyanophycinase, dipeptidase	1.1	0.1	0.001	2
Photosynthesis and carbon metabolism
psbAIII	alr4592	PSII core protein	1.4	–0.4	0.000	2
psbAIV	all3572	PSII core protein	0.7	–0.2	0.002	2
psbH	all4050	PSII reaction centre protein	1.8	0.0	0.000	2
psaE	asl4098	PSI reaction centre subunit	2.0	–0.4	0.000	2
fbaB	all3735	Fructose-bisphosphate aldolase class I	1.0	–0.2	0.007	1
glgB	all0875	Alpha-glucanotransferase	1.4	–0.9	0.000	2
asr3089	asr3089	Transglycosylase-associated protein	1.2	–0.4	0.003	1
alr1850	alr1850	Transketolase	0.8	–0.5	0.032	1
Transcriptional regulators
rbcR1	all0862	LysR-type regulator	1.0	–0.7	0.000	1
Stress-related genes
alr3199	alr3199	Iron- and oxygen-binding HHE domain protein	1.7	–1.0	0.000	2
all0457	all0457	Low temperature-induced protein	1.1	–0.5	0.003	2
all0458	all0458	Low temperature-induced protein	1.5	–0.8	0.000	2
all0459	all0459	Ferritin-like protein, nutrient stress responsive	2.0	0.0	0.000	2
alr5182	alr5182	Short-chain dehydrogenase/reductase, desiccation responsive	1.4	–0.4	0.000	2
alr0896	alr0896	Unknown protein, induced by desiccation	1.3	–1.1	0.028	2
asr1134	asr1134	CsbD stress response protein	0.9	–0.4	0.015	2
flv4	all4446	Flavodiiron protein	0.7	–1.1	0.038	—
flv2	all4444	Flavodiiron protein	0.0	–1.2	0.735b	—
Bidirectional hydrogenase
alr0761	alr0761	Peptidase domain-containing protein	–1.5	1.6	0.033	3
hoxU	alr0762	Bidirectional hydrogenase subunit	–0.4	0.0	0.227b	3
alr0763	alr0763	Hypothetical protein	–0.6	0.5	0.002	3
hoxY	alr0764	Bidirectional hydrogenase subunit	–0.9	0.7	0.000	3
alr0765	alr0765	Hypothetical protein	0.0	1.0	0.818	3
alr0750	alr0750	Hypothetical protein	–0.4	0.2	0.015	3
hoxE	alr0751	Bidirectional hydrogenase, diophorase subunit	–0.8	0.5	0.000	3
hoxF	alr0752	Bidirectional hydrogenase subunit	–0.8	0.6	0.136	3
Other proteins
alr0198	alr0198	Hypothetical protein	1.1	–0.6	0.000	2
alr0199	alr0199	Hypothetical protein	1.1	–0.3	0.000	2
alr3715	alr3715	Hypothetical protein	1.7	–1.4	0.000	2

Genes with log2 expression fold change ≥1 are considered to be differentially expressed. However, in some cases genes of special interest with log2 <1 differential expression have been included.

Data with false discovery rate (FDR)>0.05 are included when a gene belongs to an operon with member genes that have statistically robust FDRs (<0.05).

Table 2.

Selected genes with long-term expression response to Ca2+

All genes shown were <2 FC differentially expressed between the high or low Ca2+ shifts and their respective controls, and were >2 FC differentially expressed between both cultures originating from the high Ca2+ pre-culture (low Ca2+ and control for low Ca2+) and both of those originating from the low Ca2+ pre-culture (high Ca2+ and control for high Ca2+).

