Divergent Protein Redox Dynamics and Their Relationship with Electron Transport Efficiency during Photosynthesis Induction

Abstract

Various chloroplast proteins are activated/deactivated during the light/dark cycle via the redox regulation system. Although the photosynthetic electron transport chain provides reducing power to redox-sensitive proteins via the ferredoxin (Fd)/thioredoxin (Trx) pathway for their enzymatic activity control, how the redox states of individual proteins are linked to electron transport efficiency remains uncharacterized. Here we addressed this subject with a focus on the photosynthetic induction phase. We used Arabidopsis plants, in which the amount of Fd–Trx reductase (FTR), a core component in the Fd/Trx pathway, was genetically altered. Several chloroplast proteins showed different redox shift responses toward low- and high-light treatments. The light-dependent reduction of Calvin–Benson cycle enzymes fructose 1,6-bisphosphatase (FBPase) and sedoheptulose 1,7-bisphosphatase (SBPase) was partially impaired in the FTR-knockdown ftrb mutant. Simultaneous analyses of chlorophyll fluorescence and P700 absorbance change indicated that the induction of the electron transport reactions was delayed in the ftrb mutant. FTR overexpression also mildly affected the reduction patterns of FBPase and SBPase under high-light conditions, which were accompanied by the modification of electron transport properties. Accordingly, the redox states of FBPase and SBPase were linearly correlated with electron transport rates. In contrast, ATP synthase was highly reduced even when electron transport reactions were not fully induced. Furthermore, the redox response of proton gradient regulation 5-like photosynthetic phenotype1 (PGRL1; a protein involved in cyclic electron transport) did not correlate with electron transport rates. Our results provide insights into the working dynamics of the redox regulation system and their differential associations with photosynthetic electron transport efficiency.

Introduction

Thiol-based redox regulation is a post-translational modification mechanism that controls enzyme activity by changing the reduction/oxidation states of Cys residues in target proteins (e.g. cleavage/formation of a disulfide bond). Thioredoxin (Trx), a small ubiquitous protein, is pivotal for the redox regulation system. Trx contains the highly conserved WCGPC amino acid sequence at the active site. Using the redox-active Cys pair in this sequence, Trx catalyzes dithiol–disulfide exchange reactions with its target proteins, thereby modulating their enzymatic activity. This Trx-mediated redox regulation system is present in both prokaryotic and eukaryotic cells.

Plant chloroplasts also have the redox regulation system, but its mode of action is unique as it is closely linked to light (Buchanan 2016). The photosynthetic electron transport chain in the thylakoid membrane converts light energy into reducing power. Trx receives some of the reducing power from the electron carrier protein ferredoxin (Fd) via Fd–Trx reductase (FTR), then transfers it to Trx-targeted redox-sensitive proteins and reductively cleaves their regulatory disulfide bonds. Various chloroplast proteins, directly or indirectly involved in photosynthesis, are subjected to Trx-dependent redox regulation. Four Calvin–Benson cycle enzymes [glyceraldehyde 3-phosphate dehydrogenase, fructose 1,6-bisphosphatase (FBPase), sedoheptulose 1,7-bisphosphatase (SBPase) and phosphoribulokinase] are the well-known examples (Michelet et al. 2013, Gurrieri et al. 2021). Most Trx-targeted chloroplast proteins are activated by reduction, but a few proteins, including glucose 6-phosphate dehydrogenase and phosphofructokinase (PFK), are deactivated by reduction (Yoshida et al. 2019a, Yoshida and Hisabori 2021). Thus, the Fd/Trx pathway serves to transfer the reducing power as a light signal to redox-sensitive proteins, thereby enabling the light-responsive control of various biological processes in chloroplasts (Buchanan 2016).

