State transitions: an example of acclimation to low-light stress

Transcription

1 Journal of Experimental Botany, Vol. 56, No. 411, Light Stress in Plants: Mechanisms and Interactions Special Issue, pp , January 2005 doi: /jxb/eri064 Advance Access publication 6 December, 2004 State transitions: an example of acclimation to low-light stress Conrad W. Mullineaux* and Daniel Emlyn-Jones Department of Biology, University College London, Darwin Building, Gower Street, London WC1E 6BT, UK Received 26 March 2004; Accepted 25 October 2004 Abstract State 1 State 2 transitions ( state transitions ) are a rapid physiological adaptation mechanism that adjusts the way absorbed light energy is distributed between photosystem I and photosystem II. They occur in both green plants and cyanobacteria, although the light-harvesting complexes involved are very different. Which aspects of the mechanism are conserved in green plants and cyanobacteria and which may be different, are discussed. It is shown that phycobilisome mobility is necessary for state transitions in cyanobacteria. A conserved cyanobacterial gene (rpac) that plays a very specific role in state transitions has been identified. There is still debate about the physiological role of state transitions. Comparison of the growth properties of the rpac deletion mutant with the wildtype gives us a way of directly addressing the question. It was found that state transitions are physiologically important only at very low light intensities: they play no role in protection from photoinhibition. Thus state transitions are a way to maximize the efficiency of light-harvesting at low light intensities. Key words: Cyanobacteria, light-harvesting, photoinhibition, photosynthesis, phycobilisomes, state transitions. State transitions in cyanobacteria and green plants State 1 State 2 transitions ( state transitions ) are a rapid mechanism for reconfiguring the photosynthetic light-harvesting apparatus in response to changing conditions. The phenomenon was first described in a red alga (Murata, 1969) and a green alga (Bonaventura and Myers, 1969) more than 30 years ago. Although green plants have a very different light-harvesting apparatus from the cyanobacteria and red algae, state transitions are conceptually similar in both groups of organisms. Illumination conditions which lead to excess excitation of photosystem II (PSII) compared with photosystem I (PSI) induce a transition to State 2, in which more absorbed excitation energy is diverted to PSI. When PSI is over-excited relative to PSII this induces a transition to State 1, in which more energy is transferred to PSII. Thus state transitions appear to act as a mechanism to balance excitation of the two photosystems under changing light regimes (reviewed by van Thor et al., 1998; Allen and Forsberg, 2001). State transitions can be triggered by changes in the redox state of electron carriers between PSII and PSI. In green plants, a specific site on the cytochrome b 6 f complex has been implicated (Vener et al., 1997). It is likely that the triggering mechanism is similar in cyanobacteria, where it is clear that state transitions must be triggered by something that is in redox equilibrium with plastoquinone (Mullineaux and Allen, 1990). In green plants the subsequent signal transduction pathway involves the activation or deactivation of a protein kinase, which phosphorylates a part of the pool of LHCII light-harvesting complexes. This leads to redistribution of LHCII between PSII and PSI (reviewed in Allen and Forsberg, 2001; Fig. 1). The biochemical mechanism of state transitions in cyanobacteria is not known, but it is likely to be significantly different from that in green plants. An excellent candidate for the LHCII kinase has recently been identified in the green alga Chlamydomonas reinhardtii (Depège et al., 2003). It has no obvious orthologue in cyanobacteria. One gene specifically required for state transitions in cyanobacteria has been identified (Emlyn-Jones et al., 1999), but it has no known orthologues in green plants (see below). * To whom correspondence should be addressed. Fax: +44 (0) c.mullineaux@ucl.ac.uk y Present address: Research School of Biological Sciences, Australian National University, Canberra, Australian Capital Territory 2601, Australia Journal of Experimental Botany, Vol. 56, No. 411, ª Society for Experimental Biology 2004; all rights reserved

