Volume 214, Issue 3 p. 967-972
Rapid report
Free Access

Alternative electron transport mediated by flavodiiron proteins is operational in organisms from cyanobacteria up to gymnosperms

Petr Ilík

Corresponding Author

Petr Ilík

Department of Biophysics, Centre of the Region Haná for Biotechnological and Agricultural Research, Palacký University, 783 71 Olomouc, Czech Republic

Author for correspondence:

Petr Ilík

Tel: +420 585634153

Email: [email protected]

Search for more papers by this author
Andrej Pavlovič

Andrej Pavlovič

Department of Biophysics, Centre of the Region Haná for Biotechnological and Agricultural Research, Palacký University, 783 71 Olomouc, Czech Republic

Search for more papers by this author
Roman Kouřil

Roman Kouřil

Department of Biophysics, Centre of the Region Haná for Biotechnological and Agricultural Research, Palacký University, 783 71 Olomouc, Czech Republic

Search for more papers by this author
Alessandro Alboresi

Alessandro Alboresi

Department of Biology, University of Padova, 35131 Padova, Italy

Search for more papers by this author
Tomas Morosinotto

Tomas Morosinotto

Department of Biology, University of Padova, 35131 Padova, Italy

Search for more papers by this author
Yagut Allahverdiyeva

Yagut Allahverdiyeva

Department of Biochemistry, Molecular Plant Biology, University of Turku, 20014 Turku, Finland

Search for more papers by this author
Eva-Mari Aro

Eva-Mari Aro

Department of Biochemistry, Molecular Plant Biology, University of Turku, 20014 Turku, Finland

Search for more papers by this author
Hiroshi Yamamoto

Hiroshi Yamamoto

Department of Botany, Graduate School of Science, Kyoto University, 606-8502 Kyoto, Japan

CREST, Japan Science and Technology Agency, Chiyoda-ku, 102-0076 Tokyo, Japan

Search for more papers by this author
Toshiharu Shikanai

Toshiharu Shikanai

Department of Botany, Graduate School of Science, Kyoto University, 606-8502 Kyoto, Japan

CREST, Japan Science and Technology Agency, Chiyoda-ku, 102-0076 Tokyo, Japan

Search for more papers by this author
First published: 17 March 2017
Citations: 108

Summary

  • Photo-reduction of O2 to water mediated by flavodiiron proteins (FDPs) represents a safety valve for the photosynthetic electron transport chain in fluctuating light. So far, the FDP-mediated O2 photo-reduction has been evidenced only in cyanobacteria and the moss Physcomitrella; however, a recent phylogenetic analysis of transcriptomes of photosynthetic organisms has also revealed the presence of FDP genes in several nonflowering plant groups. What remains to be clarified is whether the FDP-dependent O2 photo-reduction is actually operational in these organisms.
  • We have established a simple method for the monitoring of FDP-mediated O2 photo-reduction, based on the measurement of redox kinetics of P700 (the electron donor of photosystem I) upon dark-to-light transition. The O2 photo-reduction is manifested as a fast re-oxidation of P700. The validity of the method was verified by experiments with transgenic organisms, namely FDP knock-out mutants of Synechocystis and Physcomitrella and transgenic Arabidopsis plants expressing FDPs from Physcomitrella.
  • We observed the fast P700 re-oxidation in representatives of all green plant groups excluding angiosperms.
  • Our results provide strong evidence that the FDP-mediated O2 photo-reduction is functional in all nonflowering green plant groups. This finding suggests a major change in the strategy of photosynthetic regulation during the evolution of angiosperms.