Although the Fd/Trx pathway was identified nearly half a century ago (Buchanan et al. 1979), the molecular basis and physiological significance of the overall redox regulation system in chloroplasts remain underexplored. Another feature of this system is the existence of a large set of Trx subtypes containing Trx-f, -m, -x, -y and -z (Serrato et al. 2013). Although their functional diversity remains debatable, previous biochemical and reverse genetic studies have partially clarified the specific and redundant roles of each Trx subtype. Trx-f is suggested as the preferable regulator of multiple metabolic enzymes in the stroma and ATP synthase (Marri et al. 2009, Thormahlen et al. 2013, Yoshida et al. 2015, Naranjo et al. 2016, Sekiguchi et al. 2020). For example, among the five Trx subtypes, Trx-f acts solely, at least in vitro, as an efficient reducing factor for FBPase and PFK (Yoshida et al. 2015, Yoshida and Hisabori 2021). This specificity might result from the positive charge around the active site of Trx-f (Yokochi et al. 2019). Trx-m also likely mediates the reduction of several metabolic enzymes, but its specific role may be the regulation of cyclic electron transport (CET) by interacting with the proton gradient regulation 5-like photosynthetic phenotype1 (PGRL1) protein (Okegawa and Motohashi 2020). Trx-x and Trx-y are considered efficient donors of reducing power to antioxidant enzymes (Collin et al. 2004, Pulido et al. 2010, Laugier et al. 2013). A recent reverse genetic study clearly indicated that Trx-x and Trx-y synergistically maintain the redox balance of the photosystem (PS) I acceptor side (Okegawa et al. 2023). Trx-z has been suggested to play a critical role in chloroplast biogenesis by regulating plastid gene expression (Arsova et al. 2010, Diaz et al. 2018) and RNA editing (Wang et al. 2021). Another notable factor in the redox regulation system in chloroplasts is NADPH-Trx reductase C (NTRC), which contains both an NADPH-Trx reductase domain and a Trx domain within a single polypeptide (Serrato et al. 2004). Using NADPH as a source of reducing power, NTRC can work independently of the Fd/Trx pathway. A major role of NTRC is likely to maintain the redox balance of 2-Cys peroxiredoxin (Perez-Ruiz et al. 2006, 2017), but other diverse functions have also been suggested, including the regulation of chlorophyll synthesis and starch metabolism (Richter et al. 2013, Skryhan et al. 2018). A recent study suggested a new role for NTRC, i.e. the regulation of CP12 protein to activate the Calvin–Benson cycle during cold acclimation (Teh et al. 2023). Furthermore, in addition to typical Trx subtypes, other proteins containing Trx-like CXXC motifs have been identified in chloroplasts (Meyer et al. 2006, Chibani et al. 2009, 2021). Some of these Trx-like proteins [Trx-like2 and atypical Cys His-rich Trx (ACHT)] were shown to oxidize several chloroplast enzymes and reduce 2-Cys peroxiredoxin with high efficiency (Yoshida et al. 2018, 2019a, Yokochi et al. 2019). These studies led to the identification of the protein-oxidation machinery that deactivates photosynthetic metabolism using the reactive oxygen species (ROS)–derived oxidizing force (Yoshida et al. 2019b). Thus, a complex redox network that includes multiple factors and pathways has been unveiled as the new landscape of the redox regulation system in chloroplasts; however, further studies are required to understand its whole nature (Cejudo et al. 2021, Meyer et al. 2021, Yoshida and Hisabori 2023). In particular, how the working efficiency of the system is modulated toward environmental changes remains elusive.

We recently reported an indispensable role of the Fd/Trx pathway in connecting the photosynthetic electron transport chain and redox-sensitive proteins in chloroplasts (Yoshida et al. 2022). Interestingly, the light-dependent reduction of stromal enzymes (e.g. FBPase and SBPase) was completely suppressed by the genetic disruption of the Fd/Trx pathway, while that of ATP synthase could be partially maintained. Furthermore, we previously found that inhibitors of photosynthetic electron transport [e.g. 3-(3,4-dichlorophenyl)-1,1-dimethylurea (DCMU)] differentially impair the light-dependent reduction of stromal enzymes and ATP synthase (Yoshida et al. 2014). These unexpected observations motivated us to firmly evaluate the relationship between the electron transport efficiency and the redox states of individual redox-sensitive proteins. In this study, we assess their relationships and dynamics during the photosynthetic induction phase using FTR knockdown and overexpression plants.