Gene name	Accession	Description	FDR	Cluster (Fig. 1)
Up-regulated in high calcium/down-regulated in low calcium
hglE	alr5351	Heterocyst glycolipid synthase	0.007	5
hglD	alr5354	Heterocyst glycolipid synthase	0.014	5
hglC	alr5355	Heterocyst glycolipid synthase	0.009	5
nifB	all1440	Nitrogen fixation protein	0.008	5
nifN	all1437	Nitrogenase iron-molybdenum protein	0.023	5
nifE	all1438	Nitrogen iron/molybdenum cofactor biosynthesis	0.007	5
nifK	all1440	Nitrogenase iron-molybdenum protein	0.008	5
nifH	all1455	Nitrogenase iron protein	0.003	5
nifU	all1456	Nitrogen fixation protein	0.018	5
nifS	all1457	Nitrogenase cofactor synthesis protein	0.004	5
chlN	all5076	Protochlorophyllide reductase subunit	0.012	5
chlL	all5078	Protochlorophyllide reductase iron-sulfur ATP-binding protein	0.006	5
clhB	alr3441	Protochlorophyllide reductase	0.003	5
chlH	all4365	Protoporphyrin IX magnesium chelatase	0.014	5
all0156	all0156	Hypothetical protein	0.003	5
all0157	all0157	Hypothetical protein	0.002	5
all0158	all0158	Hypothetical protein	0.012	5
Down-regulated in high calcium/up-regulated in low calcium
rpl14	all4205	50S ribosomal protein	0.004	6
rpl29	asl4207	50S ribosomal protein	0.028	6
rpl16	all4208	50S ribosomal protein	0.019	6
rps3	all4209	30S ribosomal protein	0.007	6
rpl22	all4210	50S ribosomal protein	0.006	6
rps19	asl4211	30S ribosomal protein	0.026	6
rpl2	all4212	50S ribosomal protein	0.039	6
rpl23	all4213	50S ribosomal protein	0.043	6
rpl4	all4214	50S ribosomal protein	0.003	6
sigG	alr3280	Group 3 sigma factor	0.002	6
sigB2	alr3800	Group 2 sigma factor	0.006	6

Fig. 1.

Hierarchically clustered heatmap of gene expression in Anabaena cells treated with high and low Ca2+, as well as in the controls for each. Columns show the absolute expression of significantly differentially expressed genes from triplicate samples (n=3) for each condition. Major clusters are indicated with numbers corresponding to Table 1 (and are discussed in text). (A) Genes responsive to 1h of Ca2+ treatment, called ‘short-term responsive genes’; (B) Genes resistant to short-term response but responsive to 3 days of Ca2+ treatment, called ‘long-term responsive genes’.

All genes shown were <2 FC differentially expressed between the high or low Ca2+ shifts and their respective controls, and were >2 FC differentially expressed between both cultures originating from the high n class="Chemical">Ca2+ pre-culture (low Ca2+ and control for low Ca2+) and both of those originating from the low Ca2+ pre-culture (high Ca2+ and control for high Ca2+).

Hierarchically clustered heatmap of gene expression in Anabaena cells treated with high and low n class="Chemical">Ca2+, as well as in the controls for each. Columns show the absolute expression of significantly differentially expressed genes from triplicate samples (n=3) for each condition. Major clusters are indicated with numbers corresponding to Table 1 (and are discussed in text). (A) Genes responsive to 1h of Ca2+ treatment, called ‘short-term responsive genes’; (B) Genes resistant to short-term response but responsive to 3 days of Ca2+ treatment, called ‘long-term responsive genes’.

Short-term Ca2+ shifts modulate the expression of carbon and nitrogen transport and photosynthesis genes

Genes related to bicarbonate (n class="Chemical">HCO3 –) uptake were among the most strongly affected short-term Ca2+-responsive genes. All eight components of the three known HCO3 – transporters in Anabaena, which are encoded on three separate operons, underwent 4–6.5 FC up-regulation (log2 FC of 2–2.6) in high Ca2+ shift. Conversely, the same genes were strongly down-regulated in low Ca2+ shift (Table 1). According to their expression profile, the HCO3 − transporter genes populate Cluster 1 (Fig. 1a), which also includes all subunits of the mrp Na+:H+ antiporter complex and the LysR-type transcriptional regulator rbcR1 (all0862). Additionally, several Calvin cycle and metabolic enzymes, including fructose bisphosphate aldolase, transketolase and the petF ferredoxin (all4148), shared the Cluster 1 expression profile.

Cluster 2 genes were strongly up-regulated in the short-term response to high Ca2+ and moderately down-regulated by low n class="Chemical">Ca2+. Several photosynthetic subunits were encoded in Cluster 2, as well as stress-responsive genes including a low-temperature induced operon (all0457–all0459; Sato ), an oxygen-binding HHE domain protein (alr3199), and a desiccation-responsive dehydrogenase (alr5182; Katoh, 2012). The cyanophycinase cphB2 (all0571) was also strongly up-regulated in high Ca2+.

Operons encoding the subunits of the bidirectional ‘hox’ hydrogenase were down-regulated in response to high Ca2+ and up-regulated in low n class="Chemical">Ca2+ (Table 1). These genes were included in expression Cluster 3 together with nifJ and hoxR, which had <2FC differential expression but were nonetheless clearly down-regulated in high Ca2+ shift and up-regulated in low Ca2+ in comparison with the controls.