Introduction

In plants, the linear electron transport from water via photosystem II (PSII) and photosystem I (PSI) to NADP+ is complemented by a wide range of alternative electron transfer pathways (Fig. 1a). These pathways are crucial for plants experiencing dynamic environmental changes, including stress or abrupt changes in light intensity. A decade ago, we discovered in lichens with Trebouxia as symbiotic green algae an interesting phenomenon – a very fast re-oxidation of PSI upon dark-to-light transition (0.2–0.4 s after onset of illumination, Ilík et al., 2006), suggesting fast activation of an alternative electron flow from PSI. Interestingly, this phenomenon was not detectable in experiments with pea leaves. At that time, this fast PSI re-oxidation was attributed to a fast activation of ferredoxin–NADP+ oxidoreductase (FNR) or O2 photo-reduction (Mehler reaction) at the acceptor side of PSI (Ilík et al., 2006). Later, this fast PSI re-oxidation was detected also in green alga Chlamydomonas reinhardtii and it was shown to be oxygen-dependent, which indicated that Mehler type reaction is involved (Franck & Houyoux, 2008). Recently, this fast oxygen-dependent PSI re-oxidation upon dark-to-light transition was observed also in some representatives of gymnosperms (Shirao et al., 2013; Pavlovič et al., 2016).

Details are in the caption following the image
Redox kinetics of P700 upon dark-to-light transition in photosynthetic organisms with deleted/introduced flv genes. (a) A scheme of electron transport pathways in green plants. The Calvin–Benson–Bassham (CBB) cycle is dependent on functioning of ferredoxin(Fd)-NADP+ oxidoreductase (FNR), cyclic electron transport around photosystem I (PSI) (indicated by red arrows) consists of two pathways: the PGR5/PGR5-LIKE PHOTOSYNTHETIC PHENOTYPE 1 (PGRL1)- and NADH dehydrogenase-like complex (NDH)-dependent pathways. Photo-reduction of O2 takes place via the Mehler reaction (photo-reduction of O2 to superoxide anion radical) and via the ‘Mehler-like’ reaction (flavodiiron proteins-mediated photo-reduction of O2 to water). (b–i) Redox kinetics of P700 in: Synechocystis sp. PCC 6803 – wild type (WT) and its ∆flv1 and ∆flv3 mutants (b, c, f, g); Physcomitrella patens – WT and its ∆flvA and ∆flvB mutants (d, h); and Arabidopsis thaliana – WT and transgenic plants expressing flvA and flvB genes from Physcomitrella patens (e, i). The kinetics of redox changes were measured in vivo upon exposure of dark-adapted sample to actinic light (2000 μmol photons m−2 s−1). (b–e) P700 kinetics measured under aerobic conditions, while (f–i) show the P700 kinetics measured under anaerobic conditions. Panels (b, c) and (f, g) show the same curves on different timescales (linear vs logarithmic).

These new findings, directly connecting the fast PSI re-oxidation to oxygen reduction, led us to an idea that the paper published by Radmer and his coworkers 50 years ago (Radmer & Kok, 1976) may describe the same phenomenon that we have observed in our experiments with lichens. They studied the onset of O2 evolution in photosynthetic organisms upon dark-to-light transition (Radmer & Kok, 1976; Marsho et al., 1979) and they noticed a several seconds long lag phase in O2 evolution in cyanobacteria and green algae, but not in spinach cells or chloroplasts. They proved that the lag phase was caused by a transient fast O2 uptake, which was insensitive to inhibitors of the Calvin–Benson–Bassham (CBB) cycle. At that time, the authors interpreted the results as photo-reduction of O2 at the acceptor side of PSI and suggested that this process is much slower in higher plants (Marsho et al., 1979).

