Cyanobacteriochrome CcaS regulates phycoerythrin accumulation in Nostoc punctiforme, a group II chromatic adapter

Yuu Hirosea, Rei Narikawab, Mitsunori Katayamac, and Masahiko Ikeuchia,b,1

a Department of Biological Science, Graduate School of Sciences, and bDepartment of Life Sciences (Biology), University of Tokyo, Komaba 3-8-1, Meguro, Tokyo 153-8902, Japan; and cCollege of Industrial Technology, Nihon University, Izumicho 1-2-1, Narashino, Chiba 275-8575, Japan

Edited by J. Clark Lagarias, University of California, Davis, CA, and approved March 26, 2010 (received for review January 6, 2010)

Responding to green and red light, certain cyanobacteria change the composition of their light-harvesting pigments, phycoerythrin (PE) and phycocyanin (PC). Although this phenomenoncomplementary chromatic adaptationis well known, the green lightsensing mechanism for PE accumulation is unclear. The lamentous cyanobacterium Nostoc punctiforme ATCC 29133 (N. punctiforme) regulates PE synthesis in response to green and red light (group II chromatic adaptation). We disrupted the green/red-perceiving histidine-kinase gene (ccaS) or the cognate response regulator gene (ccaR), which are clustered with several PE and PC genes (cpeC-cpcG2-cpeR1 operon) in N. punctiforme. Under green light, wild-type cells accumulated a signicant amount of PE upon induction of cpeC-cpcG2-cpeR1 expression, whereas they accumulated little PE with suppression of cpeC-cpcG2-cpeR1 expression under red light. Under both green and red light, the ccaS mutant constitutively accumulated some PE with constitutively low cpeC-cpcG2-cpeR1 expression, whereas the ccaR mutant accumulated little PE with suppression of cpeC-cpcG2-cpeR1 expression. The results of an electrophoretic mobility shift assay suggest that CcaR binds to the promoter region of cpeC-cpcG2cpeR1, which contains a conserved direct-repeat motif. Taken together, the results suggest that CcaS phosphorylates CcaR under green light and that phosphorylated CcaR then induces cpeCcpcG2-cpeR1 expression, leading to PE accumulation. In contrast, CcaS probably represses cpeC-cpcG2-cpeR1 expression by dephosphorylation of CcaR under red light. We also found that the cpeBcpeA operon is partially regulated by green and red light, suggesting that the green light-induced regulatory protein CpeR1 activates cpeB-cpeA expression together with constitutively induced CpeR2.
complementary chromatic adaptation phytochrome

| phycobilisome | cyanobacteria |

hytochromes are photoreceptors that perceive red and far-red light in plants, bacteria, cyanobacteria, and fungi (1). Recently, unique phytochrome-related photoreceptors have been discovered in cyanobacteria. These receptors bind a linear tetrapyrrole molecule in their cGMP phosphodiesterase/adenylyl cyclase/FhlA (GAF) domain that is distinct in amino acid sequence from the GAF of phytochromes (2). To date, several GAF domains of these photoreceptors have been analyzed and shown to photoconvert reversibly between violet-/yellow (3), blue-/green (46), or green-/red light-absorbing forms (7, 8); this photoconversion contrasts with the red/far-red photoconversion of phytochromes. Mass spectrometry and biochemical studies suggest that their chromophore species and conguration in the GAF domain are highly divergent from those of the phytochromes (5, 711). Because these photoreceptors are found only in cyanobacteria, they have been called cyanobacteriochromes (2). Cyanobacteria use phycobilisome (PBS) as a light-harvesting antenna. Phycobilisome is composed of a rod and a core structure, each of which is composed of pigmented phycobiliproteins and associated linker proteins (12). In certain species, the rod contains two kinds of phycobiliproteinsphycoerythrin (PE) and phycocyanin (PC), which harvest green light (GL) and red light (RL), re-

spectively. It has long been known that many of these species are able to increase their PE/PC ratio under GL and reduce it under RL in a process called complementary chromatic adaptation (1315). There are two groups of cyanobacteria that change the PE/PC ratio in different manners (15, 16). One group regulates only PE synthesis by GL and RL (group II), whereas the other group regulates both PE and PC synthesis (group III). The molecular mechanism of complementary chromatic adaptation has been studied extensively in Fremyella diplosiphon (Calothrix or Tolypothrix sp. PCC 7601), which performs group III chromatic adaptation (15). A series of mutational studies has suggested that a GAF/histidine-kinase type cyanobacteriochrome, RcaE, induces the expression of PC and associated linker genes via phosphorylation of two response regulators, RcaF and RcaC, under RL (1719). Although RcaE is suggested to repress the expression of PE and associated linker genes under RL (20), these genes still are induced by GL in both rcaE and rcaC mutants (2123), suggesting another pathway for induction of PE genes. In addition, the biochemical characteristics of RcaE protein, such as its spectral properties, chromophore species, and phosphorylation state, still are largely unknown (24). Recently, another GAF/histidine-kinase type cyanobacteriochrome, CcaS, was discovered in Synechocystis sp. PCC 6803 (S. 6803), which possesses PC but not PE in the PBS. CcaS covalently binds the linear tetrapyrrole, phycocyanobilin, at a conserved Cys residue of the GAF domain and undergoes reversible photoconversion between green- and red-absorbing forms with Z-E isomerization of the phycocyanobilin (8). Notably, the GAF domain of CcaS is homologous to that of RcaE, suggesting that RcaE also perceives GL and RL by a similar mechanism. However, CcaS has enhanced phosphorylation activity upon GL irradiation, an activity which apparently contrasts with the putative RL-activated phosphorylation of RcaE (8). In addition, CcaS directly transfers phosphates to the OmpR-class response regulator (CcaR) in vitro, differing from the putative three-component phosphorelay of RcaE to RcaF to RcaC. Both ccaS and ccaR are clustered with the rodcore linker of PC (cpcG2) in the genome of S. 6803 and have been suggested by microarray analysis to be involved in its transcriptional regulation (25). Interestingly, ccaS and ccaR orthologs are clustered with several PE and PC genes in the lamentous N2-xing cyanobacterium,

