OUP user menu

NaCl enhances cellular cAMP and upregulates genes related to heterocyst development in the cyanobacterium, Anabaena sp. strain PCC 7120

Masahiko Imashimizu , Hidehisa Yoshimura , Hiroshi Katoh , Shigeki Ehira , Masayuki Ohmori
DOI: http://dx.doi.org/10.1016/j.femsle.2005.08.035 97-103 First published online: 1 November 2005


Cellular cAMP was rapidly increased in the nitrogen-fixing cyanobacterium, Anabaena sp. PCC 7120, by the addition of 200 mM NaCl to the culture medium. Other alkaline-metal chlorides such as KCl or LiCl caused a lesser increase. The increase in cellular cAMP was transient and diminished when an adenylate cyclase, CyaC, which contains the conserved domains of the bacterial two-component regulatory system, was disrupted. DNA microarray analysis showed that expression of a gene cluster containing all5347 and alr5351 (hglE) was upregulated by NaCl in the wild-type strain but not in the cyaC mutant. Primer extension analysis indicated that transcription levels of all5347 and hglE were rapidly increased in response to the NaCl addition, and that these genes have NaCl-dependent transcription start sites. It was concluded that NaCl induced expression of genes related to heterocyst envelope formation in this cyanobacterium, possibly via a CyaC–cAMP signal transduction system.

  • Cyanobacteria
  • Anabaena
  • cAMP
  • NaCl
  • Signal transduction
  • DNA microarray

1 Introduction

Salt stress induces many biological processes that assist organisms to survive in environments with high salt concentrations. The mechanisms responsible for tolerance to salt stress seem to be activated when cells are exposed to high salt concentrations. It has been reported that in Synechocystis sp. PCC 6803, the addition of NaCl to the medium enhanced expression of genes related to salt tolerance, such as those that express glucosylglycerol-phosphate synthase [ [, []. There must be a signal transduction pathway for environmental salt stress. However, the regulatory mechanism of gene expression has not yet been elucidated.

In the nitrogen-fixing filamentous cyanobacterium, Anabaena cylindrica, cellular cAMP levels exhibit dramatic changes in response to various environmental signals such as nitrogen depletion [3], light–dark, aerobic–anaerobic [4] and low pH–high pH [5]. cAMP is a ubiquitous signalling molecule well known as an intracellular second messenger that alters enzyme activities via phosphorylation in animal cells or regulates gene expression by binding to cAMP receptor protein (CRP) in bacteria [ [, []. Recently, we have found that blue light stimulates cell motility via a CRP-mediated cAMP signal-transduction system in the unicellular cyanobacterium, Synechocystis sp. PCC 6803 [ [, [].

In Anabaena sp. strain PCC 7120 (hereafter, Anabaena PCC 7120), cAMP is synthesized from ATP by adenylate cyclases with various domain structures (CyaA, B1, B2, C, D and E) [ [0, [1]. Of these, CyaC has a unique structure composed of the domains of a bacterial two-component regulatory system [ [2– [4], and is considered to play a key role in the cAMP signalling pathway coupled with receptors for extracellular signals [ [0, [5].

In this study, we found that NaCl increased the cellular cAMP in Anabaena PCC 7120. The effect of NaCl on gene expression was also determined using a DNA microarray. A comparison of genome-wide expression changes between wild-type and cyaC mutant suggested that the expression of genes for the glycolipid layer of the heterocyst envelope is enhanced via a CyaC–cAMP signal transduction system.

2 Materials and methods

2.1 Organisms and growth conditions

Anabaena PCC 7120 and its derivatives bearing inactivated cyaA, cyaB1, cyaB2, cyaC or cyaD genes with a spectinomycin and streptomycin-resistant cassette [10] were grown in 50 ml of modified Detmer's medium (MDM) [16] at 30 °C. Air containing 1% CO2 was bubbled through the medium under continuous illumination at 50 μE m−2 s−1. The culture medium contained 1.7 mM NaCl and 1.4 mM K2HPO4 as alkaline metal cation sources. Cells at the late log growth phase (optical density at 750 nm was about 0.8) were harvested and washed twice with nitrogen-free MDM (MDM0). The cell suspension was diluted to an optical density of 0.2 at 750 nm with MDM0 and incubated for one day under the same culture conditions before use.

