Correspondence: Hirofumi Yoshikawa, Department of Bioscience, Tokyo University of Agriculture, 1-1-1 Sakuragaoka, Setagaya-ku, Tokyo 156-8502, Japan. Tel.: + 81 3 54772758; fax: + 81 3 54772668; e-mail: firstname.lastname@example.org
The freshwater cyanobacterium Synechococcus elongatus PCC 7942 exhibits light-dependent growth. Although it has been reported that DNA replication also depends on light irradiation in S. elongatus 7942, the involvement of the light in the regulation of DNA replication remains unclear. To elucidate the regulatory pathway of DNA replication by light, we studied the effect of several inhibitors, including two electron transport inhibitors, 3-(3,4-dichlorophenyl)-1,1-dimethylurea (DCMU) and 2,5-dibromo-3-methyl-6-isopropyl-p-benzoquinone (DBMIB), on DNA replication in S. elongatus 7942. DCMU inhibited only DNA replication initiation, whereas DBMIB blocked both the initiation and progression of DNA replication. These results suggest that DNA replication depends on the photosynthetic electron transport activity and initiation and progression of DNA replication are regulated in different ways.
Chromosome replication is the most fundamental and essential process in the cell cycle. Generally, bacterial DNA replication efficiency is coupled with the rate of growth through nutrient-dependent changes (Wang & Levin, 2009). Photosynthetic microorganisms can produce energy and synthesize cellular components via photosynthesis, and their growth depends on photosynthetic products.
Cyanobacteria are prokaryotic microorganisms that have developed an oxygen-producing photosynthetic system similar to that in chloroplasts of higher plants, and they are therefore used as model organisms for phototrophs. The freshwater cyanobacterium Synechococcus elongatus PCC 7942 (hereafter referred to as S. 7942) is an oligoploid organism harbouring multiple genomic copies in each cell, as seen in chloroplasts (Mann & Carr, 1974; Mori et al., 1996; Griese et al., 2011), and it also exhibits light-dependent cell proliferation (Binder & Chisholm, 1990; Mori et al., 1996). In our previous study, we showed that S. 7942 replicates its DNA in a light-dependent manner (Watanabe et al., 2012). Cyanobacteria utilize light in various cellular processes [e.g. production of energy (Bryant, 1994), regulation of gene expression (Asayama, 2006), and cell proliferation (Asato, 2006)]. However, the process or processes by which light affect DNA replication are unknown.
To reveal the factors regulating DNA replication by light, we performed an in vivo DNA replication assay using various kinds of inhibitors. Our results clearly showed that DNA replication depends on the photosynthetic electron transport activity and de novo gene expression of the DNA replication component. Furthermore, we found initiation and progression of DNA replication are regulated in different ways.
Materials and methods
Bacterial strains and growth condition
Synechococcus elongatus PCC 7942TK strain (Watanabe et al., 2012), was grown at 30 °C in 40 mL of BG-11 medium with 40 μg mL−1 of spectinomycin under 30 μmol m−2 s−1 of white light irradiation. To synchronize cell proliferation, we cultured the cells in BG-11 medium until they reached the stationary phase. The cells were then diluted to OD750 = 0.2 with fresh BG-11 medium. After culturing for 18 h in the dark, the culture was transferred to the light condition (time 0) to restart cell growth.
For inhibitor treatments, rifampicin (Rif), chloramphenicol, nalidixic acid (NDX), cefalexin (CEX), 3-(3,4-dichlorophenyl)-1,1-dimethylurea [DCMU, an inhibitor of electron transport between the PSII complex and the plastoquinone (PQ) pool], and 2,5-dibromo-3-methyl-6-isopropyl-p-benzoquinone (DBMIB, an inhibitor of electron transport between PQ and the cytochrome b6f complex) were added to the cultures at final concentrations of 200, 100, 3, 2.5 μg mL−1, 5 and 5 μM, respectively. Each inhibitor was dissolved in organic solvents (Rif and DCMU: dimethyl sulfoxide, Cm, NDX and CEX: ethanol, DBMIB: methanol, respectively).
Immunoblot analysis of 5-bromo-2′-deoxyuridine (BrdU)-labelled DNA
BrdU 1 mM was added to the culture, and BrdU-labelled DNA was extracted and analysed by immunoblotting as previously described (Watanabe et al., 2012). The intensity of immunoblots was quantified using image labtm 2.0 software (Bio-Rad Laboratories, Hercules, CA).
Flow cytometry analysis
Cell preparation and flow cytometry using a FACSCalibur instrument (Becton-Dickinson, Palo Alto, CA) were performed as previously described (Watanabe et al., 2012).