Author contributions: Y.H., R.N., and M.K. designed research; Y.H. performed research; Y.H. analyzed data; and Y.H. and M.I. wrote the paper. The authors declare no conict of interest. This article is a PNAS Direct Submission. Data deposition: The sequences reported in this paper have been deposited in CyanoBase or GenBank: CcaS (Npun_F3797), CcaR (Npun_R3793), CpeC (Npun_F3794), CpcG2 (Npun_F3795), CpeR1 (Npun_F3796), CpeB (Npun_R3807), CpeA (Npun_R3806), CpeE (Npun_F0736), CpcG1 (Npun_F3811), CpeR2 (Npun_F0739), CpcB (Npun_F5289), CpcA (Npun_F5290), RcaE (AAB08575), RcaF (AAC45448), RcaC (Q01473), CcaS of S. 6803 (ABI83649), and CcaR of S. 6803 (slr1584).
1

To whom correspondence should be addressed. E-mail: mikeuchi@bio.c.u-tokyo.ac.jp.

This article contains supporting information online at www.pnas.org/cgi/content/full/ 1000177107/DCSupplemental.

88548859 | PNAS | May 11, 2010 | vol. 107 | no. 19

www.pnas.org/cgi/doi/10.1073/pnas.1000177107

Nostoc punctiforme ATCC 29133 (PCC 73102), which performs group II chromatic adaptation (16). In this study, we deleted ccaS or ccaR in N. punctiforme and examined the responses of the mutants to GL and RL. Our results suggest that CcaS and CcaR regulate PE accumulation under both GL and RL in N. punctiforme, providing insights into the GL-sensing mechanism of PE accumulation and also into the evolution of the two different types of complementary chromatic adaptation. Results The PBS proteins in N. punctiforme are encoded by several gene clusters in the genome (Fig. S1). In one such group, cyanobacteriochrome ccaS and response regulator ccaR are clustered with genes for a rod linker of PE (cpeC), a rod-core linker of PC (cpcG2), and a putative transcriptional activator of PE (cpeR1) (Fig. 1A). The gene arrangement of ccaS, ccaR, and cpcG2 in N. punctiforme is similar to that found for S. 6803 (8). CcaS protein consists of an Nterminal transmembrane helix, a cyanobacteriochrome-type GAF domain, two Per/ARNT/Sim (PAS) domains, and a histidine-kinase domain (Fig. 1B). The domain organization of N. punctiforme CcaS is identical to that of S. 6803 CcaS, and their sequences are very similar (77% similar residues, 62% identical residues). The GAF domain of N. punctiforme CcaS was puried from phycocyanobilinproducing Escherichia coli and showed reversible photoconversion between the green-absorbing form (Pg; absorbance maximum at 536 nm) and the red-absorbing form (Pr; absorbance maximum at 672 nm) (Fig. 1 C and D). Covalent binding of a linear tetrapyrrole chromophore was conrmed by detecting uorescence of the denatured protein after SDS/PAGE (Fig. 1E). The molar extinction coefcients of the Pg and Pr peaks were calculated as 27,000 M1 cm1 and 30,000 M1 cm1, respectively. These features are almost identical to those of S. 6803 CcaS (8). CcaR is an OmpR-class response regulator that consists of a two-component receiver domain and a DNA-binding domain, and its sequence also is very similar to that of S. 6803 CcaR (83% similar residues, 73% identical residues). These ndings indicate that N. punctiforme CcaS and CcaR are orthologous to S. 6803 CcaS and CcaR, leading to predictions that CcaS/CcaR regulates cpeC, cpcG2, and cpeR1 expression by GL and RL in N. punctiforme. We allowed for the complete acclimation of N. punctiforme cells by culturing the cells in liquid medium for long periods (68 days) under GL or RL. We then analyzed their absorption spectra and their transcription of the PE and PC genes. For the absorption spectra of wild-type N. punctiforme cells (normalized to the PC peak maxima), the absorbance of the PE peak was substantially