2 390 Mullineaux and Emlyn-Jones fluorescence data it can be estimated that up to about 50 60% of phycobilisomes are decoupled from PSII on transition to State 2 (Mullineaux et al., 1990), and energy storage studies indicate that these phycobilisomes must then be functionally coupled to PSI (Mullineaux et al., 1991). The biochemical mechanism involved is not known, but one gene specifically required for the process has been identified. Fig. 1. Comparison of state transitions in green plants and cyanobacteria. Mechanism of state transitions in cyanobacteria Energy transfer and mutant studies have shown that phycobilisomes in cyanobacteria can transfer energy directly to PSI as well as to PSII (Mullineaux, 1994; Rakhimberdieva et al., 2001). State transitions change the relative energy transfer from phycobilisomes to PSI and PSII, and also the distribution of chlorophyll-absorbed energy (van Thor et al., 1998). Although the two effects normally occur together, the phycobilisome effect can be specifically inhibited by mutagenesis (Emlyn-Jones et al., 1999; McConnell et al., 2002). This suggests two independent mechanisms, both triggered by the same initial redox signal (Fig. 1). In cyanobacteria grown under standard conditions, the phycobilisomes are the major accessory light-harvesting complexes, and the phycobilisome effect is quantitatively more significant than the chlorophyll effect, which is not well-characterized. Studies using Fluorescence Recovery after Photobleaching (FRAP) have shown that the phycobilisomes are mobile complexes, diffusing rapidly on the surface of the thylakoid membrane. By contrast, PSII is completely immobile under normal conditions (Mullineaux et al., 1997; Sarcina et al., 2001). This indicates that the association between phycobilisomes and reaction centres is transient and unstable. Recently a direct connection between phycobilisome mobility and state transitions has been established. When cyanobacterial cells are immersed in buffers of high osmotic strength, phycobilisome diffusion is strongly inhibited (Joshua and Mullineaux, 2004). Under the same conditions, cells are locked into either State 1 or State 2, depending on how they were adapted prior to addition of the buffer. This indicates that the diffusion of phycobilisomes from reaction centre to reaction centre is required for state transitions. The results suggest a dynamic equilibrium model for state transitions, in which the signal transduction pathway leads to a change in the binding constant of phycobilisomes for PSII and/or PSI, leading to a change in the steady-state populations of phycobilisomes coupled to each type of reaction centre. From time-resolved A gene specifically required for state transitions in cyanobacteria A random mutagenesis approach was used to look for genes required for state transitions in the cyanobacterium Synechocystis 6803 (Emlyn-Jones et al., 1999). Libraries of tagged, random mutants were generated by random cartridge mutagenesis (Chauvat et al., 1989) and screened by using a fluorescence video imaging system to visualize state transitions in cell colonies. Marker rescue and sequencing were used to identify the sites of insertions/deletions in the mutants, and targeted interposon mutagenesis was used to confirm the phenotype (Emlyn-Jones et al., 1999). Five mutants specifically deficient in state transitions were isolated. All five mutants were the result of different insertional events, but all proved to have lesions in the same open reading-frame (Emlyn-Jones, 2000). This suggests that only one gene is specifically required for state transitions in cyanobacteria. However, the random cartridge mutagenesis method (Chauvat et al., 1989) often produces large deletions. Hence genes which are close to other, essential, genes may be difficult to identify by this method. The open reading-frame which had been identified was designated sll1926 in the CyanoBase Synechocystis 6803 database ( Nakamura et al., 2000) and the gene was named rpac (Regulator of Phycobilisome Association C) (Emlyn-Jones et al., 1999). rpac codes for a previously uncharacterized putative membrane protein, with 85 amino acids and a molecular weight of about 9 kda. Initially, the open reading-frame was predicted to code for a 14 kda protein with a significant hydrophilic domain at the N-terminus (Emlyn-Jones et al., 1999). However, this part of the sequence is not conserved, and it is therefore likely that translation starts at an alternative downstream site. There are two predicted transmembrane alpha-helices (Fig. 2) but no other recognizable structural or functional motifs. Hence the identification of rpac unfortunately provides no clues to the biochemical mechanism of state transitions in cyanobacteria. Since 1999 the genome sequences of nine more cyanobacteria have been completed, or are close to completion. rpac is strongly conserved in all of them, with one exception. There is no rpac orthologue in MED4, a high-light ecotype of Prochlorococcus marinus (sequenced by Rocap et al., 2003). However, there are clear rpac orthologues in the low light-ecotypes of Prochlorococcus marinus. In SS120 (sequenced by Dufresne et al., 2003) the open reading-frame is