The nir operon that encodes the nitrate/n class="Chemical">nitrite transporter complex was down-regulated in high Ca2+, and up-regulated in low Ca2+ shift (Table 1), although these genes had low expression in both high Ca2+ and its control culture, and high expression in both low Ca2+ and its control (see Fig. 1A). This suggests that expression of the nir operon and other Cluster 4 genes (Fig. 1A) was predominantly influenced by the conditions of the pre-culture, which were common for each shift and its control; however, these genes also demonstrated a short-term response to Ca2+ that led to differential regulation 1h after the shift.

Calcium has a long-term effect on genes encoding nitrogen metabolism

Genes that were induced by long-term high Ca2+ conditions and repressed by long-term low n class="Chemical">Ca2+ included heterocyst glycolipid synthesis (hgl) genes and several members of the nif operon that encode the heterocyst-specific nitrogen fixing machinery in Anabaena (Table 2). Subunits of the light-independent protochlorophyllide reductase (DPOR) were also among long-term Ca2+ up-regulated genes that are shown in Cluster 5 (Fig. 1B). Cluster 6 genes that underwent significant repression and induction by long-term high and low Ca2+ conditions, respectively, encoded nine proteins of the 50S and 30S ribosome, as well as the group 2 sigma factor sigB2 (also called sigE) and ECF sigma factor sigG (Table 2).

Effects of calcium on Anabaena growth and biomass composition

Biomass accumulation and total protein and total sugars content of n class="Species">Anabaena cells were followed for 24h after cultures were shifted to the same high and low Ca2+ media and using the same control samples as described above. Anabaena cells shifted to low Ca2+ resulted in a biomass penalty. These cells showed a decreasing trend in biomass accumulation throughout the experiment in comparison with the control, from a ratio of approximately 1 early in the experiment to less than 0.9 at 24h (Fig. 2A). This was in contrast to no relative change in biomass observed for cells shifted to high Ca2+.

Fig. 2.

Biomass (A), total protein (B) and total sugars (C) composition of Anabaena sp. PCC 7120 cells shifted to high Ca2+ (0.25mM to 1mM) or low Ca2+ (1mM to 0.25mM) relative to the controls. Data presented are averages of three biological replicates (n=3). Error bars show standard deviations. Significant differences between high Ca2+ and low Ca2+ samples are indicated with asterisks (t-test P<0.05).

Biomass (A), total protein (B) and total sugars (C) composition of n class="Species">Anabaena sp. PCC 7120 cells shifted to high Ca2+ (0.25mM to 1mM) or low Ca2+ (1mM to 0.25mM) relative to the controls. Data presented are averages of three biological replicates (n=3). Error bars show standard deviations. Significant differences between high Ca2+ and low Ca2+ samples are indicated with asterisks (t-test P<0.05).

Whilst the low Ca2+ medium penalized biomass accumulation over time, the relative protein composition of this biomass was higher than that of control cells throughout the 24h growth period (Fig. 2B). The relative protein composition of cells in low n class="Chemical">Ca2+ reached the highest ratio at 8h after the shift in Ca2+ concentration, which correlates with the observed up-regulation of nitrate transport genes. There was less variation in the relative protein composition of cells in high Ca2+, which remained fairly stable at around 1 until 16h, but was significantly lower than the low Ca2+ shift at 24h.

The shift in Ca2+ concentration had no observable effect on the total n class="Chemical">sugars content of Anabaena cells in the current experiment, with the relative concentrations remaining close to 1 and similar between high and low Ca2+ shifts throughout (Fig. 2C).

Effects of calcium on the photosynthetic activity of Anabaena

Fluorescence kinetics of PSII and oxygen evolution in n class="Species">Anabaena cultures shifted to low and high Ca2+ conditions showed an increase in PSII yield in all cultures in the first 2h after the shift (Fig. 3A). After 2h, the PSII yield stabilized at 0.35–0.38 for all cells, except in the low Ca2+ control cells. These cells experienced a further 28% drop in PSII yield from 2 to 4h and then stabilized until the 24h time point. Consequently, a relatively higher photosynthetic yield was observed after 2h for the Anabaena cells shifted to high Ca2+. Similarly, a rapid increase in gross oxygen evolution was observed for all cells over the first 2h, resulting in an approximate doubling of recorded values followed by a slow decline throughout the remainder of the experiment (Fig. 3B).