greater under GL than under RL (Fig. 2A), indicating that N. punctiforme cells accumulate PE upon exposure to GL. It should be noted that the PE peak never completely disappeared under RL, even after full acclimation. Interestingly, Northern blotting showed that the expression levels of cpeC and cpcG2 were significantly higher under GL, whereas they were hardly detectable under RL (Fig. 3 A and B). The maximum size of the cpeC and cpcG2 transcript was 2 kbp. Although we could not detect cpeR1 transcripts by Northern blotting analysis, RT-PCR analysis showed that cpeR1 expression also was induced by GL but not by RL (Fig. 3G). Similarly, transcripts that span cpcG2-cpeR1 and cpeC-cpcG2-cpeR1 were induced by GL but not RL. These points suggest that cpeR1 is cotranscribed with cpeC and cpcG2 as a polycistron under GL (designated as the cpeC-cpcG2-cpeR1 operon). We also found that the expression level of the phycobiliprotein operon of PE (cpeB-cpeA) was 2.5-fold higher under GL than under RL in the wild-type cells, although the expression was not completely suppressed under RL (Fig. 3C). Conversely, the expression levels of another rod linker of PE (cpeE), another rodcore linker of PC (cpcG1), and the phycobiliprotein operon of PC (cpcB-cpcA) were 1.0- to 1.5-fold higher under RL than under GL (Fig. 3 DF). RT-PCR analysis showed the expression of another putative transcriptional activator of PE (cpeR2) was not signicantly changed under GL and RL (Fig. 3G). To study the roles of ccaS and ccaR in PE accumulation, we replaced these genes with the neomycin-resistance cassette from pSCR9 (26). We evaluated the full segregation of the ccaS or ccaR gene by PCR (Fig. S2 A and B) and examined the pigmentation and expression of PBS genes under GL or RL. In the absorption spectra of the ccaS mutant, the height of the PE peak under both GL and RL was 70% that of the wild type under GL (0% is the level in the wild type grown under RL) (Fig. 2B). This nding indicates that the ccaS mutant constitutively accumulates a certain level of PE under both GL and RL. Northern blotting showed that the expression of cpeC and cpcG2 under both GL and RL was 40% that of the wild type under GL (Fig. 3 A and B). In addition, RT-PCR showed that the expression of cpeR1, cpcG2-cpeR1, and cpeC-cpcG2-cpeR1 also was constitutively expressed under GL and RL (Fig. 3G). These results indicated that, in the ccaS mutant cpeC-cpcG2-cpeR1 was constitutively expressed under both GL and RL at a lower level than found for the wild type under GL. The expression of cpeB-cpeA under both GL and RL was similar to that of wild type under GL. Thus, the ccaS mutant also constitutively expressed cpeB-cpeA, although, unlike cpeC-cpcG2-cpeR1 expression, cpeB-cpeA expression was not decreased by the mutation (Fig. 3C).
PLANT BIOLOGY

Fig. 1. Gene arrangement, domain organization, and purication of CcaS and CcaR. (A) Arrangement of cyanobacteriochrome ccaS and cognate response regulator ccaR in the N. punctiforme genome. They are clustered with a rod linker of PE (cpeC), a rod-core linker of PC (cpcG2), and a transcriptional regulator of PE (cpeR1). (B) Domain architecture of CcaS (Upper) and CcaR (Lower). Phycocyanobilin is covalently bound to Cys147 of CcaS. His537 of CcaS and Asp51 of CcaR are conserved phosphorylation residues of the histidineasparagine (HisAsp) phosphorelay system. DB, DNA-binding domain; His kinase, histidinekinase domain; REC, receiver domain; TM, transmembrane. (C) Absorption spectra of the chromophore-binding GAF domain of N. punctiforme CcaS, which was puried from phycocyanobilin-producing E. coli. The spectrum of the green-absorbing form (Pg; red line) has a maximum at 536 nm, whereas that of the red-absorbing form (Pr; green line) has a maximum at 672 nm. (D) Photographs of Pg (Left) and Pr (Right). Pg is converted to Pr by irradiation with GL. Pr is converted to Pg by irradiation with RL. These photoconversions were fully reversible. (E) An SDS polyacrylamide gel showing the puried GAF domain of CcaS (lane 1) and full-length CcaR (lane 2). Fluorescence of covalently bound phycocyanobilin was detected for CcaS (lane 3) but not for CcaR (lane 4).

Hirose et al.

PNAS | May 11, 2010 | vol. 107 | no. 19 | 8855

Fig. 2. Pigmentation of PE and PC. Absorption spectra of wild type (A), ccaS (B), and ccaR (C) mutants that had been incubated under GL (green line) or RL (red line). The spectra were normalized to the PC peak maxima. The four major peaks are those for the absorbances of PE, PC, and Chl. Their photographs in liquid medium also are included in each graph.

Whether the ccaR mutant had been cultured under GL or RL, the PE peaks of its absorption spectra were of the same intensity as that of wild type under RL, for which a very slight PE peak was seen (Fig. 2C). This result indicates that the ccaR mutant is defective in GL-induced PE accumulation. The expression of cpeC and cpcG2 was hardly detectable under GL or RL by Northern hybridization (Fig. 3 A and B). RT-PCR showed that the expression of cpeR1, cpcG2-cpeR1, and cpeC-cpcG2-cpeR1 also was suppressed under both GL and RL (Fig. 3G). These results indicate that the ccaR mutant is defective in GL-induced cpeC-cpcG2-cpeR1 expression in addition to PE accumulation. The expression level of cpeB-cpeA in the ccaR mutant was as low as that of wild type in RL (Fig. 3C), although expression was slightly higher in RL than in GL. Conversely, the expression levels of cpeE, cpcG1, cpeR2, and cpcBcpcA were not much changed in either the ccaS or ccaR mutant under either light condition (Fig. 3 DG), indicating that expression of these genes is not regulated by CcaS or CcaR. Because deletion of ccaR resulted in complete loss of GL-induced cpeC-cpcG2-cpeR1 expression, we postulated that CcaR directly

regulates this expression. We examined the ability of CcaR to bind to the promoter region of cpeC-cpcG2-cpeR1 using the electrophoretic mobility shift assay. CcaR was puried from the soluble fraction of E. coli as an N-terminal His-tagged protein (Fig. 1D). When the puried CcaR protein was mixed with a DNA fragment containing the upstream region of cpeC, the CcaR-DNA complex was formed (Fig. 4A, lanes 14). The formation of this complex was inhibited by addition of unlabeled probes but not by addition of a nonspecic DNA fragment (Fig. 4A, lanes 5 and 6), suggesting that the CcaR protein specically binds upstream of this operon. Sequence alignment shows that there is a conserved direct repeat of [CTTTNCNATTT] 2 (designated as the G-box) in the upstream region of N. punctiforme cpeC and S. 6803 cpcG2 (Fig. 4B). These results suggest that CcaR directly binds to the G-box in the promoter of cpeC-cpcG2-cpeR1 and regulates its expression. Discussion In this study, we found that cpeC-cpcG2-cpeR1 expression and PE content are strictly regulated by GL and RL in N. punctiforme. We