Results

Growth phenotypes of FTR knockdown and overexpression plants

FTR acts as a central hub in the Fd/Trx pathway linking the electron transport chain and the redox-sensitive proteins in chloroplasts (Dai et al. 2004). Therefore, genetically modifying its accumulation level may provide insights into the relationship between electron transport efficiency and the redox regulation system. In Arabidopsis, the catalytic subunit of FTR is encoded by a single FTRB gene. In this study, we used the Arabidopsis ftrb mutant containing T-DNA in the fourth intron of the FTRB gene (Yoshida and Hisabori 2016). Furthermore, we introduced the FTRB gene into Arabidopsis plants under the control of the cauliflower mosaic virus 35S promotor and eventually isolated two lines of transgenic plants (35S::FTRB #3-1 and #9-5; Fig. 1A). When grown under short-day conditions, the accumulation level of FTR protein was estimated to be approximately 40%, 700% and 900% of the wild-type level in ftrb, 35S::FTRB #3-1 and 35S::FTRB #9-5, respectively (Fig. 1B, C). We therefore used ftrb and 35S::FTRB as FTR knockdown and overexpression plants, respectively.

Fig. 1.

Growth phenotypes of the ftrb mutant and 35S::FTRB plants under short-day conditions. (A) The wild-type plant (WT), ftrb mutant and 35S::FTRB plants were grown for 43 d. (B) Immunoblotting analysis of the FTR catalytic subunit (FTRc). The same amount of total leaf protein (except for the WT dilution series) was loaded onto each lane. As a loading control, the Rubisco large subunit was stained with Coomassie Brilliant Blue R-250 (CBB). (C) Quantification of the FTRc protein amount. Relative values to the WT level are shown. Each value represents the mean ± SD (three different samples). Asterisks denote a significant difference (*P < 0.05, **P < 0.01, Student’s t-test). (D) Fresh weight of aboveground tissues. Each value represents the mean ± SD (10 different samples). (E) Chlorophyll content in the leaf. Each value represents the mean ± SD (five different samples). (D and E) Different letters denote significant differences (P < 0.01, Tukey–Kramer multiple comparison test).

We examined their growth phenotypes under several conditions. Under short-day conditions, the fresh weight of the ftrb mutant was significantly lowered, while that of the 35S::FTRB plants remained unchanged (Fig. 1A, D). The chlorophyll content was unaltered in the ftrb mutant and 35S::FTRB plants (Fig. 1A, E). The growth impairment in the ftrb mutant was also evident under long-day and continuous-light conditions (Supplementary Fig. S1A, B). It was more pronounced under fluctuating light conditions where the light intensity was repeatedly switched every 15 min between 60–70 and 350–370 μmol photons m⁻² s⁻¹ (Supplementary Fig. S1C). Under fluctuating light conditions, the growth impairment was also observed in 35S::FTRB plants (Supplementary Fig. S1C). These results indicate that changes in FTR activity significantly affect plant growth. Plants grown under short-day conditions (Fig. 1) were used for subsequent experiments.

Protein redox shift responses in FTR knockdown and overexpression plants

We next examined the protein redox shift responses during dark to light transitions. In this study, we focused on four proteins: the ATP synthase CF₁-γ subunit, the Calvin–Benson cycle enzymes FBPase and SBPase, and PGRL1 involved in CET. Fig. 2 shows the results obtained under low-light conditions (60 μmol photons m⁻² s⁻¹). The CF₁-γ showed a rapid reduction response, reaching approximately 80% reduction within 1 min after exposure to light. This redox shift pattern was almost identical among all examined plants. FBPase was reduced by 40% 3 min after exposure to light in the wild-type and 35S::FTRB plants. SBPase was reduced more slowly by light in these plants and reached approximately 30% reduction 3 min after exposure to light. In the ftrb mutant, the light-dependent reduction of FBPase and SBPase was mildly suppressed. While it is established that CF₁-γ, FBPase and SBPase form an intramolecular disulfide bond in an oxidized state, the redox regulation mechanism of PGRL1 remains unclear. Nevertheless, the most plausible mechanism is that the PGRL1 directly interacts with a specific isoform of Trx (Trx-m4 in Arabidopsis) via the intermolecular disulfide bond, lowering CET activity (Okegawa and Motohashi 2020). In agreement with this idea, we detected two major bands in the immunoblotting analysis using the PGRL1 antibody. According to Okegawa and Motohashi (2020), the upper (approx. 45 kDa) and lower (approx. 30 kDa) bands would likely correspond to the disulfide-linked Trx-m4–PGRL1 complex and the reduced monomer of PGRL1, respectively. Under dark conditions, approximately half of PGRL1 was present in the complex. PGRL1 was nearly completely shifted to the monomer just after exposure to light and then slightly formed a complex again. This redox-based formation/dissociation of the PGRL1 complex was commonly observed in all plants.