3 designated Pro0741, and in MIT9313 (sequenced by Rocap et al., 2003) it is PMT0493. Comparison of the predicted RpaC polypeptide sequences from nine cyanobacteria allows the identification of strongly conserved residues (Fig. 2). The Synechocystis rpac deletion mutant shows a very clear and specific phenotype. State transitions are completely absent as judged from fluorescence measurements with phycobilin excitation. However, an effect of state transitions can still be detected with chlorophyll excitation (Emlyn-Jones et al., 1999; McConnell et al., 2002). The mutant is not significantly affected in electron transport as judged from oxygen evolution at saturating light, nor is there any marked effect on the cellular content of phycobilisomes and reaction centres. Apart from the absence of state transitions, there is no obvious effect on light harvesting. Phycobilisomes still transfer energy to PSII and PSI reaction centres, and only the fine-tuning due to state transitions is lost. Judging from low-temperature fluorescence spectra, cells of the rpac deletion mutant are trapped in State 1 (Emlyn-Jones et al., 1999). The mutant phenotype suggests that the RpaC gene product is involved in the phycobilisome branch of the state transition signal transduction pathway (Fig. 1). However, one anomaly is that there are clear rpac genes in the lowlight Prochlorococcus marinus ecotypes, which have no phycobilisomes, although they do retain some residual phycobiliproteins (Hess et al., 1999). No rpac orthologue has been identified in any green plant, and the completion of the Arabidopsis thaliana genome sequence (The Arabidopsis Genome Initiative, 2000) makes it possible to state with confidence that rpac is not present in this organism. Expression of rpac rpac mrna was detected by Hihara et al. (2001) using a Synechocystis DNA microarray. Cells were grown at Fig. 2. The predicted sequence and membrane topology of the Synechocystis 6803 rpac gene product. The gene product is likely to be a thylakoid membrane protein, but this is not yet experimentally demonstrated. Polypeptide sequence is obtained from CyanoBase ( Residues underlined are highly-conserved among different cyanobacteria. State transitions 391 a relatively low light intensity of 20 le m ÿ2 s ÿ1. When cells were exposed to high light (300 lem ÿ2 s ÿ1 ) the rpac mrna level fell by about 50% within 15 min (Y Hihara, personal communication). Thus the rpac gene appears to be preferentially expressed at low light intensities. This correlates with the authors experience of measuring state transitions in Synechocystis: state transitions can only reliably be observed in cells grown at low light intensities, and cultures are routinely grown at 5 10 le m ÿ2 s ÿ1 for this purpose. Cyanobacterial state transitions are physiologically important at very low light intensities The very specific phenotype of the Synechocystis rpac deletion mutant provides an ideal system for testing the physiological role of state transitions. Growth experiments were carried out comparing the doubling time of the wildtype and the rpac ÿ mutant under continuous illumination over a range of light intensities. There is no significant difference in doubling time under white light at 500 lem ÿ2 s ÿ1 (Emlyn-Jones, 2000) 100 le m ÿ2 s ÿ1,or10le m ÿ2 s ÿ1 (Emlyn-Jones et al., 1999). However, under very weak white light at 2 lem ÿ2 s ÿ1 the doubling time for the rpac ÿ mutant becomes 30% longer than for the wild-type (Emlyn- Jones et al., 1999). The effect becomes more extreme when cells are grown under yellow light which is specifically absorbed by the phycobilisomes. Under yellow light at 2 le m ÿ2 s ÿ1, the doubling time for rpac ÿ is nearly 40% longer than for the wild-type. By contrast, there is little difference in doubling time when the cells are grown under weak red light (also 2 lem ÿ2 s ÿ1 ), which is specifically absorbed by chlorophyll (Emlyn-Jones et al., 1999). It was concluded that state transitions become physiologically important only at very low light intensities. The RpaC gene product is needed to maximize the efficiency of utilization of light energy absorbed by the phycobilisomes. Cyanobacterial state transitions play no role in protection from photoinhibition The results described above suggest that state transitions are not important under high light, and therefore they probably play no role in protection against photoinhibition. To test this point more specifically cell cultures of the wildtype and the rpac ÿ mutant were exposed to light at 1600 le m ÿ2 s ÿ1. Neither strain showed a significant decline in oxygen evolution over 3 h (Emlyn-Jones, 2000). In the presence of lincomycin, which inhibits the PSII repair cycle, both strains showed a 50% decline in oxygen evolution after 3 h (Emlyn-Jones, 2000). Thus there is no indication that the rate of PSII photodamage is greater in the rpac ÿ mutant, or that the PSII repair cycle is impeded.