Construction of the strain carrying the FLAG-tagged dnaB gene
To express DnaB fused with three copies of the FLAG tags at the C-terminus, a recombinant DNA fragment was constructed as follows. The genomic regions including two genes (dnaB and Synpcc7942_1326, downstream of dnaB ORF) were amplified from the S. 7942 genomic DNA using the primer sets 5′-GGGCGGATCC GTGCAGGAACTTCGCTTCGA-3′ and 5′-TCACTTGTCATCGTCATCCTTGTAATCGATGTCATGATCTTTATAATCTCCGTCATGGTCTTTGTAGTCACCCGTAGGCGCATG-3′ (FLAG sequence underlined) and 5′-CGGGGCGTAATTTCCAGCGATCGCCCAAAC-3′ and 5′-TCCCGTAGTCGCTCCAGCTC-3′, respectively. The chloramphenicol (Cm)-resistance gene was amplified from pAM990 using the primer sets 5′-TGACAAGTGAAGAATAAATAAATCC-3′ and 5′-TCGCTGGAAATTACGCCCCGCCCTGCCACT-3′. The three amplified fragments were recombined using PCR, and the resulting DNA fragment was used for the transformation of S. 7942. The resulting strain was named DB-F, which expresses the DnaB-FLAG fusion protein from the endogenous dnaB promoter.
Light-dependent DNA replication in S. 7942
Synechococcus elongatus PCC 7942 replicates its DNA under light culture condition (Binder & Chisholm, 1990; Mori et al., 1996). To investigate the effect of light on DNA replication in S. 7942, we first measured BrdU incorporation to monitor DNA replication activity under light and dark culture conditions. To measure de novo DNA synthesis, we used the S. 7942TK strain, which can incorporate BrdU, an analogue of thymidine, into its chromosome (Watanabe et al., 2012). The stationary phase culture grown under continuous light for 10 days was diluted and incubated in the dark for 18 h and then transferred to the light condition to restart cell growth. Aliquots of the culture were withdrawn and divided into two batches at various time points after the culture was transferred to the light condition. One batch was BrdU-labelled under the light condition, and the other was labelled under the dark condition (Fig. 1a). BrdU incorporation gradually increased under the light condition (Fig. 1b and c). When the culture was transferred to the dark condition, the signal of BrdU incorporation strikingly decreased (Fig. 1b and c), indicating that DNA replication in S. 7942 depends on light irradiation.
Effect of the transcriptional inhibitor on DNA replication
Rifampicin prevents the initiation but not the progression of DNA replication in Escherichia coli and Bacillus subtilis (Skarstad et al., 1986). We investigated the effect of Rif on DNA replication in cyanobacteria. As DNA replication is initiated as early as 0.5–1 h after transfer to the light culture condition (Watanabe et al., 2012), BrdU-labelling was started when the culture was transferred to the light condition and either Rif or NDX (a DNA gyrase inhibitor) was added 1 h post-transfer (Fig. 2a). After BrdU-labelling with or without Rif, the cells were harvested and the incorporation of BrdU was measured. In the absence of the inhibitor, the incorporation of BrdU was dependent on the time of incubation (Fig. 2b and c, Control). When Rif was added to the culture 1 h post-transfer, BrdU incorporation increased until 3 h post-transfer and inhibited thereafter, and NDX immediately blocked the BrdU incorporation (Fig. 2b and c). The cell viabilities of Rif- and NDX-treated culture at all time points were comparable to that of the control, indicating that the decrease of BrdU incorporation at those points is not caused by loss of cell viability (Supporting Information, Fig. S1C and Appendix S1). These results suggest that the new round of DNA replication was not initiated in the presence of Rif (i.e. Rif inhibits replication initiation but not progression) and NDX inhibits both replication initiation and progression. In E. coli and B. subtilis, it has been reported that the Rif inhibits initiation of RNA synthesis, whereas DNA synthesis continued until all cells complete replication (Gellert et al., 1977; Lampe & Bott, 1984). Our results are consistent with those observed in E. coli and B. subtilis.
