The kaiABC gene cluster is important for maintaining circadian rhythms in the cyanobacterium Synechococcus PCC 7942. An extensive PCR-based survey of phylogenetically diverse cyanobacteria was conducted using degenerate primers designed to identify the presence of the kaiC gene. Hybridization and sequence analyses showed that the observed amplification products had a high degree of similarity to kaiC. Forty cyanobacterial strains possessed kaiC related sequences, suggesting that a clock system is universal among cyanobacteria.
Circadian rhythms are biological programs that follow an approximate 24 h cycle even in the absence of external signals. They can be reset by light and dark cues, and are temperature-compensated. In the cyanobacterium Synechococcus, the circadian clock has a global effect on cell metabolism . The Synechococcus clock regulates cell division, nitrogen fixation, photosynthesis, amino acid uptake, carbohydrate synthesis, and respiration , and is essential for generating and maintaining circadian oscillations in Synechococcus PCC 7942 .
Cyanobacteria may benefit from having an internal clock on a similar light/dark cycle as the environment under competitive circumstances. In non-competitive situations cyanobacterial growth is not inhibited by a circadian rhythm out of phase with the environment. When two strains are in competition with one another, the strain with the internal clock more similar to the present light/dark cycle will survive [4,5]. Whether a clock system is vital to the survival of all cyanobacteria living in competitive habitats has yet to be investigated. It has been suggested that circadian cycles allow for cellular processes to occur in a temporal organization most beneficial to the organism. Some cyanobacteria display a 24-h photosynthesis rhythm that is 180° out of phase with the rhythm of nitrogen fixation. Limiting photosynthesis (an oxygen-producing process) to daylight hours allows the oxygen-sensitive enzyme nitrogenase (essential for nitrogen fixation) to function at night.
The kaiABC system of Synechococccus PCC 7942 encodes a circadian oscillator and has been extensively characterized (for review see Golden, 1999 ). The luxA reporter gene has been employed to monitor kaiABC regulation in vivo. KaiA is a positive regulator of the kaiBC promoter, whereas overexpression of kaiC inhibits transcription at its own promoter [1,3,4,7–9]. The negative feedback mechanism of the kaiABC cluster in Synechococcus sp. strain PCC 7942 is very similar to the regulatory systems of the Drosophila, Neurospora and Mus musculus clock genes . The evolution of circadian regulatory systems still remains unclear, and it has yet to be seen whether information learned about the Kai gene system can be applied to the investigation of more complex organisms.
The proteins of the kai system show no sequence similarity to the clock proteins of eukaryotic model organisms . A homologue to kaiABC is present in the genome sequence of Synechocystis[10–12], but analysis of other cyanobacteria has not been described. A variety of cyanobacterial species including representatives of all five sections  were investigated in this study for the presence of the kaiC gene. The results suggest that a kaiC-regulated circadian system is universal to the cyanobacteria.
2Materials and methods
2.1Source of material
The legend of Fig. 2 lists the cyanobacterial strains examined and the sources of materials. A crude DNA preparation, adequate for PCR, was generated from cultures by boiling a small sample of cells and then extracting the lysate with chloroform .
2.2Primers and amplification
PCR amplification was carried out using Taq enzyme and reagents from Gibco BRL Products. Fig. 1 shows the relative position of the primers used. Primers 336, 337 and 338 were based on conserved regions inferred from the alignment of kaiABC and related deduced protein sequences retrieved from public databases: Synechocystis PCC 6803 , Methanobacterium thermoautotrophicum, Pyrococcus horikoshii and Archaeoglobus fulgidus. Primers 488 and 489 were based on data obtained in this study. All primers were used at a final concentration of 1.0 μM. Primer sets 1 and 2 were used in a PCR program of 45 cycles comprised of 94°C for 30 s, 47°C for 50 s, and 72°C for 3 min, in a Robocycler Gradient 96 (Stratagene). Primer set 3 was used in a PCR program of 40 cycles comprised of 92°C for 45 s, 51°C for 45 s, and 72°C for 1 min, in a programmable Thermal Controller (MJ Research). Plasmid pJS1, consisting of Synechococcus PCC 7942 kaiABC sequences ligated into pGEM-T (Promega), was used as a positive control template for all PCR reactions. The 3.3-kb insert DNA (kaiABC) was obtained from Synechococcus genomic DNA by PCR using primers described by Ishiura et al. .