2.2 Determination of cellular cAMP level

Cellular cAMP concentrations were determined according to the method of Ohmori [5]. One-millilitre aliquots of cell cultures were mixed with trichloroacetic acid to a final concentration of 5% (w/v) and centrifuged. Supernatants of the mixtures were collected, washed with 3 ml of ethyl ether six times to remove trichloroacetic acid and were then lyophilized. The amount of cAMP in each lyophilized sample was determined using an enzyme immunoassay kit (EIA system; Amersham Biosciences, NJ, USA) according to the manufacturer's protocol. The amount of cAMP in the medium was measured and found to be less than 5 nmol/mg chlorophyll, before and after the experiment. The change in the cAMP level of the lyophilized extract was thus considered to reflect the change in cellular cAMP level.

2.3 DNA microarray analysis

Cells were harvested from 25 ml of culture by centrifugation at 4000g for 5 min at 4 °C, frozen immediately in liquid nitrogen and stored at −80 °C until use. Total RNA was isolated by the hot-phenol method of Mohamed and Jansson [17]. The crude total RNA was treated with 5 U of DNase I (Takara, Shiga, Japan) at 37 °C for 1 h to remove DNA contamination. The microarray of Anabaena PCC 7120 consisted of 2407 DNA segments of approximately 3 kb in size corresponding to about 90% of the entire genomic sequence, and each spot was duplicated [18]. Synthesis of Cy3-labelled and Cy5-labelled cDNA probes, hybridization with the cDNA probes and washing of the microarrays were performed as described previously [19]. For quantification of the fluorescent intensity of spots, microarrays were scanned with a laser fluorescent scanner (Scanarray 4000; GSI Lumonics, MA, USA), and raw data obtained with the optimum photo-multiplier gain were then analysed with Quantarray software (GSI Lumonics). The local background signal was subtracted, and the signals were normalized using the ratio of the total signal intensity, except for rRNA signals. Changes in the transcript, relative to the total amount of mRNA, were then calculated.

2.4 RT-PCR

One-microgram aliquots from the total RNA, isolated under the same culture conditions as the DNA microarray analysis, were used as the RNA template. Reverse transcription and the following PCR were performed with an RNA-PCR kit (Takara). The primers for RT-PCR are shown in Table 1. Each pair of primers was designed to produce PCR products approximately 300 bp in length. In the PCR amplification, 25 cycles, consisting of 30 s at 94 °C, 30 s at 57 °C and 30 s at 72 °C, were applied. The whole process was performed three times using total RNAs purified from independent cell cultures to verify reproducibility.

View this table:

Primer sequences used in this study

Primer extension probes

2.5 Primer extension analysis

The amount and 5′-end of the mRNAs for all5347 and alr5351 (hglE) were analysed by the primer extension method as described previously [20]. Briefly, 30 μg of total RNA was annealed with 100 ng of 32P-labelled primers 1–5 (Table 1). Primers 1, 2 and 4, whose 5′-ends were located at +47, −622 and −905 with respect to the translation start of alr5351, respectively, were used for the analysis of alr5351. Primers 3 and 5, whose 5′-ends were located at +49 and −300 with respect to the translation start of all5347, respectively, were used for the analysis of all5347. The annealed mixtures were extended with 200 U of AMV Reverse Transcriptase XL (Takara) for 1 h at 42 °C. The products were electrophoresed on a 6% polyacrylamide gel together with a DNA size marker or sequencing ladder.