Fig. 3. Transcription of PE and PC genes. Northern blots of transcribed PE and PC genes in the wild type and in the ccaS (ccaS) and ccaR (ccaR) mutants that had been incubated under GL (green bars) or RL (red bars). Extracted RNA was separated, blotted, stained with methylene blue for control of rRNA, and then hybridized with 32P-labeled probes of cpeC (A), cpcG2 (B), cpeB-cpeA (C), cpeE (D), cpcG1 (E), or cpcB-cpcA (F). Quantitation of each band is shown by the height of the corresponding bar below the gel lane. Wild type under GL was set to 100% to compare relative expression levels. Error bars indicate SD of three independent experiments. (G) RT-PCR analysis of cpeR1, cpcG2-cpeR1, cpeC-cpcG2-cpeR1, and cpeR2 using spanning primer sets shown in Fig S1. RT-PCR analysis of 16S rRNA also was included as positive (+RT) and negative (RT) controls.

8856 | www.pnas.org/cgi/doi/10.1073/pnas.1000177107

Hirose et al.

Fig. 4. Specic binding of CcaR to cpeC-cpcG2-cpeR1 promoter. (A) Electrophoretic mobility shift assay of CcaR. Puried CcaR [0 ng (lane 1), 100 ng (lane 2), 200 ng (lane 3), or 300 ng (lane 4)] was mixed with the radiolabeled DNA fragment that encodes the promoter region of cpeC-cpcG2-cpeR1. For the competition assay, 300 ng of CcaR was mixed with the radiolabeled probe and 1 g of unlabeled probe (lane 5) or 1 g of a nonspecic DNA fragment (lane 6). (B) Alignment of the promoter regions of N. punctiforme (N. pun) cpeC and S. 6803 (S.6803) cpcG2. A conserved direct-repeat motif of [CTTTNCNATTT] 2 (gray box) is designated the G-box. The numbers to the right of the sequences indicate the nucleotide positions with respect to the start codon.

demonstrated that chromatic regulation of cpeC-cpcG2-cpeR1 expression and chromatic regulation of PE content were completely lost in both ccaS and ccaR mutants. In addition, the recombinant CcaR protein directly bound to the promoter region of cpeC-cpcG2cpeR1. These results suggest that CcaS regulates cpeC-cpcG2-cpeR1 expression via phosphorylation of CcaR, leading to chromatic regulation of PE content in the PBS. In the ccaR mutant, cpeC-cpcG2-cpeR1 expression was completely suppressed under both GL and RL. This result indicates that CcaR is a transcriptional activator for cpeC-cpcG2-cpeR1 expression and is consistent with our earlier nding that cpcG2 expression is signicantly decreased in the S. 6803 ccaR mutant (25). In addition, this result suggests that there is no other dominant transcriptional activator for cpeC-cpcG2-cpeR1 expression, although N. punctiforme harbors a total of 102 response regulators (27). We speculate that the conserved G-box motif enables the specic interaction between CcaR and the cpeC-cpcG2-cpeR1 promoter. We showed that N. punctiforme CcaS and CcaR are orthologs of those in S. 6803. In our previous in vitro assay, S. 6803 CcaR was directly

phosphorylated by CcaS under GL rather than under RL (8), suggesting that N. punctiforme CcaR also is phosphorylated by CcaS under GL. In this study, GL irradiation induced cpeC-cpcG2-cpeR1 expression in wild-type N. punctiforme. These results strongly suggest that the phosphorylated state of CcaR is responsible for activation of cpeC-cpcG2-cpeR1 expression. Deletion of ccaS resulted in decreased cpeC-cpcG2-cpeR1 expression under GL, also suggesting that CcaS phosphorylates CcaR under GL in vivo. Notably, this low level of cpeC-cpcG2cpeR1 expression never disappeared under either light condition in the ccaS mutant, whereas the expression was suppressed completely in the ccaR mutant. This difference suggests that a CcaSindependent but CcaR-dependent pathway constitutively activates cpeC-cpcG2-cpeR1 expression in the ccaS mutant. In fact, N. punctiforme possesses a total of 153 histidine kinases (27), which may phosphorylate CcaR in a GL/RL-independent manner, leading to a constitutively low level of cpeC-cpcG2-cpeR1 expression. However, cpeC-cpcG2-cpeR1 expression was suppressed in wild type under RL. This suppression indicates that CcaR is not phosphorylated under RL in the presence of CcaS, despite the weak phosphorylation by other histidine kinases, and strongly suggests that CcaS not only phosphorylates CcaR under GL but also dephosphorylates CcaR under RL. It is commonly accepted that many histidine kinases catalyze both phosphorylation and dephosphorylation of the cognate response regulator (28). This bifunctionality is thought to prevent the undesirable phosphorylation of the response regulator by noncognate histidine kinases. We assume that both GL-induced phosphorylation and RL-induced dephosphorylation activities are essential so that CcaS can strictly control the phosphorylation state of CcaR by GL and RL. Notably, mutation of rcaE resulted in constitutive expression of PC genes under both GL and RL in F. diplosiphon, suggesting that RcaE also possesses both phosphorylation and dephosphorylation activities (15). From these results, we propose a molecular mechanism for complementary chromatic adaptation in N. punctiforme (Fig. 5). Under GL, CcaS phosphorylates CcaR, and the phosphorylated CcaR binds to the G-box in the cpeC-cpcG2-cpeR1 promoter, activating its expression. Under RL, CcaS dephosphorylates CcaR, so the dephosphorylated CcaR does not bind to the cpeC-cpcG2-cpeR1 promoter, and its expression remains suppressed. We also found that cpeB-cpeA expression is partially regulated by GL and RL. In F. diplosiphon, the regulatory protein CpeR is