Fig. 3.

Photosynthetic activity of Anabaena sp. PCC 7120 cells shifted to high Ca2+ (0.25mM to 1mM, open squares) or to low Ca2+ (1mM to 0.25mM, open circles) and of the controls (for high Ca2+, closed squares; and for low Ca2+, closed circles). Dual-PAM measurements of PSII yield (A) and oxygen evolution rates (B) shown here are from single experiments (n=1) that represent photosynthetic response to change in [Ca2+]e. Lines connecting data points are included to aid visualization of the series.

Photosynthetic activity of Anabaena sp. PCC 7120 cells shifted to high n class="Chemical">Ca2+ (0.25mM to 1mM, open squares) or to low Ca2+ (1mM to 0.25mM, open circles) and of the controls (for high Ca2+, closed squares; and for low Ca2+, closed circles). Dual-PAM measurements of PSII yield (A) and oxygen evolution rates (B) shown here are from single experiments (n=1) that represent photosynthetic response to change in [Ca2+]e. Lines connecting data points are included to aid visualization of the series.

Changes in calcium concentration have no effect on heterocyst differentiation in Anabaena in the presence of combined nitrogen

Heterocyst frequency was evaluated at 10h, 20h and 2 d after the change in Ca2+ conditions, and was determined not to be statistically different between low and high n class="Chemical">Ca2+ shifted cells (t-test, 0.95 confidence; P=0.32). Heterocyst frequency data are shown in Fig. 4 and Supplementary Table S1 at JXB online. To investigate this further, the combined nitrogen levels were determined in the filtered media taken from all time points, and shown to be approximately 200mg L−1 NO3 −-N for both high and low Ca2+ media.

Fig. 4.

Box and whisker plot showing the heterocyst frequency rates (A) and representative bright field and fluorescence micrographs of Anabaena sp. PCC 7120 shifted to high Ca2+ (0.25mM to 1mM; B) or to low Ca2+ (1mM to 0.25mM; C) and corresponding controls (A only) 2 days post-shift. Arrows indicate nitrogen-fixing heterocysts.

Box and whisker plot showing the heterocyst frequency rates (A) and representative bright field and fluorescence micrographs of n class="Species">Anabaena sp. PCC 7120 shifted to high Ca2+ (0.25mM to 1mM; B) or to low Ca2+ (1mM to 0.25mM; C) and corresponding controls (A only) 2 days post-shift. Arrows indicate nitrogen-fixing heterocysts.

Discussion

External calcium affects the gene expression and primary metabolism of Anabaena

The role of Ca2+ ions in signal transduction is widely recognized in eukaryotic organisms (Clapham, 1995) and has increasingly become a focus of study in prokaryon class="Chemical">tes in recent years (Onek and Smith, 1992; Norris ; Michiels ; Dominguez ). The maintenance of very low basal intracellular Ca2+ concentrations by active efflux from the cell and through the activity of Ca2+-binding proteins (Gangola and Rosen, 1987; Carafoli ; Batistic and Kudla, 2012; Stael ; Dominguez ) allows increases in cytosolic Ca2+ to have a potent signalling effect. The intracellular Ca2+ concentration of the filamentous cyanobacterium Anabaena was shown to rise transiently as a result of increased external Ca2+, as well as in response to temperature and osmotic stress (Torrecilla ). However, the major signalling role of Ca2+ is associated with the differentiation of vegetative cells into nitrogen-fixing heterocyst cells under nitrogen-limited conditions, through the activity of several regulator proteins that are sensitive to the metabolic status of the cell (Torrecilla ; Zhao ; Shi ; see Kumar for a review). In this work, a 4-fold increase in external Ca2+ concentration caused rapid and strong up-regulation of the HCO3 – transporters Cmp (also called BCT1), Bic and Sbt, as well as the Mrp operon that is involved in supplying Na+ to the latter two HCO3 − symporters. These transporters are components of the carbon-concentrating mechanism (CCM) that also includes CO2 transport machinery in the plasma and thylakoid membranes (Shibata ; Badger ; Badger and Price, 2003). The opposite transcription response of HCO3 − transporters occurred under a 4-fold decrease in Ca2+. In the unicellular cyanobacterium Synechocystis sp. PCC 7002, Cmp and Sbt are expressed in response to limiting Ci conditions (Woodger ); however, cultures used in the current study were grown under 3% CO2 and did not experience Ci limitation. Neither was this caused by any change in media pH, which measured 7.5–8.0 throughout the experiment and was identical in all cultures. Low Ci-induced expression of the Cmp operon in Anabaena is controlled by the LysR-type regulator All0862, which is itself up-regulated in response to low nitrogen (Lopez-Igual ). In this work, the 2FC up-regulated expression of all0862 observed in high Ca2+ may explain up-regulation of the Cmp transporter, although these cultures were not under nitrogen deprivation (discussed below). Notably, translocation of extracellular HCO3 − by the cmpA subunit requires co-binding of Ca2+ and HCO3 − ions to cmpA (Koropatkin ). It has not been established whether these Ca2+ translocation cofactors are released into the cell; however, in this way increases in [Ca2+]e could rapidly translate to an internal signal through increasing the intracellular concentration of Cmp-transported HCO3 −, rather than directly through increased [Ca2+]i. A transient rise in [Ca2+]i in response to increased [Ca2+]e was shown to be rapidly reversed in Anabaena (Torrecilla ), whereas an increase in HCO3 − could induce a more long-term signal through metabolic intermediates such as 2-oxoglutarate (2-OG). Carbohydrate measurements in this study showed no difference between high and low Ca2+ that would reflect the up- and down-regulation of HCO3 − transporters. Rapid induction of several photosynthetic genes under high Ca2+ may be connected with the increased uptake of HCO3 −, which is converted to CO2 for photo-assimilation (reviewed in Burnap ). However, analysis of O2 evolution and PSII yield in this work did not provide evidence of an increased rate of photosynthesis under high Ca2+.