Based on all these findings, we began to search for a relatively widespread alternative electron pathway, effectively and quickly routing electrons from PSI to oxygen upon dark-to-light transition, and one possible route came into focus – flavodiiron-dependent pathway. Flavodiiron proteins (FDPs) are flavoproteins that have originally been found in anaerobic and some aerobic prokaryotes and in eukaryotic Protozoa and that are able to reduce O2 to water without the formation of reactive oxygen species (‘Mehler-like’ reaction, Allahverdiyeva et al., 2013) (for a review see Allahverdiyeva et al., 2015). The involvement of this type of reaction in photosynthetic electron transport was first described in 2003 by Helman et al. in a cyanobacterium Synechocystis sp. PCC6803 (hereafter Synechocystis). Shortly upon dark-to-light transition, when CO2 fixation is still not active, FDPs Flv1 and Flv3 mediate photo-reduction of O2 at the acceptor side of PSI (Helman et al., 2003) (see Fig. 1a). Originally, it has been assumed that these FDPs are present only in phylogenetically old organisms and that they do not have wider importance. This was probably the reason why this pathway did not attract the attention of a broader plant scientist community. However, < 10 years ago, the analysis of available transcriptomes has shown that homologs to flv1 and flv3 genes, flvA and flvB, are present also in phylogenetically younger plant groups – green algae, bryophytes and lycophytes (Zhang et al., 2009; Peltier et al., 2010; Dang et al., 2014; Jokel et al., 2015) (for phylogenetic tree of green plants see Fig. 2a). Recently, the flvA and flvB genes were also found in a number of representatives of gymnosperms (Allahverdiyeva et al., 2015; Yamamoto et al., 2016). Obviously, the sole presence of the genes does not necessarily mean that the respective alternative pathway is active in all the earlier-mentioned plant groups. However, very recently it has been shown that in the moss Physcomitrella patens (hereafter Physcomitrella), the FDP-dependent pathway is indeed active upon dark-to-light transition and that this pathway represents the main electron sink during the first seconds after the onset of illumination (Gerotto et al., 2016). These results are regarded as the first evidence of the FDP function in terrestrial plants.

Details are in the caption following the image
Redox kinetics of P700 upon dark-to-light transition in representatives of different plant groups from the phylogenetic tree. (a) Phylogenetic tree of green plant groups according to Clarke et al. (2011) with indicated loss of flavodiiron proteins (FDPs), genera used in this study are listed in brackets (for full names of the plant species see the Materials and Methods section). (b) Redox kinetics of P700 in the selected species under aerobic conditions. (c) Redox kinetics of P700 in the same species under anaerobic conditions. Redox kinetics were measured in vivo upon exposure of dark-adapted sample to actinic light (2000 μmol photons m−2 s−1) and the curves were normalized.

All this evidence lead us to a hypothesis that the FDP-dependent pathway is much more widespread in the plant kingdom than originally thought and we have decided to use the measurement of PSI redox kinetics to verify the hypothesis.

Materials and Methods

Biological samples

Synechocystis sp. PCC 6803 and its ∆flv1 or ∆flv3 knock-out mutants (Helman et al., 2003), as well as Chlamydomonas reinhardtii CC-002 strains, were grown at temperature 23 ± 1°C in BG-11 medium with periodic 14 h : 10 h, day : night cycle (40 μmol photons m−2 s−1, PAR). For the measurement, cells were harvested and resuspended in BG-11 medium to OD750 nm of 0.35 for Synechocystis strains and 0.6 for Chlamydomonas (Unicam UV 550; ThermoSpectronic, Cambridge, UK). Terrestrial plants were taken from a glasshouse collection of the Department of Botany (Palacký University in Olomouc, Czech Republic) or from the wild close to Olomouc city. Wild type and transgenic Arabidopsis thaliana expressing flvA and flvB genes from Physcomitrella patens were grown in peat substrate at 23 ± 1°C with periodic 9 h : 15 h, day : night cycle (50 μmol photons m−2 s−1, PAR) for 7 wk according to Yamamoto et al. (2016). Protonemata of Physcomitrella patens, Gransden, and its flvA or flvB knock-out mutants were cultured on PpNO3 minimum medium at 24°C in 16 h : 8 h, light : dark photoperiod and the light intensity of 50 μmol photons m−2 s−1 (PAR) for 10 d (Gerotto et al., 2016). A list of other species used in this study: Marchantia polymorpha, Sphagnum girgensohnii, Phaeoceros carolinianus, Selaginella kraussiana, Cyathea cooperi, Cycas circinalis, Ginkgo biloba, Thuja plicata, Pinus nigra, Welwitschia mirabilis, Gnetum sp., Nicotiana tabacum, Zea mays. For their position in the phylogenetic tree, see Fig. 2(a).