3 Results and discussion

3.1 Increase of cAMP concentration upon addition of NaCl

When 200 mM NaCl was added to the cells of Anabaena PCC 7120 under nitrogen-fixing conditions, the cellular cAMP level increased more than 4.5-fold within 3 min and then decreased (Fig. 1(a)). The rapid cAMP increase and subsequent decrease within 3 min suggests that cAMP functioned as a signalling molecule. A significantly smaller increase in cAMP levels was observed following the addition of 200 mM KCl (2-fold) and LiCl (1.5-fold), indicating that NaCl was the most effective of the alkaline-metal chlorides at increasing cellular cAMP. The addition of 200 mM sorbitol as an organic osmoticum did not increase the cellular cAMP level, suggesting that salt stress but not osmotic stress increased cellular cAMP. It is postulated that the change in NaCl concentration was sensed by a specific signal-receptor histidine kinase, such as Hik33, which is located in the cytoplasmic membrane of Synechocystis PCC 6803 [21], and that the signal was transmitted to an adenylate cyclase with which cAMP was formed.


(a) Effect of ionic stimuli on cellular cAMP levels. Wild-type cells cultured for 24 h in nitrogen-free MDM were used. NaCl (filled circle), KCl (filled triangle), LiCl (filled square) or sorbitol (filled rhombus) at a final concentration of 200 mM, and culture medium as a control (open circle) were added at time zero. (b) Effect of NaCl on cellular cAMP levels of cya-inactivated mutants. cyaA (filled circle), cyaB1 (filled triangle), cyaB2 (filled square), cyaC (open circle), and cyaD (open triangle) mutants cultured for 24 h in nitrogen-free MDM were used with 200 mM NaCl added at time zero. Data are expressed as an average of the results of two independent experiments.

A line of adenylate cyclase (cya)-inactivated mutants had been constructed by interposition of cya genes with an Spr–Smr cassette [10]. Using these mutants, we determined the change in cellular cAMP levels after the addition of NaCl to the medium. Fig. 1(b) shows that the cellular cAMP level of the cyaC mutant did not increase when 200 mM NaCl was added. In cyaA, cyaB1 and cyaD mutants, the cAMP levels were increased by the addition of NaCl even though the levels were less than the wild-type. In the cyaB2 mutant, a smaller increase in the cellular cAMP level than in the wild-type was observed. These results show that CyaC contributes principally to the increase of cAMP by the addition of NaCl, and that CyaB2 may be partially involved in this response.

3.2 Microarray analysis of NaCl-dependent gene expression in wild-type and cyaC mutant

The Anabaena DNA microarray consists of 2,407 contigs of 3–4 kb genomic segments, each containing 1–8 open reading frames (ORFs) [18]. Table 2 shows a list of ORFs that were expressed more than 3-fold higher in the wild-type than in the cyaC mutant 10 min after NaCl addition. These genes, such as the nif genes [ [2– [4], cox genes [25], hglEDC and hetMN[ [6– [8] are mainly related to nitrogen fixation and heterocyst differentiation. In addition, several DNA segments containing putative ABC transporters of a devBCA homologue and putative glycosyl transferases, which could be involved in the biosynthesis of glycolipid or polysaccharide, were upregulated. devBCA is known to be involved in the formation of the heterocyst-specific glycolipid layer in Anabaena PCC 7120 [29]. It has been reported that the 29 kb region is upregulated in the form of an “expressed island” on the Anabaena PCC 7120 chromosome when cells are exposed to nitrogen depletion for 24 h [30].