Fig. 5. Model of complementary chromatic adaptation in N. punctiforme. (Left) Under GL, CcaS phosphorylates CcaR, and phosphorylated CcaR binds to the G-box of the cpeC-cpcG2-cpeR1 promoter, thus activating its expression. Both GL-induced CpeR1 and constitutively induced CpeR2 activate cpeB-cpeA expression. (Right) Under RL, CcaS dephosphorylates CcaR; dephosphorylated CcaR does not bind to the cpeC-cpcG2-cpeR1 promoter, and its expression remains suppressed. In this case, only CpeR2 activates cpeB-cpeA expression. GL- and RL-activated pathways are shown as green and red lines, respectively; the constitutively activated pathway is shown as a black line.

Hirose et al.

PNAS | May 11, 2010 | vol. 107 | no. 19 | 8857

PLANT BIOLOGY

induced by GL and activates the expression of cpeB-cpeA (23, 29), although the mechanism for its activation is not known. In N. punctiforme, one of the cpeR genes (cpeR1) was induced by GL (Fig. 3G), suggesting that CpeR1 activates cpeB-cpeA expression under GL. Interestingly, N. punctiforme possesses a second copy of the cpeR gene (cpeR2) that is homologous to cpeR1 (84% similar residues, 64% identical residues), and cpeR2 was not signicantly regulated by GL and RL (Fig. 3H). These facts suggest that both CpeR1 and CpeR2 activate cpeB-cpeA expression under GL, whereas only CpeR2 activates expression under RL (Fig. 5). Indeed, cpeB-cpeA expression and PE accumulation never stopped under RL in wild type even after full acclimation (Fig. 2A and Fig. 3H), whereas cpeR1 expression was shut down completely. It has been reported that several cyanobacteria species do not completely suppress PE synthesis under RL (16), possibly suggesting a common cpeR1 and cpeR2 organization in these species. We demonstrated that CcaS/CcaR is the GL-sensing system for PE accumulation in N. punctiforme. Although the genomes of other group II chromatic adapters have not yet been sequenced, we assume that CcaS/CcaR is conserved in these strains because their acclimation is very similar to that found for N. punctiforme (16, 30). In F. diplosiphon, a group III chromatic adapter, RcaE/ RcaF/RcaC serves as the RL-sensing system to induce PC accumulation, whereas an additional unidentied GL-sensing system has been implicated to induce the expression of PE genes. Could this GL-sensing system be CcaS/CcaR? Although the complete sequence of the F. diplosiphon genome has not been reported yet, that of another group III chromatic adapter, Synechococcus sp. PCC 7335 (GenBank; ABRV00000000), harbors RcaE/RcaF/RcaC but not CcaS/CcaR. This point may mean that the additional GL-sensing system for PE accumulation of group III strains is not CcaS/CcaR. Conversely, a system homologous to RcaE/RcaF/RcaC is not found in the genome of N. punctiforme, suggesting that N. punctiforme uses only CcaS/CcaR for group II chromatic adaptation. Our results and those of previous studies (15) suggest that group II and group III chromatic adaptations are regulated by two distinct cyanobacteriochromes, CcaS and RcaE, respectively. Notably, the GAF domain of CcaS shows the greatest sequence similarity (68% similar residues, 51% identical residues) to that of RcaE among all sequenced cyanobacteriochrome GAF domains. Residues that form the chromophore-binding pocket of CcaS also are highly conserved in RcaE (8). In addition, the N-terminal receiver domain of CcaR shows a high level of sequence similarity to that of RcaC (74% similar residues, 54% identical residues). These features strongly suggest that CcaS/CcaR is closely related to RcaE/RcaF/ RcaC. We assume that CcaS/CcaR is the prototype for RcaE/RcaF/ RcaC because the domain organization, phosphorylation pathway, and target PBS genes of CcaS/CcaR are much simpler than those of RcaE/RcaF/RcaC. The evolution of CcaS/CcaR to RcaE/RcaF/ RcaC may have enabled certain group II chromatic adapters to regulate many more PBS genes by GL and RL, leading to the emergence of group III chromatic adapters that change the PE/PC composition more dynamically (16, 31). On the other hand, some marine cyanobacteria recently have been reported to change their chromophore (phycourobilin and phycoerythrobilin) composition of PE responding to blue light and GL, a change that is called group IV (type IV) chromatic adaptation (32). However, we assume that this type of acclimation is regulated by other molecular mechanism(s), because no cyanobacteriochrome gene was found in the genome of these species. It is generally known that composition of PE, PC, and chlorophyll (Chl) is affected not only by light color but also by light intensity (33). This fact led to confused arguments in early times about whether light color or light intensity regulates chromatic adaptation. Subsequent analyses have suggested that the ratio of PE/PC is regulated by light color to harvest light energy maximally (16), whereas the ratio of PBS/Chl is regulated by both light color
8858 | www.pnas.org/cgi/doi/10.1073/pnas.1000177107