The concentration of Ca2+ demonstrated a short-term effect on the expression of the n class="Chemical">nirA operon that encodes the nitrate assimilation machinery, which was induced in low Ca2+ and repressed in high Ca2+. This correlates with the higher relative protein composition of cells shifted to low Ca2+ (Fig. 2B), and was the opposite trend to that observed for HCO3 − transport and photosynthesis genes. Expression of the nirA operon requires a lack of ammonium and is increased by both nitrate levels (indicated by extracellular or intracellularly generated nitrite) and by 2-OG, a metabolite which represents cellular C:N levels. These indicators respectively activate the ntcB and ntcA transcriptional activators of nirA (Ohashi, 2011). NtcA, considered the global nitrogen regulator of cyanobacteria (Herrero ), is crucial for the expression of the nirA operon across all cyanobacteria (Maeda ). NtcA activity is also increased by PipX, which is sequestered by the signalling protein PII under low 2-OG (Espinosa ). Interestingly, modification of the PII protein was suggested to be influenced by Ca2+ (Zhao ), though evidence for this is lacking. PII is involved in cyanophycin distribution between vegetative cells and heterocysts (Laurent ) and the up-regulation of cyanophycinase (cphB2) was observed under high Ca2+ conditions in this study (Table 1). Thus it appears likely that Ca2+ affects the sensing of intracellular nitrogen availability, possibly through interaction with PII, resulting in the differential regulation of the nirA operon.

Anabaena heterocyst differentiation is not affected by increased extracellular calcium under nitrate-replete conditions

It is interesting that the heterocyst-specific nitrogen fixing (nif) genes are up-regulated by high n class="Chemical">Ca2+ in media replete with nitrate (Fig. 1, Cluster 5), whilst the heterocyst frequency did not differ significantly between high and low Ca2+ conditions. In Anabaena sp. PCC 7120, nitrogenase is protected from oxygen-induced inactivation by being spatially segregated in specialized heterocyst cells where nif expression has been reported to occur late during heterocyst development, at 12–24h after nitrogen deprivation (Elhai and Wolk, 1990; Golden ; Herrero ). Previous work has demonstrated that intracellular Ca2+ is directly involved in heterocyst differentiation (Zhao ; Shi ), but this has generally been studied under conditions of combined nitrogen deprivation. Under such conditions the levels of 2-OG are increased, and NtcA is activated and binds to the promoter of the cyanobacterial Ca2+-binding protein (CcbP), decreasing ccbP expression and increasing free [Ca2+]i. Free [Ca2+]i is also increased in heterocysts through the degradation of CcbP by HetR, the master regulator of heterocyst development. Collectively, the [Ca2+]i of heterocyst cells reaches concentrations severalfold higher than that of vegetative cells (Zhao ; Shi ). Unlike the effect on nif expression, significant effects of [Ca2+]e were not detected in this study for HetR or NtcA expression, which have been reported to occur in the early and intermediate stages of heterocyst development (Herrero ). Nor was a Ca2+-mediated differential expression of PatS or HetN observed, both of which are involved in the negative regulation of heterocyst differentiation (Orozco ). However, the expression level of these genes does not necessarily represent their activity. On the other hand, the effects of high [Ca2+]e in the nitrate-replete conditions of the current work appear to act only on specific functional genes, including the heterocyst glycolipid synthase (hgl) gene cluster (Table 2), coding for a glycolipid layer deposited between the outer membrane and polysaccharide layer and required for preventing diffusion of oxygen into the heterocyst (Fan ). This may indicate the structural and functional importance of heterocysts in maintaining a high [Ca2+]i once differentiation has actually occurred.