Fig. 2.

Protein redox shift responses during dark to low light transitions. (A) The WT, ftrb mutant and 35S::FTRB plants were dark-adapted and then irradiated with low light (60 μmol photons m⁻² s⁻¹) for the indicated times. Finally, the redox states of CF₁-γ, FBPase, SBPase and PGRL1 were determined. As a loading control, the Rubisco large subunit was stained with CBB. Ox, oxidized form: Red, reduced form. (B) The reduction level was calculated as the ratio of the reduced form to the total. Each value represents the mean ± SD (three different samples).

Similar experiments were performed under high-light conditions (360 μmol photons m⁻² s⁻¹; Fig. 3). As in the case of low light, CF₁-γ was shifted to a highly reduced state (>80% reduction) immediately after exposure to light in all plants. FBPase and SBPase were gradually reduced by light, reaching approximately 60% and 50% reduction 3 min after exposure to light, respectively, in the wild-type plant. The light-dependent reduction of FBPase and SBPase was substantially delayed in the ftrb mutant. In 35S::FTRB plants, FBPase and SBPase were reduced with the same efficiency as the wild-type plant until 0.5–1 min after exposure to light, but then their redox states were kept stable until 3 min. The band pattern of PGRL1 was nearly identical among the plants, which was also similar to that observed under low-light conditions.

Fig. 3.

Protein redox shift responses during dark to high light transitions. (A) The WT, ftrb mutant and 35S::FTRB plants were dark adapted and then irradiated with high light (360 μmol photons m⁻² s⁻¹) for the indicated times. Finally, the redox states of CF₁-γ, FBPase, SBPase and PGRL1 were determined. As a loading control, the Rubisco large subunit was stained with CBB. Ox, oxidized form: Red, reduced form. (B) The reduction level was calculated as the ratio of the reduced form to the total. Each value represents the mean ± SD (three different samples).

Electron transport efficiency in FTR knockdown and overexpression plants

We then examined the photosynthetic electron transport efficiency by simultaneously measuring the chlorophyll fluorescence and P700 absorbance change (Fig. 4). The maximal operating efficiency of PSII, F_v/F_m, was slightly lowered in the ftrb mutant (Supplementary Fig. S2). Actinic light was applied to the leaf at three different intensities (60, 120 and 360 μmol photons m⁻² s⁻¹). The electron transport rates of PSII (ETR II) and PSI (ETR I) were gradually elevated by light; their induction patterns differed among the three light intensities. For example, ETR II and ETR I rapidly (<1 min after exposure to light) reached a plateau at low levels under low-light conditions (60 μmol photons m⁻² s⁻¹), whereas they continuously increased until 3 min under high-light conditions (360 μmol photons m⁻² s⁻¹). In the ftrb mutant, the increases in ETR II and ETR I were partially suppressed at all light intensities. Lower ETR II and ETR I in the ftrb mutant were accompanied by modifications of other parameters, including the increase in 1—qL (an indicator of the reduced state of the plastoquinone pool) and the increase in Y (NA) (an indicator of the PSI acceptor side limitation). The 35S::FTRB plants did not show obvious differences in the induction patterns of photosynthetic parameters from those in the wild-type plant, although they exhibited weak tendencies of higher non-photochemical quenching (NPQ; an indicator of thermal dissipation of light energy) and lower ETR I under high-light conditions.

Fig. 4.

Photosynthetic electron transport dynamics during dark to light transitions. The WT, ftrb mutant and 35S::FTRB plants were dark-adapted and then irradiated with low (60 μmol photons m⁻² s⁻¹), middle (120 μmol photons m⁻² s⁻¹) or high light (360 μmol photons m⁻² s⁻¹) for the indicated times. During this period, several parameters, including ETR II, 1—qL, NPQ, ETR I, Y (ND) and Y (NA), were determined. Each value represents the mean ± SD (five different samples).