4 392 Mullineaux and Emlyn-Jones Discussion: what are state transitions for? Studies of the growth phenotype of the rpac ÿ mutant strongly indicate that state transitions in cyanobacteria are physiologically important only at very low light intensities. They are a way to maximize the efficiency of utilization of absorbed light energy under conditions when light is strongly limiting for growth. Some further indications in support of this idea come from the expression pattern of the rpac gene (Hihara et al., 2001) and the absence of rpacin MED4, the high-light ecotype of Prochlorococcus marinus. These conclusions confirm the original ideas of Murata (1969) and Bonaventura and Myers (1969) as to the function of state transitions. It has been suggested, mainly in the context of green algae, that state transitions may play a second role in the protection against photoinhibition (Finazzi et al., 2001). This could conceivably arise if cells enter State 2 at very high light intensities. This could act to reduce the rate of PSII photodamage by minimizing PSII antenna size. At the same time, an increase in cyclic electron transport around PSI (Wollman, 2001) could provide a supply of ATP required for the PSII repair cycle (Finazzi et al., 2001). This may be a peculiarity of state transitions in green algae such as Chlamydomonas, where there is evidence that the transition to State 2 leads to a major switch from linear to cyclic electron flow (Wollman, 2001). State transitions do not appear to play a role in protection from photoinhibition in Arabidopsis (Lunde et al., 2003). Results with the rpac ÿ mutant also indicate that state transitions are not physiogically important at high light intensities in cyanobacteria. However, it is possible that other adaptation mechanisms, perhaps bearing some resemblance to state transitions, become active under high light conditions. These results simply indicate that any such mechanisms do not require the same gene products as classic low-light state transitions. A different term should be found to describe them. Acknowledgements We thank John Allen for helpful discussions, Peter Nixon for help with the photoinhibition experiment, and Yukako Hihara for communicating an unpublished result. DE-J was supported by a BBSRC research studentship. Work in CWM s laboratory is supported by BBSRC and The Wellcome Trust. References Allen JF, Forsberg J Molecular recognition in thylakoid structure and function. Trends in Plant Science 6, Bonaventura C, Myers J Fluorescence and oxygen evolution from Chlorella pyrenoidosa. Biochimica et Biophysica Acta 189, Chauvat F, Rouet P, Bottin H, Boussac A Mutagenesis by random cloning of an Escherichia coli kanamycin resistance gene into the genome of the cyanobacterium Synechocystis PCC6803; selection of mutants defective in photosynthesis. Molecular and General Genetics 216, Depège N, Bellafiore S, Rochaix J-D Role of chloroplast protein kinase Stt7 in LHCII phosphorylation and state transition in Chlamydomonas. Science 299, Dufresne A, Salanoubat M, Partensky F, et al Genome sequence of the cyanobacterium Prochlorococcus marinus SS120, a nearly minimal oxyphototrophic genome. Proceedings of the National Academy of Sciences, USA 100, Emlyn-Jones D A gene required for regulation of photosynthetic light harvesting in the cyanobacterium Synechocystis PCC6803. PhD thesis, University of London. Emlyn-Jones D, Ashby MK, Mullineaux CW A gene required for the regulation of photosynthetic light-harvesting in the cyanobacterium Synechocystis Molecular Microbiology 33, Finazzi G, Barbagallo RP, Bergo E, Barbato R, Forti G Photoinhibition of Chlamydomonas