View this table:

DNA segments whose upregulation in response to the presence of 200 mM NaCl depends upon CyaC

Location on chromosomeaWT/cyaCbORF
1870741901243.6 ± 1.5all0177:flavoprotein, all0178:flavoprotein
2856632882923.3 ± 0.9alr0267:unknown protein, all0268:hypothetical protein
4336574368793.8 ± 1.3alr0370:unknown protein, all0371:probable integral membrane efflux protein, all0372:transcriptional regulator, all0373:serine/threonine protein phosphatase
8980229015813.0 ± 0.8all0775:squalene–hopene–cyclase, alr0776:hypothetical protein, all0777:unknown protein
8997479032747.4 ± 4.1alr0776:hypothetical protein, all0777:unknown protein, all0778:similar to serine protease inhibitor, asl0779:unknown protein
9327859362973.4 ± 1.6all0807:ABC transporter ATP-binding subunit, all0808:ABC transporter membrane spanning subunit, all0809:ABC transporter membrane fusion protein, alr0810:transcriptional regulator
113127611343375.3 ± 2.4alr0970:ABC transporter ATP-binding protein, alr0971:hypothetical protein
169464616975343.5 ± 1.0all1436(nifX):nitrogen fixation protein NifX, all1437(nifN):nitrogenase olybdenum-iron protein NifN, all1438(nifE):nitrogen Fe/Mo cofactor biosynthesis E
171142117150523.2 ± 1.5all1454(nifD):nitrogenase molybdenum-iron protein alpha chain, all1455(nifH):nitrogenase ron protein NifH, all1456(nifU):nitrogen fixation protein NifU
302283030262493.0 ± 0.3alr2515(coxA2):cytochrome c oxidase subunit I, alr2516(coxC2):cytochrome c oxidase subunit III, alr2517:hypothetical protein, alr2518:hypothetical protein
332729633308634.0 ± 1.0alr2729:hypothetical protein, alr2730:hypothetical protein, alr2731(coxB3):cytochrome c oxidase subunit II, alr2732(coxA3):cytochrome c oxidase subunit I
444936944525726.0 ± 3.2alr3689:hypothetical protein, alr3690:hypothetical protein, all3691:serine/threonine kinase with two-component sensor domain
637425263776514.3 ± 2.3all5341:probable glycosyl transferase, all5342:unknown protein, all5343:unknown protein
637655063796444.2 ± 1.3all5343:unknown protein, all5344:hypothetical protein, all5345:oxidoreductase, all5346:ABC-transporter membrane spanning subunit
637884363823994.1 ± 0.9all5345:oxidoreductase, all5346:ABC-transporter membrane spanning subunit, all5347:ABC-transporter membrane fusion protein
638386663866913.7 ± 0.9alr5348:unknown protein, asr5349:unknown protein, asr5350:unknown protein, alr5351(hglE):heterocyst glycolipid synthase
638720063901767.5 ± 0.8alr5351(hglE):heterocyst glycolipid synthase
638924663921415.9 ± 3.4alr5351(hglE):heterocyst glycolipid synthase, alr5352:unknown protein, alr5353:hypothetical protein
639428163976164.1 ± 2.1alr5354(hglD):heterocyst glycolipid synthase HglD, alr5355(hglC):heterocyst glycolipid synthase HglC
639725363999053.5 ± 0.4alr5355(hglC):heterocyst glycolipid synthase HglC, alr5356:hypothetical protein
639888264024543.0 ± 0.4alr5356:hypothetical protein, alr5357(hetM):polyketide synthase HetM, alr5358(hetN):ketoacyl reductase HetN
  • aLocation was referred for CyanoBase (http://www.kazusa.or.jp/cyanobase/).

  • bNormalized signal intensity ratio of wild-type to cyaC mutant. The data are expressed as ±SD of values from four measurements (two independent experiments of duplicated spots to a glass slide).

To examine changes in the expression of four ORFs, all5347, alr5351 (hglE), alr5353, and alr5355 (hglC) in the 29 kb region, levels of mRNA were individually analysed by RT-PCR (Fig. 2). It was found that the levels of mRNA of all these ORFs were substantially increased in the wild-type after the addition of NaCl, but smaller increases were observed for the cyaC mutant. Primer extension analysis confirmed the presence of the NaCl-dependent specific 5′-ends, R4 and L2 (likely transcription start sites (TSSs)), for hglE and all5347 (Fig. 3). This suggests that the CyaC protein and activity are required for the induction of gene expression by NaCl. Simultaneously, a number of genes (particularly those involved with photosynthesis, energy production and ribosomal protein) were downregulated, but this study was restricted to upregulated genes associated with the increase in cAMP.


