Proton gradient across the chloroplast thylakoid membrane governs the redox regulatory function of ATP synthase

Chloroplast ATP synthase (CFoCF1) synthesizes ATP by using a proton electrochemical gradient across the thylakoid membrane, termed ΔμH+, as an energy source. This gradient is necessary not only for ATP synthesis but also for reductive activation of CFoCF1 by thioredoxin, using reducing equivalents produced by the photosynthetic electron transport chain. ΔμH+ comprises two thermodynamic components: pH differences across the membrane (ΔpH) and the transmembrane electrical potential (ΔΨ). In chloroplasts, the ratio of these two components in ΔμH+ is crucial for efficient solar energy utilization. However, the specific contribution of each component to the reductive activation of CFoCF1 remains unclear. In this study, an in vitro assay system for evaluating thioredoxin-mediated CFoCF1 reduction is established, allowing manipulation of ΔμH+ components in isolated thylakoid membranes using specific chemicals. Our biochemical analyses revealed that ΔpH formation is essential for thioredoxin-mediated CFoCF1 reduction on the thylakoid membrane, whereas ΔΨ formation is nonessential.

Chloroplast ATP synthase (CF o CF 1 ) synthesizes ATP by using a proton electrochemical gradient across the thylakoid membrane, termed DmH + , as an energy source.This gradient is necessary not only for ATP synthesis but also for reductive activation of CF o CF 1 by thioredoxin, using reducing equivalents produced by the photosynthetic electron transport chain.
DmH + comprises two thermodynamic components: pH differences across the membrane (DpH) and the transmembrane electrical potential (DJ).In chloroplasts, the ratio of these two components in DmH + is crucial for efficient solar energy utilization.However, the specific contribution of each component to the reductive activation of CF o CF 1 remains unclear.In this study, an in vitro assay system for evaluating thioredoxinmediated CF o CF 1 reduction is established, allowing manipulation of DmH + components in isolated thylakoid membranes using specific chemicals.Our biochemical analyses revealed that DpH formation is essential for thioredoxin-mediated CF o CF 1 reduction on the thylakoid membrane, whereas DJ formation is nonessential.
In chloroplasts, solar energy is converted to chemical energy via the photosynthetic electron transport chain in the thylakoid membrane.During this process, reducing equivalents are stored as NADPH, and a proton electrochemical gradient (DmH + ) forms across the membrane.Termed the protonmotive force, DmH + drives chloroplast F o F 1 -ATP synthase (CF o CF 1 ) to catalyze ATP synthesis (1,2).Therefore, CF o CF 1 is a pivotal enzyme for energy conversion in photosynthesis.Moreover, its activity undergoes precise regulation to maintain efficient chemical energy production under varying light conditions.
DmH + comprises two thermodynamic components: pH differences across the membrane (DpH) and the transmembrane electrical potential (DJ).Both components were reported to be kinetically equivalent in F o F 1 from thermophilic bacteria (3), and early investigations into CF o CF 1 revealed that both contribute to ATP synthesis by this enzyme complex (4-6).Notably, CF o CF 1 exhibits thiol-based redox regulation, distinguishing it among all F o F 1 enzymes across species (7,8), with this regulation system closely linked to DmH + .The central axis of CF o CF 1 , the g subunit (CF 1 -g), harbors a pair of redoxactive cysteines (Cys 199 and Cys 205 in Spinacia oleracea) (9,10).To activate CF o CF 1 , chloroplast thioredoxin (Trx) reduces this Cys pair, using reducing equivalents from photosynthetic electron transport reactions (7,11).In this Trx-dependent activation process, DmH + formation is essential for CF 1 -g reduction (12)(13)(14).Conversely, we previously revealed that DmH + dissipation stimulates CF 1 -g oxidation (15,16), facilitated by chloroplast oxidizing factors, such as Trx-like proteins (16)(17)(18).Consequently, CF o CF 1 activity is finely tuned to activate only during photosynthetic conditions, promptly deactivating in the dark when DmH + formation ceases.However, the relationship between CF o CF 1 redox regulation and the DmH + components, i.e., DpH or DJ, remains unexplored.