Whilst nif and hgl genes demonstrated long term responses of up-regulation in high Ca2+ and down-regulation in low n class="Chemical">Ca2+, sigma factors sigB2 and sigG, which are both involved in, but not individually required for, heterocyst development (Brahamsha and Haselkorn, 1992; Khudyakov and Golden, 2001; Ehira and Miyazaki, 2015), demonstrated opposite long-term responses to Ca2+. Additionally, nrra, another gene involved in early heterocyst development (Ehira and Ohmori, 2006), was found to be strongly (but not significantly; FDR>0.05) down-regulated in high Ca2+.

Ca2+ influences C:N balance in Anabaena

This study has found that changes in Ca2+ concentrations lead to opposite regulation of C and N assimilation pathways that nonetheless do not lead to major changes in cellular biochemistry or development. This suggests that n class="Chemical">Ca2+ signalling in Anabaena may have a role in fine-tuning the C:N balance of cells for optimal growth under fluctuating environmental conditions. One possible mechanism for this may be through modulating the activity of a transcriptional regulator. Given the central role of the NtcA regulator, which generally activates, but can also repress, genes involved in management of the C:N balance in Anabaena (Herrero ), the expression of approximately 40 known NtcA-regulated genes (Picossi ) was tested using RNAseq data from the current study. This analysis revealed no clear correlation between NtcA regulation and Ca2+ treatment, although, several of these genes were significantly differentially expressed under shifted [Ca2+]e (see Supplementary Figure S1 at JXB online).

Considering the tight regulation of [Ca2+]i in n class="Species">Anabaena (Torrecilla ), the up-regulation of nitrogen metabolism- and heterocyst-related genes as a long-term response to increased [Ca2+]e may most likely be a metabolic adjustment in response to the up-regulation of carbon metabolism in the short term. In this case, the observed increase in nif expression may be induced through increased 2-OG levels as a consequence of increase of Cmp-transported HCO3 −. Nonetheless, the possibility that even trace quantities of [Ca2+]i can influence gene expression through Ca2+-binding transcription regulators, despite strict [Ca2+]i maintenance, should not be dismissed.

Perspective

This work shows that Ca2+ operan class="Chemical">tes as a secondary messenger that affects the primary cellular metabolism of Anabaena under conditions replete in both combined-nitrogen and inorganic carbon. Transcriptomics data revealed distinctly and strikingly opposite trends in regulation of nitrogen-related and carbon-related processes in response to Ca2+. Given the elaborate and sensitive network of regulation of C:N balance in cyanobacteria, it is tempting to speculate that Ca2+ has a prominent role in the initial stages of C:N balance adjustment, through regulation of cellular nutrient levels, photosynthesis and/or gene activation. This short-term effect of Ca2+ likely prompts a cascade of gene expression to rebalance the C:N homeostasis of the cell. However, heterocyst differentiation and nitrogen fixation in Anabaena appear to require convergence of multiple signalling pathways in addition to Ca2+, including those instigated by genuine combined nitrogen deficiency. It will be most valuable to establish the identities of the cellular factors responsible for transforming the Ca2+ signal to gene expression in Anabaena, especially metabolic intermediates such as 2-OG and protein receptors of Ca2+.

Supplementary data

Supplementary data are available at JXB online.

Figure S1. Clustered heatmap showing the absolute expression of genes shown to interact with the transcriptional regulator NtcA (Picossi ) in response to changes in [Ca2+] in the expression data of the current study.

Table S1. Heterocyst counts in Ca2+ shift experiments.