Regulation of DNA replication initiation
We investigated the effects of different inhibitors on DNA replication initiation. Inhibitors were added at the time of transfer to the light condition with BrdU (Fig. 3a), and the replication activity was compared on the basis of BrdU incorporation. BrdU incorporation was inhibited by the addition of NDX but not CEX (a cell division inhibitor) (Fig. 3b), indicating that DNA replication precedes cell division in S. 7942 as well as in E. coli (Wang & Levin, 2009). Both Rif and Cm (a translation inhibitor) inhibited BrdU incorporation after transfer to the light condition (Fig. 3b). In addition, by using flow cytometry (FACS), we assayed the DNA content per cell of the culture 6 h after the inhibitors were added. Compared with the culture before the transfer (Fig. S2, 0 h), about double the amount of DNA was detected in the culture incubated for 6 h under the light condition (Fig. S2, 6 h control), whereas the DNA profiles were not changed when the inhibitors (Rif, Cm, NDX) were added to the culture, except for the addition of CEX (Fig. S2). These results indicate that DNA replication initiation requires de novo gene expression and protein synthesis.
We investigated the expression level of DNA helicase DnaB by constructing a DB-F strain, which expresses the DnaB-FLAG fusion protein from an endogenous promoter. When the culture was transferred to the light condition, the level of DnaB significantly increased. On the other hand, Rif and Cm clearly inhibited the expression of DnaB (Fig. 3c and Appendix S1), indicating that DNA replication initiation is correlated at least with de novo gene expression of DnaB.
We also examined the effects of the electron transport inhibitors DCMU and DBMIB on DNA replication initiation. DCMU and DBMIB inhibited both BrdU incorporation and the increase of DNA contents after transferring to the light condition (Figs S2 and S3B). In addition, DCMU and DBMIB clearly inhibited the expression of DnaB, as did Rif and Cm (Fig. 3c). These results indicate that the DNA replication initiation depends on the photosynthetic electron transport activity, at least partly through de novo gene expression of the DNA replication component.
Regulation of DNA replication progression
We next investigated the effects of inhibitors after DNA replication was initiated (5 and 8 h post-transfer) (Fig. 4a). The incorporation of BrdU was completely inhibited by the addition of NDX, whereas no effect was observed when CEX was added to the culture (Fig. 4b and c). On the other hand, the BrdU signal was detected in the presence of Rif and Cm, although the levels of BrdU incorporation apparently decreased (Fig. 4b and c). Because different effects were observed in the presence of NDX and Rif, we analysed the DNA content per cell under the culturing condition with NDX and Rif. Inhibitors were added 8 h post-transfer, and the cells were harvested 9 h thereafter (i.e. 17 h post-transfer) and analysed using FACS. As shown in Fig. 4d, the copy number of chromosomes in the control culture increased, whereas no change was observed in the FACS profile of the culture treated with NDX (Fig. 4d); this finding indicates that NDX completely inhibits both replication initiation and progression. In contrast, some peaks in the FACS profile appeared 9 h after culturing with Rif, although the chromosome copy numbers did not change. In E. coli and B. subtilis, when Rif are added to the culture, RNA synthesis is inhibited, whereas DNA synthesis continues until all cells complete replication and end up with fully replicated chromosomes. DNA plots of Rif-treated E. coli and B. subtilis cultures show peaks which correspond to the cell ploidy (Skarstad et al., 1986; Seror et al., 1994). Our result was similar to those in E. coli and B. subtilis, suggesting that Rif inhibits the initiation but not the progression of DNA replication in S. 7942. This was also supported by the results shown in Fig. 2.
We observed the difference between the effects of DCMU and DBMIB on DNA replication 5 and 8 h post-transfer. The BrdU signal was detected in the presence of DCMU or Rif, whereas its signal decreased markedly in the presence of DBMIB or NDX (Fig. 4b). In the FACS analysis, some peaks were observed in the culture treated with DCMU, as with Rif, and no change was observed in the culture treated with DBMIB, as with NDX. These results indicate that DNA replication progression depends on a process that is inhibited by DBMIB but not DCMU.
Effect of far-red light on DNA replication
A number of cyanobacterial responses are known to be triggered by change in the redox state of the PQ pool (Mullineaux, 2001). To investigate the effect of redox state of the PQ pool on DNA replication in S. 7942, we performed a BrdU-labelling assay under white, PSII (orange) and PSI (far-red) light conditions. The PQ pool in cyanobacteria becomes relatively oxidized under PSI light condition, then the ratio of PSII/PSI increases as a result of PSI excitation. This light adaptation, called state transition, is triggered by change in the redox state of the PQ pool in cyanobacteria (Mullineaux & Allen, 1990). We observed that under the PSI light condition in S. 7942, the ratio of PSII/PSI complex increased, as judged by the low-temperature chlorophyll fluorescence at 695 and 720 nm (F695 for PSII and F720 for PSI) (Fig. S3C), indicating that the PSI was preferentially excited in S. 7942. The ratio of the PSII/PSI complex under PSII light was comparable to that under white light condition (Fig. S3C). However, a marked difference was not observed in the BrdU incorporation under the white, PSII or PSI light conditions at any time (Fig. S3A and B), suggesting that the redox state of the PQ pool does not affect DNA replication in S. 7942.