In chloroplasts, DpH plays an important role in regulating photosynthetic performance.Specific ion transporters, such as two-pore potassium channel three and voltage-dependent chloride channel 1, present in the thylakoid membrane contribute to the movement of counter ions for dissipating DJ (19,20), maintaining a high DpH relative to DJ. Acidification of the lumen (i.e., DpH formation) regulates the electron transfer activity of the cytochrome b 6 f complex (21,22).Additionally, DpH serves as a key signal for initiating nonphotochemical quenching (NPQ), which functions as a photoprotection mechanism, especially the qE component (23).Therefore, green plants must maintain a balanced DpHto-DJ ratio within DmH + for safe light utilization.
This study focuses on the thermodynamic components of DmH + required for CF 1 -g reduction by Trx.Our investigation involved manipulating DmH + bias in isolated spinach thylakoids using specific chemicals.We also established an in vitro assay system using freshly prepared thylakoid membranes to facilitate CF 1 -g reduction by Trx on the membrane.

Results and discussion
Establishing biased DmH + conditions in the thylakoid membrane using ionophores thylakoid membrane or neither form.Using this approach, we aimed to discern differences in the contributions of DpH and DJ to CF 1 -g redox regulation.Specifically, we used the fluorescent reagents 9-amino-6-chloro-2-methoxyacridine (ACMA) and 8-anilinonaphthalene-1-sulfonic acid (ANS) to monitor DpH and DJ formation on the thylakoid membrane, respectively (Fig. 1).ACMA fluorescence intensity decreases upon protonation (24,25), whereas ANS fluorescence increases upon potentiated membrane binding (26,27).The addition of the artificial electron mediator 1-methoxy-5methylphenazinium methylsulfate (PMS) to the thylakoid membrane induces DmH + formation across the membrane under light due to pseudocyclic electron transport around photosystem I (28).We then monitored DpH formation by observing the ACMA fluorescence decrease (120 s; Fig. 1A,  Control).Subsequently, red light irradiation further enhanced DpH formation (300 s; highlighted area of the graph in Fig. 1), dependent on light intensity (14-130 mmol photons m −2 s −1 ).DpH remained constant during red light irradiation and dissipated upon its cessation (900 s), finally returning to initial levels upon FCCP addition (1200 s).Similar temporal events were observed for DJ formation using ANS (Fig. 1B, Control), with DJ gradually declining under low light conditions (14 and 43 mmol photons m −2 s −1 ).
The effects of FCCP, valinomycin, and nigericin on DpH and DJ formation under the aforementioned light conditions were then examined.FCCP selectively permeates H + across the membrane, abolishing both DpH and DJ (i.e., DmH + ).Pretreatment of the thylakoid membrane with FCCP almost abolished fluorescence changes in both ACMA and ANS (Fig. 1, A and B examined light conditions, albeit to a lesser extent (Fig. 1B, +1 mM nigericin).These results highlight the successful identification of varying ionophore actions in the thylakoid membrane as well as the conditions distinguishing both DpH and DJ.
Historically, the electrochromic shift in endogenous carotenoid absorbance at 515 nm (DA 515 ) has been used to monitor the transmembrane DJ level (29).DA 515 can be observed by briefly exposing isolated thylakoids to actinic light flashes.However, we employed ACMA and ANS in this study.This is because these fluorescent reagents have a history of being used in combination with ionophores and uncoupler to measure the enzymatic activity of ATP synthase and other ion transporters in in vitro studies, and their effectiveness has been confirmed (16,(30)(31)(32).