and light intensity to compensate their excitation imbalance (34). In this study, we demonstrated that the PE/PC ratio of N. punctiforme is regulated predominantly by light color via the CcaS/CcaR system, which is specialized to perceive GL and RL. Notably, we also found that the expression of cpeE, cpcG1, cpcB, and cpcA was 1.0- to 1.5-fold higher under RL than under GL in wild-type cells and in ccaS and ccaR mutants (Fig. 3). This expression seems to reect the increase of PBS/Chl under irradiation of RL that excites Chl more preferentially than PBS. The difculty of early studies might be that they mixed the adaptation of PE/PC ratio (regulated by CcaS or RcaE) with that of PBS/Chl ratio. To date, group II chromatic adapters have been thought not to regulate PC synthesis by GL and RL (16). However, the expression of cpcG2 is induced, along with cpeC and cpeR1, by GL in N. punctiforme. In S. 6803, disruption of cpcG2 results in a decrease of PC content (35). This nding suggests that GL-induced CpcG2 posttranslationally stabilized PC, leading to PC accumulation, although this effect was not so dramatic in this study. Interestingly, CpcG2 possesses a hydrophobic helix at its C terminus and is implicated in energy distribution from PBS to photosystem I in S. 6803 (3537). This nding may suggest that N. punctiforme increases not only the PE/PC ratio but also energy transfer to photosystem I under GL. It will be important to resolve the relationship between chromatic regulation of linker genes and the roles of their proteins. Material and Methods
Strains and Growth Conditions. The sequenced substrain (UCD153) of N. punctiforme ATCC 29133 was used as wild type (38). N. punctiforme cells were grown in liquid BG11 medium (39) bubbled with air containing 1% (vol/vol) CO2 at 31 C. For complementary chromatic adaptation, cells were irradiated with 20 E m2s1 of GL (peak at 520 nm, half bandwidth 50 nm) or 15 E m2s1 of RL (peak at 660 nm, half bandwidth 20 nm), which was supplied by color uorescent lamps (FL20S-G or FL20SL-R; Panasonic). SDS/PAGE and Spectral Assays. SDS/PAGE and detection of uorescence of phycocyanobilin was performed as described previously (8). Absorption spectra of the GAF domain were measured at room temperature using a UV2400PC spectrophotometer (Shimadzu). Absorption spectra of N. punctiforme cells were measured at room temperature using a U3500 spectrophotometer (Hitachi) with an end-on photomultiplier. Plasmid Construction. The sequence of each primer is listed in Table S1. Construction of ccaS and ccaR deletion plasmids is shown in Fig. S2. To make the expression construct of the GAF domain of N. punctiforme CcaS, the coding region was amplied with the primer set NpCcaS-16/NpCcaS-15, cloned, sequenced, and subcloned into the NdeI/BamHI sites of pET28a (Novagen). To make the expression construct of N. punctiforme CcaR, the primer set NpCcaR-1/NpCcaR-2 was used for PCR, and the subsequent procedures were as described above. Protein Expression and Purication. The His-tagged GAF domain of N. punctiforme CcaS was expressed in BL21(DE3) harboring the phycocyanobilinproducing vector, pKT271 (40), as described (8) with slight modications. We added 50 M -aminolevulinic acid but no FeCl3 to the LB medium. We used 1 mM isopropyl-thio--D-galactoside for induction. His-tagged N. punctiforme CcaR was expressed in BL21 (DE3) as described above but without addition of -aminolevulinic acid. Cell breakage and protein purication were performed as described (8), except that CcaR was isolated with a purication buffer that contained 500 mM NaCl. RNA Analysis. RNA was extracted from 70-mL N. punctiforme liquid cultures at midlogarithmic growth phase (A750 nm = 1.01.5). Liquid cultures were cooled quickly with ice and then centrifuged at 4,000 g for 5 min at 4 C. Cell pellets were moved to microtubes and centrifuged again; then 300-L volumes of TrisEDTA (pH 7.5) and 500-L volumes 0.1 M Tris-buffered phenol (pH 7.5) were added to each pellet. Cells were broken by vortexing in a bead beater (Microsmash MS100R; Tomy) in the presence of zirconia/silica beads at 5,000 rpm for 3 min. The aqueous phase was collected and sequentially treated with acidic phenol, phenol-chloroform, and DNase I; then the RNA was precipitated using ethanol. For Northern blotting, the total RNA (5 g/lane) was separated by electrophoresis through an agarose-formaldehyde gel. RNA was transferred overnight to a nylon membrane (Biodyne-plus; Pall) using 20 SSC (3 M

Hirose et al.

NaCl; 300 mM C6H5O7Na3) and cross-linked by UV irradiation. The rRNA, which served as the control, was stained with methylene blue. For probes, the coding region of each PBS gene was amplied by PCR using the following primer sets shown in Fig S1: NpCpeC-1/NpCpeC-2 for cpeC, NpCpcG2-1/NpCpcG2-2 for cpcG2, NpCpeB-1/NpCpeA-2 for cpeB-cpeA, NpCpcB-1/NpCpcA-2 for cpcBcpcA, NpCpeE-1/NpCpeE-2 for cpeE, and NpCpcG1-1/NpCpcG1-2 for cpcG1. Then each PCR product was gel-puried, labeled with [-32P]dCTP using the reagents of a random primer DNA-labeling kit (Takara), and puried through a Sephadex G-25 spin column (GE Healthcare). Blotted RNA was hybridized with each probe at 50 C for 1518 h in hybridization buffer [5 SSPE (3 M NaCl; 173 mM NaH2PO4; 25 mM EDTA); 50% formamide; 0.1% BSA; 0.1% Ficoll 400; 0.1% polyvinylpyrrolidone; 0.5% SDS; 0.02 mg/mL salmon sperm DNA] and then washed with 25 mL buffer 1 (2 SSPE; 0.1% SDS), buffer 2 (1 SSPE; 0.1% SDS), and buffer 3 (0.1 SSPE; 0.1% SDS) for 15 min at 50 C. The radiolabeled probes on the membrane were quantied by BAS-2500 (Fuji). For RT-PCR analysis, extracted RNA was reverse-transcribed with random primer using SuperScript III reverse transcriptase (Invitrogen) according to the manufacturers instructions. Reverse-transcribed RNA (25 ng) were subjected to PCR of 33 cycles with the primer sets shown in Fig S1: NpCpeR1-8/NpCpeR1-9 for cpeR1, NpCpcG2-8/NpCpeR1-6 for cpcG2-cpeR1, NpCpeR2-5/NpCpeR2-6 for cpeR2, and PCR of 37 cycles with NpCpeC-8/NpCpeR1-6 for cpeC-cpcG2-cpeR1.