2.3Analysis of amplicons by hybridization
Amplicons obtained by PCR were analyzed by agarose gel electrophoresis. For Southern hybridization studies, approximately 5–10 ng of the PCR products generated using primer set 3 were resolved on a 1% gel and transferred to Hybond-N+ membrane (Amersham Pharmacia Biotech). The blot was probed with a 290-bp EcoRV/BspHI fragment of kaiC (Fig. 1) at 45°C for 8 h in hybridization buffer (7% SDS/1% BSA/0.25 M Na2HPO4/1 mM EDTA) and washed twice at 45°C in 2×SSC/0.1% SDS. The processed blot was placed on film for 1.5 h at −80°C for autoradiography.
Selected amplicons were either sequenced directly, using the PCR primers, or were cloned in pGEM-T or pGEM-T Easy (Promega), or PCR2.1-TOPO (Invitrogen) and sequenced from the plasmids. When necessary, PCR products were purified using the QIAquick PCR Purification kit, or were recovered from agarose using the QIAquick Gel Extraction kit (Qiagen). All sequencing was done using the ABI PRISM Dye Terminator Cycle Sequencing Ready Reaction DNA Sequencing kit (PE Applied Biosystems) and the 373 DNA sequencer STRETCH (Applied Biosystems). The results were compared to sequences in public databases using the BLASTX program [18,19]. Deduced protein sequences of putative kaiC homologues were aligned using CLUSTAL W . All sequence data obtained in this study were deposited in GenBank (accessions AF222599–AF222605, AF239752–AF239755).
3Results and discussion
A survey of cyanobacteria was carried out using degenerate primers designed to identify genes of the kai cluster by PCR. Primer sets 1 and 2 were based on a small set of available related sequences, including two from cyanobacteria (Synechococcus PCC 7942 and Synechocystis PCC 6803). Amplicons of the expected sizes (1.2 and 0.6 kb) were observed with at least one primer set for 11 of 15 strains tested; additional, smaller amplicons were also observed in some cases, but these were always less abundant and were considered spurious products (results not shown). Preliminary partial sequences for seven of the strains (obtained by directly sequencing the PCR products) indicated that the amplicons were very similar to each other, as well as to kaiC in Synechococcus PCC 7942 and Synechocystis PCC 6803. These results were used to design a third set of primers that would better reflect sequences conserved in cyanobacteria.
Primer set 3 was developed to amplify a 0.65-kb fragment of the kaiC gene, and was very effective. An amplicon of the expected size was obtained from the DNA of all 40 strains tested. The amplicons from 36 strains were tested by Southern hybridization with a fragment of the the cloned Synechococcus PCC 7942 kaiABC cluster (pJS1) that is internal to the amplicon generated with primer set 3 (Fig. 1). All 36 amplicons hybridized to the probe, indicating that kaiC homologues exist in a wide distribution of cyanobacteria (Fig. 2).
Amplicons from four additional strains, as well as representatives from the first 36 strains (selected to represent all sections in Fig. 2, legend), were cloned and sequenced. Fig. 3 shows the deduced translation products of primer set 1 PCR products from two strains (PCC 73104 and PCC 7417), and primer set 3 PCR products generated from nine strains (see Fig. 3, legend) aligned with the equivalent portion of the Synechococcus PCC 7942 KaiC protein, and the KaiC homologue of Synechocystis PCC 6803. The aligned sequences shared 64% identity in the 198-residue region shown. It has been recently shown in PCC 7942 that ATP-binding activity of the P loop motif (residues 15–22) is necessary for circadian oscillation . This motif is conserved (100% identity) in all KaiC sequences obtained in this study. The primer set 1 products also included a region homologous to the expected portion of KaiB (not shown), indicating that the order of kaiB and kaiC in these cyanobacteria is the same as that of Synechococcus PCC 7942.
These results suggest that cyanobacteria as a group possess the genetic capability to establish a circadian rhythm. The kaiC gene sequences of 13 cyanobacteria are highly similar, and PCR results indicate that a wide range of cyanobacteria possess a kaiC homologue. Perhaps the advantages offered to a cyanobacterium by the kai gene cluster [4,5] are so great that all cyanobacterial species require a clock system to survive in a competitive environment. The exact role circadian regulation plays in the developmental cycle of these photoactive bacteria is not known, however; and a number of other questions remains to be addressed. Do all cyanobacteria have the full complement of clock genes? Do symbiotic cyanobacteria (in lichens, for example) have clock genes and, if so, what are their roles? Investigaton into the organization of kai genes in diverse cyanobacteria, and into the presence of kai homologues in other micro-organisms, may also lead to an understanding of how a circadian regulatory system evolved.
We thank Julian Davies for suggesting this project, S. Turner and D. Mazel for their generosity in providing materials, and S. Turner for commenting on an earlier draft of this manuscript.