reinhardtii in State 1 and State 2. Damages to the photosynthetic apparatus under linear and cyclic electron flow. Journal of Biological Chemistry 276, Hess WR, Steglich C, Lichtlé C, Partensky F Phycoerythrins of the oxyphotobacterium Prochlorococcus marinus are associated to the thylakoid membranes and are encoded by a single large gene cluster. Plant Molecular Biology 40, Hihara Y, Kamei A, Kanehisa M, Kaplan A, Ikeuchi M DNA microarray analysis of cyanobacterial gene expression during acclimation to high light. The Plant Cell 13, Joshua S, Mullineaux CW Phycobilisome diffusion is required for light-state transitions in cyanobacteria. Plant Physiology 135, Lunde C, Jensen PE, Rosgaard L, Haldrup A, Gilpin MJ, Scheller HV Plants impaired in state transitions can to a large degree compensate for their defect. Plant and Cell Physiology 44, McConnell MD, Koop R, Vasil ev S, Bruce D Regulation of the distribution of chlorophyll and phycobilin-absorbed excitation energy in cyanobacteria. A structure-based model for the light-state transition. Plant Physiology 130, Mullineaux CW Excitation energy transfer from phycobilisomes to photosystem I in a cyanobacterial mutant lacking photosystem II. Biochimica et Biophysica Acta 1184, Mullineaux CW, Allen JF State 1 State 2 transitions in the cyanobacterium Synechococcus 6301 are controlled by the redox state of electron carriers between photosystems I and II. Photosynthesis Research 22, Mullineaux CW, Bittersmann E, Allen JF, Holzwarth AR Picosecond time-resolved fluorescence emission spectra indicate decreased energy transfer from the phycobilisome to photosystem II in light-state 2 in the cyanobacterium Synechococcus Biochimica et Biophysica Acta 1015, Mullineaux CW, Griebenow S, Braslavsky SE Photosynthetic energy storage in cyanobacterial cells adapted to light-states 1 and 2. A laser-induced optoacoustic study. Biochimica et Biophysica Acta 1060, Mullineaux CW, Tobin MJ, Jones GR Mobility of photosynthetic complexes in thylakoid membranes. Nature 390, Murata N Control of excitation transfer in photosynthesis. 1. Light-induced change of chlorophyll a fluorescence in Porphyridium cruentum. Biochimica et Biophysica Acta 172, Nakamura Y, Kaneko T, Tabata S CyanoBase, the genome database for Synechocystis sp. strain PCC6803: status for the year Nucleic Acids Research 28, 72. Rakhimberdieva MG, Boichenko VA, Karapetyan NV, Stadnichuk IN Interaction of phycobilisomes with Photosystem II dimers and Photosystem I monomers and trimers in the cyanobacterium Spirulina platensis. Biochemistry 40,

5 State transitions 393 Rocap G, Larimer FW, Lamerdin J, et al Genome divergence in two Prochlorococcus ecotypes reflects oceanic niche differentiation. Nature 424, Sarcina M, Tobin MJ, Mullineaux CW Diffusion of phycobilisomes on the thylakoid membranes of the cyanobacterium Synechococcus 7942: effects of phycobilisome size, temperature and membrane lipid composition. Journal of Biological Chemistry 276, The Arabidopsis Genome Initiative Analysis of the genome sequence of the flowering plant Arabidopsis thaliana. Nature 408, van Thor JJ, Mullineaux CW, Matthijs HCP, Hellingwerf KJ Light-harvesting and state transitions in cyanobacteria. Botanica Acta 111, Vener AV, van Kan PJM, Rich PR, Ohad I, Andersson B Plastoquinol at the quinol oxidation site of reduced cytochrome bf mediates signal transduction between light and protein phosphorylation: thylakoid protein kinase deactivation by a single-turnover flash. Proceedings of the National Academy of Sciences, USA 94, Wollman F-A State transitions reveal the dynamics and flexibility of the photosynthetic apparatus. EMBO Journal 20,