Reverse-transcribed RNA (positive control) and RNA that was not reverse transcribed (negative control) were subjected to PCR of 25 cycles with the primer sets Npun-r020-1/Npun-r020-2, which encode 16S ribosomal RNA. Electrophoretic Mobility Shift Assay. For probes, the promoter region of cpeC was amplied with the primer set NpccaRD-3/NpCpeC-4 (total 286 bp). As a nonspecic competitor, a region downstream of CcaS was amplied with the primer set NpccaSD-3/NpCcaS-10 (total 271 bp). End-labeling of the DNA fragment, purication, binding, electrophoresis, and autoradiography were performed as described (41) with slight modications. We incubated the aliquots of the CcaR protein (0, 100, 200, or 300 ng/lane) with probes for 10 min at room temperature. For competition analysis, 300 ng of CcaR was mixed with radiolabeled probes and 1 g of unlabeled probes or 1 g of the nonspecic competitor. ACKNOWLEDGMENTS. We thank Dr. Jack Meeks (University of California, Davis) for genetic tools of N. punctiforme and for very helpful advice. We thank Dr. Takayuki Kohchi (Kyoto University) for the gift of the phycocyanobilin-coexpression system (pKT271) in E. coli. This work was supported by Grants-in-Aid for Scientic Research from the Japanese Society for the Promotion of Science to M.I. and to Y.H.

1. Karniol B, Wagner JR, Walker JM, Vierstra RD (2005) Phylogenetic analysis of the phytochrome superfamily reveals distinct microbial subfamilies of photoreceptors. Biochem J 392:103116. 2. Ikeuchi M, Ishizuka T (2008) Cyanobacteriochromes: A new superfamily of tetrapyrrole-binding photoreceptors in cyanobacteria. Photochem Photobiol Sci 7: 11591167. 3. Narikawa R, Kohchi T, Ikeuchi M (2008) Characterization of the photoactive GAF domain of the CikA homolog (SyCikA, Slr1969) of the cyanobacterium Synechocystis sp. PCC 6803. Photochem Photobiol Sci 7:12531259. 4. Yoshihara S, Katayama M, Geng X, Ikeuchi M (2004) Cyanobacterial phytochrome-like PixJ1 holoprotein shows novel reversible photoconversion between blue- and greenabsorbing forms. Plant Cell Physiol 45:17291737. 5. Rockwell NC, et al. (2008) A second conserved GAF domain cysteine is required for the blue/ green photoreversibility of cyanobacteriochrome Tlr0924 from Thermosynechococcus elongatus. Biochemistry 47:73047316. 6. Ishizuka T, et al. (2006) Characterization of cyanobacteriochrome TePixJ from a thermophilic cyanobacterium Thermosynechococcus elongatus strain BP-1. Plant Cell Physiol 47:12511261. 7. Narikawa R, Fukushima Y, Ishizuka T, Itoh S, Ikeuchi M (2008) A novel photoactive GAF domain of cyanobacteriochrome AnPixJ that shows reversible green/red photoconversion. J Mol Biol 380:844855. 8. Hirose Y, Shimada T, Narikawa R, Katayama M, Ikeuchi M (2008) Cyanobacteriochrome CcaS is the green light receptor that induces the expression of phycobilisome linker protein. Proc Natl Acad Sci USA 105:95289533. 9. Yoshihara S, et al. (2006) Reconstitution of blue-green reversible photoconversion of a cyanobacterial photoreceptor, PixJ1, in phycocyanobilin-producing Escherichia coli. Biochemistry 45:37753784. 10. Ulijasz AT, et al. (2009) Cyanochromes are blue/green light photoreversible photoreceptors dened by a stable double cysteine linkage to a phycoviolobilin-type chromophore. J Biol Chem 284:2975729772. 11. Ishizuka T, Narikawa R, Kohchi T, Katayama M, Ikeuchi M (2007) Cyanobacteriochrome TePixJ of Thermosynechococcus elongatus harbors phycoviolobilin as a chromophore. Plant Cell Physiol 48:13851390. 12. Grossman AR, Schaefer MR, Chiang GG, Collier JL (1993) The phycobilisome, a lightharvesting complex responsive to environmental conditions. Microbiol Rev 57: 725749. 13. Tandeau de Marsac N (2003) Phycobiliproteins and phycobilisomes: The early observations. Photosynth Res 76:197205. 14. Grossman AR (2003) A molecular understanding of complementary chromatic adaptation. Photosynth Res 76:207215. 15. Kehoe DM, Gutu A (2006) Responding to color: The regulation of complementary chromatic adaptation. Annu Rev Plant Biol 57:127150. 16. Tandeau de Marsac N (1977) Occurrence and nature of chromatic adaptation in cyanobacteria. J Bacteriol 130:8291. 17. Chiang GG, Schaefer MR, Grossman AR (1992) Complementation of a red-lightindifferent cyanobacterial mutant. Proc Natl Acad Sci USA 89:94159419. 18. Kehoe DM, Grossman AR (1996) Similarity of a chromatic adaptation sensor to phytochrome and ethylene receptors. Science 273:14091412. 19. Kehoe DM, Grossman AR (1997) New classes of mutants in complementary chromatic adaptation provide evidence for a novel four-step phosphorelay system. J Bacteriol 179:39143921. 20. Li L, Alvey RM, Bezy RP, Kehoe DM (2008) Inverse transcriptional activities during complementary chromatic adaptation are controlled by the response regulator RcaC binding to red and green light-responsive promoters. Mol Microbiol 68:286297. 21. Alvey RM, Karty JA, Roos E, Reilly JP, Kehoe DM (2003) Lesions in phycoerythrin chromophore biosynthesis in Fremyelia diplosiphon reveal coordinated light