Relationship between electron transport efficiency and protein redox states

By integrating the data on the redox states of individual proteins (Figs. 2, 3) and electron transport efficiency (Fig. 4), we analyzed their relationship. The reduction level of each protein was plotted against ETR II and ETR I as general indices of electron transport efficiency (Fig. 5). CF₁-γ showed the nearly completely reduced level even under conditions where ETR II and ETR I were substantially low. In contrast, the reduction level of FBPase was elevated along with the increases in both ETR II and ETR I. Their linear positive correlations were evident under both low- and high-light conditions, but the slope of the regression line was steeper under low-light conditions. The reduction level of SBPase also showed linear positive correlations with ETR II and ETR I, although their correlations under low-light conditions were not so strong. Concerning PGRL1, we could not recognize any correlation between the reduction-based monomerization shift and the increases in ETR II or ETR I, irrespective of light intensities.

Fig. 5.

Relationships between the electron transport efficiency and protein redox state during dark to low light (60 μmol photons m⁻² s⁻¹) or high light (360 μmol photons m⁻² s⁻¹) transitions. The reduction levels of CF₁-γ, FBPase, SBPase and PGRL1 (mean ± SD; n = 3) were plotted against ETR II and ETR I (mean ± SD; n = 5). For the data of FBPase and SBPase, the regression lines were shown with the R² values and slopes (× 10³ value).

Discussion

The photosynthetic electron transport chain in the thylakoid membrane produces reducing power using light energy. The Fd/Trx pathway then transfers the reducing power to several redox-sensitive proteins in chloroplasts, finally controlling chloroplast functions in a light-dependent manner. A series of these reactions have been accepted as the basic mechanism of chloroplast redox regulation (Buchanan 2016). However, our understanding of its dynamics from an ecophysiological viewpoint remains insufficient. In particular, the connection between the electron transport chain and protein redox states in the context of working efficiency requires further clarification. In this study, we addressed this issue with a focus on the relationship during the photosynthetic induction phase. We also noted the effects of the abundance of FTR, a core component of the Fd/Trx pathway.

A balanced amount of FTR is essential for optimal redox regulation and plant growth

Several studies have already focused on Arabidopsis FTR mutants. First, the virus-induced gene silencing of the FTRB gene leads to delayed leaf greening accompanied by alterations in the plastid gene expression (Wang et al. 2014). Second, a point mutation in the FTR catalytic subunit (Cys⁶⁰ to Tyr) causes a marked change in the metabolite profiles, leading to growth defects (Hashida et al. 2018). Third, the clustered regularly interspaced short palindromic repeats (CRISPR)/CRISPR-associated protein 9-based complete knockout of FTR exerts critical inhibitory effects on photosynthesis and autotrophic growth (Yoshida et al. 2022). In the present study, we used a FTR-knockdown ftrb mutant (Yoshida and Hisabori 2016). In this mutant, the T-DNA is inserted into the intron region of the FTRB gene, which lowers the FTR protein abundance while maintaining an appropriate protein size (Fig. 1B, C). The growth was significantly impaired in the ftrb mutant (Fig. 1, Supplementary Fig. S1), but its visible phenotype was apparently mild compared with those of other types of FTR mutants (Wang et al. 2014, Hashida et al. 2018, Yoshida et al. 2022). For example, the chlorophyll content in the ftrb mutant remained at the same level as that in the wild-type plant (Fig. 1A, E). Such weak phenotypes were rather advantageous for this study, which aimed to accurately characterize the relationship between the electron transport efficiency and the redox regulation system, because the potential secondary effects could be minimized. Furthermore, we used previously unreported FTR-overexpressed lines (the 35S::FTRB plants). Notably, the 35S::FTRB plants also exhibited an impaired growth phenotype under certain conditions (e.g. fluctuating light; Supplementary Fig. S1) potentially due to the disturbance of chloroplast redox balance resulting from an excess FTR accumulation (see further). These results suggest that the FTR activity should be maintained at a suitable level for optimal redox regulation and plant growth. We described in a recent study that ACHT2 overexpression in Arabidopsis leads to stunted growth and high NPQ phenotypes, also supporting the importance of a balanced activity of the redox regulation system (Yokochi et al. 2021).