Redox changes of P700

The redox changes of P700 were recorded using a Dual-PAM 100 measuring system (Heinz Walz, Effeltrich, Germany) during 6 s long illumination (2000 μmol photons m−2 s−1). Before the measurement, samples (leaves, thalli, protonemata or cells) were dark-adapted for 30 min. The measurements with terrestrial plants were performed on cut leaves or thalli. The measurements with Physcomitrella were performed at protonema stage on plants grown on solid medium. Anaerobiosis was induced by bubbling the cultivation medium with gaseous nitrogen (aquatic samples) or by incubation of the samples in nitrogen atmosphere for 15 to 20 min (all other samples).

Results and Discussion

Cyanobacterium Synechocystis was the first photosynthetic organism in which the O2 photo-reduction mediated by FDPs was discovered (Helman et al., 2003). The measurement of P700 redox kinetics upon abrupt illumination of dark-adapted wild type Synechocystis cells shows initial P700 oxidation, followed by its reduction and re-oxidation, creating a wave-like pattern (Fig. 1b,c). Similar measurements have been performed earlier, but the wave has not been observed due to a lower time resolution (Helman et al., 2003). In the experiments with Synechocystis ∆flv1 and ∆flv3 mutants, the P700 re-oxidation phase was clearly missing, as was in experiments with wild type Synechocystis performed under anaerobic conditions (Fig. 1f,g). Very similar results were observed for the moss Physcomitrella and its ∆flvA and ∆flvB mutants (Gerotto et al., 2016; Fig. 1d,h). We performed the measurement of P700 redox changes upon dark-to-light transition also in wild type and transgenic Arabidopsis expressing flvA and flvB genes from Physcomitrella patens (Yamamoto et al., 2016). While we did not observe the P700 re-oxidation phase in the wild type, in the transgenic Arabidopsis this re-oxidation phase was present (Fig. 1e), although it was slower than in Synechocystis and Physcomitrella. Again, under anaerobic conditions the re-oxidation phase disappeared (Fig. 1i).

The experiments with these three phylogenetically different photosynthetic organisms and their mutants/transgenic plants allowed us to make a connection between the operation of FDP-dependent pathway and P700 redox changes detectable upon dark-to-light transition. The initial oxidation of P700 reflects the intrinsic activity of PSI (i.e. the charge separation in P700 and electron transfer to ferredoxin) and the subsequent P700 reduction reflects the intake of electrons originating in water splitting at PSII (Schansker et al., 2003, 2005). The final re-oxidation of P700 can be attributed, in general, to the outflow of electrons from PSI. As we have observed this re-oxidation phase only under aerobic conditions and only in plants where FDPs are present (Fig. 1), we can attribute it to the cooperative action of Flv1 and Flv3 proteins (in Synechocystis) or their analogs FlvA and FlvB proteins (in Physcomitrella and transgenic Arabidopsis).

Once the correspondence between the operative FDP-dependent pathway and P700 re-oxidation phase was established, we have used the measurement of the P700 redox kinetics upon dark-to-light transition for a systematic screening of FDP-mediated electron transport in representatives of all groups of green plants from the phylogenetic tree (Fig. 2a). Our experiments revealed that the fast oxygen-dependent P700 re-oxidation phase is present in all green plants up to gymnosperms and that the only plant group where this phase is missing is angiosperms (Fig. 2b,c). Thus, our results strongly indicate that the FDP-mediated alternative electron transport from PSI to O2 is an universal pathway, operational in all photosynthetic organisms from cyanobacteria up to gymnosperms.

Given the presence of the FDP-dependent pathway in such a wide range of plant species, it is obvious that it has to play an important physiological role. Indeed, recent studies with Synechocystis, Anabaena sp. PCC 7120, Chlamydomonas reinhardtii and Physcomitrella and their mutants defective in the synthesis of Flv1 (FlvA) or Flv3 (FlvB) proteins have clearly demonstrated the importance of FDPs for photosynthetic organisms exposed to fluctuating light (Allahverdiyeva et al., 2013; Dang et al., 2014; Gerotto et al., 2016). Upon sudden increase in light intensity, these FDPs function as an effective electron sink downstream of PSI, minimizing the risk of over-reduction of the photosynthetic electron transport chain. When the FDP-dependent pathway is not functional, there is a bottleneck at the acceptor side of PSI and a sudden increase in light intensity results in PSI photo-inhibition and damage (Allahverdiyeva et al., 2013; Gerotto et al., 2016). Besides acting as a route for safe disposal of excessive electrons, the FDP pathway also helps to build up a proton gradient across the thylakoid membrane. The gradient is crucial for triggering other protective mechanisms, including induction of nonphotochemical quenching of excitations (NPQ) and downregulation of electron transport (see Allahverdiyeva et al., 2015).