22. 23.

24.

25.

26.

27. 28. 29.

30.

31. 32. 33. 34. 35.

36.

37.

38.

39. 40.

41.

regulation of apoprotein and pigment biosynthetic enzyme gene expression. Plant Cell 15:24482463. Li L, Kehoe DM (2005) In vivo analysis of the roles of conserved aspartate and histidine residues within a complex response regulator. Mol Microbiol 55:15381552. Seib LO, Kehoe DM (2002) A turquoise mutant genetically separates expression of genes encoding phycoerythrin and its associated linker peptides. J Bacteriol 184: 962970. Terauchi K, Montgomery BL, Grossman AR, Lagarias JC, Kehoe DM (2004) RcaE is a complementary chromatic adaptation photoreceptor required for green and red light responsiveness. Mol Microbiol 51:567577. Kaytayama M, Ikeuchi M (2006) Perception and transduction of light signals by cyanobacteria. Frontier in Life Sciences, Perception and Transduction of Light Signals by Cyanobacteria, eds Fujiwara M, Sato N, Ishiura S (Research Signpost, Kerala, India), pp 6590. Cohen MF, Meeks JC (1997) A hormogonium regulating locus, hrmUA, of the cyanobacterium Nostoc punctiforme strain ATCC 29133 and its response to an extract of a symbiotic plant partner Anthoceros punctatus. Mol Plant Microbe Interact 10: 280289. Meeks JC, et al. (2001) An overview of the genome of Nostoc punctiforme, a multicellular, symbiotic cyanobacterium. Photosynth Res 70:85106. Laub MT, Goulian M (2007) Specicity in two-component signal transduction pathways. Annu Rev Genet 41:121145. Cobley JG, et al. (2002) CpeR is an activator required for expression of the phycoerythrin operon (cpeBA) in the cyanobacterium Fremyella diplosiphon and is encoded in the phycoerythrin linker-polypeptide operon (cpeCDESTR). Mol Microbiol 44:15171531. Neuschaefer-Rube O, Boger P, Ernst A (2002) Interference of an apcA insertion with complementary chromatic adaptation in the diazotrophic Synechocystis sp strain BO 8402. Biochim Biophys Acta 1553:279295. Bryant DA (1982) Phycoerythrocyanin and phycoerythrinproperties and occurrence in cyanobacteria. J Gen Microbiol 128:835844. Everroad C, et al. (2006) Biochemical bases of type IV chromatic adaptation in marine Synechococcus spp. J Bacteriol 188:33453356. Grossman AR, Schaefer MR, Chiang GG, Collier JL (1995) The Molecular Biology of Cyanobacteria (Kluwer, Dordrecht, The Netherlands), pp 641675. Jones LW, Myers J (1965) Pigment variations in Anacystis nidulans induced by light of selected wavelengths. J Phycol 1:714. Kondo K, Geng XX, Katayama M, Ikeuchi M (2005) Distinct roles of CpcG1 and CpcG2 in phycobilisome assembly in the cyanobacterium Synechocystis sp. PCC 6803. Photosynth Res 84:269273. Kondo K, Ochiai Y, Katayama M, Ikeuchi M (2007) The membrane-associated CpcG2phycobilisome in Synechocystis: A new photosystem I antenna. Plant Physiol 144: 12001210. Kondo K, Mullineaux CW, Ikeuchi M (2009) Distinct roles of CpcG1-phycobilisome and CpcG2-phycobilisome in state transitions in a cyanobacterium Synechocystis sp. PCC 6803. Photosynth Res 99:217225. Campbell EL, Christman H, Meeks JC (2008) DNA Microarray comparisons of plant factorand nitrogen deprivation-induced hormogonia reveal decision-making transcriptional regulation patterns in Nostoc punctiforme. J Bacteriol 190:73827391. Rippka R (1988) Isolation and purication of cyanobacteria. Methods Enzymol 167: 327. Mukougawa K, Kanamoto H, Kobayashi T, Yokota A, Kohchi T (2006) Metabolic engineering to produce phytochromes with phytochromobilin, phycocyanobilin, or phycoerythrobilin chromophore in Escherichia coli. FEBS Lett 580:13331338. Midorikawa T, Matsumoto K, Narikawa R, Ikeuchi M (2009) An Rrf2-type transcriptional regulator is required for expression of psaAB genes in the cyanobacterium Synechocystis sp PCC 6803. Plant Physiol 151:882892.

Hirose et al.

PNAS | May 11, 2010 | vol. 107 | no. 19 | 8859

PLANT BIOLOGY