ATP synthase is sensitively reduced by light, even at low electron transport efficiency

Using the FTR-knockdown ftrb mutant and the FTR-overexpressed 35S::FTRB plants, we assessed the relationship between the electron transport efficiency and redox states of individual proteins during the photosynthetic induction phase. In the ftrb mutant, the induction of photosynthetic electron transport reactions was impaired, as evidenced by lowered ETR II and ETR I, at all light intensities examined here (Fig. 4). In contrast, the ATP synthase CF₁-γ subunit displayed a rapid reduction pattern after exposure to light, irrespective of FTR abundance or light intensity (Figs. 2, 3). Consequently, CF₁-γ was constantly maintained in the highly reduced state under light conditions even when the electron transport reactions were not fully induced (Fig. 5). Although earlier studies have also demonstrated rapid reduction of CF₁-γ (e.g. Konno et al. 2012; Yoshida and Hisabori 2018), the underlying mechanisms remain unclear. We previously found that CF₁-γ could be partially reduced even when the linear electron transport was blocked by DCMU (Yoshida et al. 2014) or the Fd/Trx pathway was genetically disrupted (Yoshida et al. 2022). The present study indicated less dependency of CF₁-γ redox regulation on electron transport efficiency. Taken together, the CF₁-γ redox regulation may be supported not only by the Fd/Trx pathway but also by other redox pathways that are not directly coupled to the electron transport reactions. In fact, NTRC was suggested as an alternative redox regulator of CF₁-γ under low-light conditions (Carrillo et al. 2016, Nikkanen et al. 2016). However, we reported that CF₁-γ reduction under light conditions could be partly maintained even when both the Fd/Trx pathway and the NTRC pathway were inactive, raising the possibility of another unidentified pathway for transferring reducing power to CF₁-γ (Yoshida et al. 2022). Further studies would thus be required to determine the molecular mechanism of the redox regulation of chloroplast ATP synthase.

FBPase and SBPase are reduced in proportion to increased electron transport efficiency

The Calvin–Benson cycle enzymes FBPase and SBPase showed a gradual shift to the reduced active state after exposure to light (Figs. 2, 3). Plotting their reduction levels against ETR II and ETR I revealed positive linear relationships (Fig. 5). These results underscore the strong association between electron transport efficiency and the redox states of FBPase and SBPase. This correlation might be closely related to the fact that the Fd/Trx pathway is the sole pathway to transmit reducing power to FBPase and SBPase, but not to ATP synthase (Yoshida et al. 2022).

How are the electron transport efficiency and the redox states of FBPase and SBPase interconnected? The reduction of FBPase and SBPase was delayed in the ftrb mutant (Figs. 2, 3), which was accompanied by lower ETR II and ETR I (Fig. 4). In this mutant, the photosynthetic electron transport chain was more reduced, as reflected by higher 1—qL and Y (NA) values (Fig. 4). Even in 35S::FTRB plants, the partial impairment of FBPase and SBPase reduction (Fig. 3) and moderate alterations in the electron transport properties (e.g. higher NPQ and lower ETR I; Fig. 4) were evident under high-light conditions, although the mechanisms underlying these observations were unclear. Based on these data, it is reasonable to consider that the lowered activation of FBPase and SBPase and the resulting downregulation of Calvin–Benson cycle suppress the turnover of ATP and NADPH, thereby limiting electron transport reactions. Of note, the relationships between the electron transport efficiency and protein redox states were variable between the light intensities (i.e. low- versus high-light conditions) or the proteins (i.e. FBPase versus SBPase) (Fig. 5), suggesting that some additional factors affected their relationships. A possible factor could be an oxidizing force enhanced by accelerated ROS generation along with increased light intensity. As protein redox states are determined by the balance of reduction and oxidation rates, an enhanced oxidizing force under high-light conditions might lower the apparent reduction level more largely. An oxidizing force might also lower the reduction level of SBPase more susceptibly compared to that of FBPase. Indeed, SBPase (but not FBPase) was identified as one of the hydrogen peroxide-sensitive enzymes (Muthuramalingam et al. 2013). Furthermore, the redox state of SBPase might have a larger impact on the electron transport efficiency, given that SBPase has been thought to catalyze the rate-limiting step of the Calvin–Benson cycle (e.g. Driever et al. 2017). It is thus highly plausible that the Calvin–Benson cycle enzymes exert different redox dynamics, which differently affects electron transport efficiency. Further studies are warranted to elucidate this complex interplay in detail.