In this study, we observed the effects of gene expression and photosynthesis inhibitors on DNA replication initiation and progression. Rif and Cm inhibited DNA replication initiation, but not progression (Figs 3b and 4b), suggesting that gene expression (e.g. RNA synthesis and protein synthesis) is required for DNA replication initiation. In fact, the level of DnaB protein decreased when the stationary culture incubated under the dark condition, and the expression level was increased after transferring to the light condition. This induction of DnaB was clearly inhibited by the addition of Rif and Cm (Fig. 3c).
We also showed that DNA replication depends on photosynthetic electron transport. Furthermore, initiation and progression of DNA replication are regulated differently through the photosynthetic electron transport activity (Fig. 5). In the process of DNA replication initiation, both DCMU and DBMIB, which interrupt photosynthetic electron transport, inhibited DNA replication activity and DnaB expression (Fig. 3). In Synechocystis sp. PCC 6803, it has been reported that photosynthetic electron transport activity affects gene expression (Hihara et al., 2003). This is consistent with our observation that DNA replication initiation is indirectly regulated by the photosynthetic electron transport activity through de novo gene expression (Fig. 5, Pathway 1).
We additionally observed several differences in the effects of DCMU and DBMIB on the process of DNA replication progression. DBMIB clearly inhibits progression of DNA replication, as does NDX (Figs 3 and 4). However, DCMU did not affect the progression, nor does Rif (Figs 3 and 4). As regards the difference between the effects of DCMU and DBMIB on DNA replication, one of the important photosynthetic mechanisms is cyclic electron flow, which comprises only the electron transport chain and PSI (Battchikova et al., 2011). As DCMU inhibits electron flow between the PSII complex and the PQ pool, cyclic electron transport around PSI can supposedly work in the presence of DCMU. In contrast to the effect of DCMU, DBMIB, which inhibits electron flow between the PQ pool and the Cyt b6f complex, completely blocks the electron transport downstream of the Cyt b6f complex, including the cyclic electron transport. Two groups have reported that a pathway downstream of Cyt b6f, which is inhibited by DBMIB but not DCMU, affects light-induced gene expression in Chlamydomonas and Cyanidioschyzon merolae (Shao et al., 2006; Kanesaki et al., 2012). These observations and our results indicate that the integrity of the electron transport downstream of the Cyt b6f complex is required for DNA replication progression (Fig. 5, Pathway 2). In this model, we consider that the redox state of PQ pool does not affect DNA replication directly. In fact, under the PSI light, which preferentially excites PSI, initiation and progression activities of DNA replication were comparable to those under the white and PSII light (Fig. S3A and B), although the PQ pool becomes relatively oxidized under the PSI light condition. Thus, the transition of the PQ state seems to have no effect on DNA replication. Since DBMIB inhibits progression of DNA replication, as does NDX, which inhibits DNA gyrase, these observations suggest that the electron transport downstream of the Cyt b6f complex directly affects DNA replication components (e.g. activity of DNA polymerase).
Consistent with the above discussion, we also showed that DNA replication in S. 7942 is completely dependent on light irradiation. Upon transferring to the dark condition, BrdU incorporation was immediately stopped, even when DNA replication was highly active under the light condition (Fig. 1). In addition, replication in the control culture transferred to the dark condition was lower than that in the Rif-treated culture under the light condition (Fig. 4). These results suggest that not only the initiation step but also the elongation step of DNA replication is significantly depressed under the dark condition. In cyanobacteria, the electron transport between the PQ pool and the Cyt b6f complex is not completely stopped even in the dark condition because the cyanobacterial photosynthetic electron transport chain is partially shared with its respiratory chain. Nevertheless, when the culture transfers to the dark condition, both linear and cyclic electron transport to the Cyt b6f complex are impeded and the electron transport activity downstream of Cyt b6f complex eventually decreases. Therefore, we consider that the electron transport downstream of the Cyt b6f complex under the dark condition is insufficient for the DNA replication progression. However, we cannot rule out the possibility that the activity of enzyme(s) involved in DNA replication is directly controlled by light. Further research is necessary to elucidate the molecular mechanism of DNA replication in cyanobacteria.
This study was supported by the Research fund for the advancement of the Graduate School, Tokyo University of Agriculture (Research fund for PhD candidates). All the authors have approved the manuscript and agree on the submission. There are no conflicts of interest to declare.