RT-PCR analysis of several genes within a 29 kb gene cluster. mRNA levels of all5347, alr5351 (hglE), alr5353 and alr5355 (hglC) at 0, 10, 30, 60 and 180 min after the addition of 200 mM NaCl were analysed by RT-PCR. rnpB was used as an internal control. The image is displayed in inverted tone for clarity of intensity.


(a) Primer extension analysis of TSSs of all5347 and alr5351 (hglE). Results of primer extension reactions with RNAs isolated from cells grown in nitrogen-free MDM (−) and cells subjected to 200 mM NaCl for 10 min (+). Lanes 1–3 show reactions with the corresponding primer number shown in Table 1. Lane M denotes the DNA size marker. Putative TSSs are indicated by arrowheads. (b) Detailed analysis of TSSs, R-4 and L-2. +/− NaCl indicate results of primer extension reactions with RNA isolated from wild-type cells before and 10 min after NaCl addition. TSSs (R-4 and L-2) are indicated by arrows. (c) Schematic diagram of TSSs of all5347 and alr5351 (hglE). Primers 1–5 used in primer extension analysis are indicated by arrows. Distances from the putative translation start sites are indicated in parentheses for R-4 and L-2.

3.3 Determination of promoter regions of all5347 and alr5351 (hglE)

The TSSs for all5347 and alr5351 (hglE) were determined by primer extension analysis. The position of each TSS was broadly surveyed with primers 1–3 (Table 1) in the presence and absence of NaCl. The analysis of the promoter region for alr5351 (hglE) and all5347 using primers 1–3 showed that transcription with all TSSs (R-1, R-2, R-3, R-4, L-1 and L-2) was enhanced in the presence of NaCl (Fig. 3(a)). Since the predominant activation by NaCl was observed in R-4 and L-2 in the respective ORFs, we then determined the precise TSSs using primers 4 and 5 (Fig. 3(b)). The location of R-4 was −986 from the putative translation start site, while those of L-2a,b,c and d were −434, −448, −456 and −493, respectively (Fig. 3(b) and (c)). The mRNA levels of R-4 and four L-2s were significantly increased in the presence of NaCl.

Previously, we found that Anabaena PCC 7120 has at least two CRPs, AncrpA and AncrpB, with different binding capacities for cAMP [31]. These CRPs are possible candidates for a transcription factor in the promoter regions of all5347 and alr5351 (hglE), although the well-known CRP-binding sequences have not yet been detected in the promoter regions.

Further studies are needed to elucidate the actual mechanisms, which include those involved in sensing an increase in environmental NaCl concentrations, CyaC-dependent cAMP signalling and upregulation of the 29 kb gene cluster. However, our results suggest a possible link between NaCl and the expression of genes involved in heterocyst-envelope formation mediated by a cAMP-signalling cascade in Anabaena PCC 7120.


We thank Drs. M. Katayama and S. Okamoto, and Mr. T. Suzuki of the University of Tokyo for their helpful discussions. This work was supported by a Grant-in-aid for General Scientific Research (12206002) from the Ministry of Education, Culture, Sports, Science and Technology of Japan, and partly by a grant from the Research Institute for Environment/the Ministry of Economy, Trade and Industry.


  1. [1].
  2. [2].
  3. [3].
  4. [4].
  5. [5].
  6. [6].
  7. [7].
  8. [8].
  9. [9].
  10. [10].
  11. [11].
  12. [12].
  13. [13].
  14. [14].
  15. [15].
  16. [16].
  17. [17].
  18. [18].
  19. [19].
  20. [20].
  21. [21].
  22. [22].
  23. [23].
  24. [24].
  25. [25].
  26. [26].
  27. [27].
  28. [28].
  29. [29].
  30. [30].
  31. [31].
View Abstract