What are the mechanisms and consequences of PGRL1 redox regulation?

Concerning PGRL1, no consistent understanding of redox-based conformation changes has yet been established. We observed two distinct forms of PGRL1 corresponding to a monomer and a disulfide-linked complex (Figs. 2, 3). The primary candidate for the interaction partner in the latter form appears to be Trx-m4 (Okegawa and Motohashi 2020), although this subject remains controversial (Naranjo et al. 2021). Some studies have also reported another form of PGRL1, namely a disulfide bond-linked homodimer (Hertle et al. 2013, Wolf et al. 2020, Naranjo et al. 2021). Therefore, concluding how PGRL1 changes its conformation based on its redox state would be difficult at this stage. Future studies using mass spectrometry might provide a direct answer to this question. However, it seems to be a common observation that PGRL1 rapidly (<1 min) undergoes reductive monomerization after exposure to light (Figs. 2, 3; see also Okegawa and Motohashi 2020, Wolf et al. 2020, and Naranjo et al. 2021). Importantly, this response of PGRL1 was similarly observed irrespective of light intensities (Figs. 2, 3) and did not correlate with electron transport efficiency variations (Fig. 5). These data suggest that the PGRL1 complex may be reduced in a manner that is not tightly coupled to electron transport reactions. It may be worth considering that, in addition to the Fd/Trx pathway, the NTRC pathway serves to reduce PGRL1, although no direct interaction between NTRC and PGRL1 has been identified so far (Yoshida and Hisabori 2016, Nikkanen et al. 2018, Gonzalez et al. 2019).

Finally, another unanswered question is how PGRL1 redox regulation affects the activity of CET. The CET contributes to the formation of the proton motive force across the thylakoid membrane, leading to NPQ induction (Yamori and Shikanai 2016). Wolf et al. (2020) proposed that NPQ induction/relaxation during dark to light transitions is, at least partly, associated with the redox-based dissociation/reformation of the PGRL1 complex. In contrast, Naranjo et al. (2021) described that high NPQ phenotypes in NTRC-deficient plants are not accompanied by significant changes in the PGRL1 redox states. In the present study, a large fraction of PGRL1 was kept in the reduced monomer state under both low- and high-light conditions (Figs. 2, 3). Nevertheless, NPQ induction patterns largely differed under the two light conditions (Fig. 4). Actually, the dynamic range of the PGRL1 redox state was much lower than that of NPQ (Supplementary Fig. S3). Our data thus suggest that the redox state of PGRL1 is not a major determinant of CET activity. This may be related to the recent idea that even PGRL1 itself is not essential for CET activity (Ruhle et al. 2021). Further studies are required to elucidate the consequences of PGRL1 redox regulation.

Conclusion

We described divergent protein redox dynamics and their relationship with electron transport efficiency in chloroplasts. FBPase and SBPase are converted to the reduced active state synchronously with the induction of electron transport reactions. Their redox behaviors are thought to be advantageous for the coordinated operation of the Calvin–Benson cycle with electron transport. In contrast, CF₁-γ is converted to the reduced state before the full induction of electron transport reactions. Such redox behavior may favor the rapid activation of ATP synthesis required for several biosynthetic reactions in chloroplasts. It should be noted that ATP synthase regulation is also important for adjusting the proton motive force across the thylakoid membrane [see Shikanai (2024) for the latest review]. Elucidating overall regulatory mechanisms to optimize ATP synthase activity remains an important challenge. We also observed a transient shift in the PGRL1 redox state with no correlation to electron transport efficiency. Thus, the heterogeneous working dynamics of the redox regulation system are evident; however, to understand it more comprehensively, the scope of the present analysis should be extended to other redox-sensitive proteins. New approaches for analyzing protein redox changes at a proteome scale provides valuable clues for this issue (Zimmer et al. 2021, Giese et al. 2023, Huang et al. 2023).