A question arises whether the absence of FDP-dependent electron transport in angiosperms is compensated by some alternative electron pathway. One of the likely candidates for this role is the PGR5/PGRL1-dependent cyclic electron transport around PSI (CET, see Fig. 1a). Indeed, recent studies with Arabidopsis plants (Suorsa et al., 2012; Kono et al., 2014) and Physcomitrella mutants lacking FDPs (Gerotto et al., 2016) have confirmed its importance for plants exposed to fluctuating light. Similarly to the FDP-dependent electron transport, the PGR5/PGRL1-dependent CET also helps to build up the proton gradient across the thylakoid membrane, triggering photoprotective mechanisms (for a recent review see Shikanai & Yamamoto, 2017).

The involvement of other alternative pathways in the photoprotection upon sudden increase in light intensity is much less likely. There are basically two other pathways, which could hypothetically have some importance. One of them is the well-known Mehler reaction (see Fig. 1a), the other involves plastid terminal oxidase (PTOX). Mehler reaction has been shown to be too slow to act as an effective electron sink under these conditions and therefore it does not seem to have significant role in the photoprotection upon sudden increase in light intensity (Badger et al., 2000; Heber, 2002). However, it could be involved in the regulation of CET, as one of its by-products (H2O2) was shown to directly activate NDH-dependent CET (see Fig. 1a) in A. thaliana (Strand et al., 2015). The second theoretically relevant pathway acts at the level plastoquinone pool and involves the oxidation of plastoquinol molecules by the action of PTOX. Nevertheless, even this pathway seems to have only minor importance for the oxidation of over-reduced electron transport chain, most probably due to low PTOX abundance. The significance of this pathway might increase under conditions leading to the overexpression of PTOX, e.g. in plants exposed to severe stress (for a review see McDonald et al., 2011).

Another intriguing question arises as to why the apparently important FDP-mediated pathway is missing in angiosperms. To answer this question we have to look back at the evolution of this plant group. Ecophysiological characteristics of early extant angiosperms suggest that their common ancestor has evolved in an understorey habitat, which is typically characterized by a relatively stable, but light-limited conditions (Feild et al., 2004; Feild & Arens, 2005). Under such low-light conditions, plants have to manage the photosynthesis-induced reducing power very carefully. One can assume that the FDP-mediated pathway, carelessly ‘dissipating’ the precious reducing power by routing electrons to O2, was not a strategy that would provide any evolutionary advantage in such habitat.

Acknowledgements

The authors thank Ursula Ferretti and Lenka Kuchařová for the cultivation of cyanobacteria and alga cultures, Dr Martin Dančák for the selection and provision of various plant species of different phylogenetic age, Dr Zbyněk Hradílek, Štěpán Koval and their colleagues for providing the authors with a sufficient quantity of liverworts. Further, the authors thank Dr Iva Ilíková for manuscript editing and Prof. Wolfgang Bilger for stimulating discussions. This work was supported by the grant project LO1204 (Sustainable development of research in the Center of the Region Haná) from the National Program of Sustainability I from the Ministry of Education, Youth and Sports, Czech Republic. R.K. was supported by a Marie Curie Career Integration Grant call FP7-PEOPLE-2012-CIG (322139).

    Author contributions

    P.I., A.P. and R.K. planned and designed the research, performed experiments and analyzed data; A.A. and T.M. performed a complete set of experiments with Physcomitrella; H.Y. prepared Arabidopsis plants; P.I. wrote the main part of the manuscript; Y.A., T.M., H.Y, E-M.A. and T.S. revised the manuscript.