Materials and Methods

Plant materials and growth conditions

Arabidopsis thaliana ecotype Columbia-0 was used as the wild-type plant. The T-DNA-inserted ftrb homozygous mutant (GK-686B09) was prepared in our previous study (Yoshida and Hisabori 2016). Transgenic plants overexpressing FTR were generated in this study (see further). Plants were grown in soil in a controlled growth chamber under specified conditions (short-day conditions: 130–140 μmol photons m⁻² s⁻¹, 22°C, 8 h day/16 h night; long-day conditions: 80–90 μmol photons m⁻² s⁻¹, 22°C, 16 h day/8 h night; continuous light conditions: 80–90 μmol photons m⁻² s⁻¹, 22°C). For fluctuating light treatments, the light intensity during the day period was changed every 15 min between 60–70 and 350–370 μmol photons m⁻² s⁻¹ at 22°C and under a 16 h day/8 h night cycle.

Overexpression of the FTR catalytic subunit

The full-length FTRB (At2g04700)-coding region was amplified by PCR and inserted into the pRI 201-AN vector (Takara, Kusatsu, Japan). Arabidopsis plants were transformed with the resulting plasmid using the Agrobacterium-mediated floral dip method (Clough and Bent 1998). Plants showing kanamycin resistance and high expression of the FTR catalytic subunit (examined by immunoblotting analysis) were selected from the T₂ generation. Among them, plants in which all progeny seeds showed kanamycin resistance were isolated as homozygous transgenic plants.

Determination of the protein redox state

A detailed protocol to determine the protein redox state in vivo can be found in our previous paper (Yoshida and Hisabori 2019). Briefly, the plants were directly frozen using liquid nitrogen under the indicated conditions. The extracted proteins were labeled with the specific thiol-modifying reagent 4-acetamido-4′-maleimidylstilbene-2,2′-disulfonate (Invitrogen, Carlsbad, CA, USA). This reagent has a molecular mass of 536.44 and thereby lowers the protein mobility on the nonreducing SDS–PAGE, allowing for band-shift-based determination of the protein redox state in the subsequent immunoblotting analysis. We used antibodies against the ATP synthase CF₁-γ subunit, FBPase, SBPase and PGRL1 to detect each protein as described previously (Yoshida et al. 2014, 2015, Yoshida and Hisabori 2016).

Measurements of chlorophyll fluorescence and P700 absorbance change

The chlorophyll fluorescence and P700 absorbance change around 830 nm were measured simultaneously using a Dual-PAM/F (Walz, Effeltrich, Germany) with the intact leaf. A saturating pulse of red light (800 ms, >5,000 μmol photons m⁻² s⁻¹) was applied to calculate several parameters. After measuring F_v/F_m and the maximal P700 absorbance change in the dark-adapted state, the leaf was irradiated with actinic light (red light) at several intensities (60, 120 or 360 μmol photons m⁻² s⁻¹). A saturating pulse was then applied every 10 s. Each parameter, including 1 − qL, NPQ, Y (ND) and Y (NA), was calculated according to Kono et al. (2014). The ETR II and ETR I were calculated as 0.5 × absI × [light intensity (μmol photons m⁻² s⁻¹)] × [Y (II) or Y (I)], where 0.5 is the fraction of the absorbed light allocated to PSII or PSI, and absI is the absorbed irradiance taken as 84% of the incident irradiance (Yamori et al. 2020). Y (II) and Y (I) denote the operating efficiencies of PSII and PSI, respectively.

Chlorophyll content

The chlorophyll content was determined after extraction with 80% (v/v) acetone according to Porra et al. (1989).

Statistical analysis

Statistical analyses were performed using Microsoft Excel software for the Student’s t-test and SPSS 12.0 J software (SPSS Inc., Chicago, IL, USA) for the Tukey–Kramer multiple comparison test.

Supplementary Data

Supplementary Data are available at PCP online.

Data Availability

The data underlying this article will be shared upon reasonable request to the corresponding author.

Funding

The Japan Society for the Promotion of Science KAKENHI (21H02502 to K.Y. and T.H., 22K19130 to K.Y., 23H02498 to K.Y. and 23H04961 to K.Y.).

Disclosures

The authors declare no conflicts of interest.