Trx-mediated CF 1 -g reduction on the thylakoid membrane fails to occur without DpH formation Using the abovementioned conditions to distinguish DpH and DJ (Fig. 1), we performed Trx-mediated reduction assays of CF 1 -g on the thylakoid membrane (Fig. 2).In vitro reduction experiments involving CF 1 -g were performed using the same concentrations of thylakoid membrane [5 mg chlorophyll (Chl)/ml] and 1-methoxy PMS (2 mM) shown in Figure 1.The redox state of CF 1 -g was determined using 4-acetamido-4 0maleimidylstilbene-2,2 0 -disulfonic acid (AMS), as described previously (16).CF 1 -g is known to be reduced by Trx-f, a major isoform of chloroplast Trx proteins (33,34).This reduction process requires prior DmH + formation across the thylakoid membrane (7,13,14).As expected, CF 1 -g on the To further explore the relationship between DmH + and CF 1g reduction, we used higher concentrations of the thylakoid membrane (50 mg Chl/ml) and 1-methoxy PMS (100 mM) under more intense light conditions (600-650 mmol photons m −2 s −1 ), following our previous study (16,34) (Fig. 3).Although CF 1 -g was not reduced in the dark [Fig.3, A and B, Dark (−PMS)], it was reduced by approximately 80% when 1 mM Trx and 100 mM DTT were added in the light (Fig. 3, A and B, Control).When FCCP or nigericin was added to the thylakoid membrane beforehand, CF 1 -g was not reduced, similar to dark conditions (Fig. 3, A and B, +5 mM FCCP, +5 mM nigericin) as well as the results shown in Figure 2.Both FCCP and nigericin dissipated DpH formation across the thylakoid membrane, as confirmed via fluorescence measurements (Fig. 1A, +2 mM FCCP, +1 mM nigericin).However, even in the presence of valinomycin, CF 1 -g was reduced by approximately 70% (Fig. 3, A and B, +5 mM valinomycin).Under these experimental conditions, Trx-f was almost completely reduced when sufficient amounts of DTT were added (Fig. 3, C and D).We also examined whether FCCP and ionophores affect the Trx-dependent reduction of other target enzymes by testing FBPase reduction via Trx-f in the presence of FCCP or ionophores.Notably, FBPase is the major target enzyme of Trx-f in chloroplasts (35,36).As shown in Figure 4, Trx-f reduced FBPase efficiently even in the presence of FCCP or ionophores in the reaction mixture, implying that the inhibited CF 1 -g reduction shown in Fig- ures 2 and 3 was due to DpH dissipation caused by FCCP or nigericin.Hence, DpH but not DJ formation across the thylakoid membrane was required for CF 1 -g reduction.The varying results for valinomycin with different thylakoid membrane concentrations may be attributed to differences in membrane stability under the respective experimental conditions.Higher membrane concentrations may maintain stability and reduce H + leakage.
Our results raise an important question: does DJ have any effect on CF 1 -g reduction?As shown in Figure 2, CF 1 -g reduction did not occur with pretreatment of either nigericin or valinomycin but was observed in the control experiment, especially under 130 mmol photons m −2 s −1 .These results imply that DJ supports CF 1 -g reduction when DpH is low.Notably, the extent of DpH with added was slightly lower than that under control conditions (Fig. 1A, Control, +1 mM valinomycin).However, CF 1 -g was reduced when valinomycin was added under more intense light conditions (Fig. 3, A and B, +5 mM valinomycin).Thus, when sufficient DpH forms across the thylakoid membrane, CF 1 -g reduction occurs regardless of DJ formation, i.e., the extent of DpH governs the CF 1 -g reduction process.This reduction process observed in this study under different light conditions was illustrated, along with the effect of ionophores and uncoupler used (Fig. 5).Alkaline conditions near the surface of the thylakoid membrane may be favorable for the dithiol-disulfide exchange reaction between CF 1 -g and Trx.However, this detailed molecular mechanism is not yet clear, and further studies are required.As DpH across the thylakoid membrane induces NPQ, it is considered crucial in plant physiology, whereas DJ is used exclusively to regulate DpH.
Overall, the ability to activate CF o CF 1 reductively via Trx under fluctuating light conditions without relying on DJ formation is likely advantageous for plants.

Experimental procedures
Preparation of thylakoid membranes from spinach leaves Thylakoid membranes were prepared from spinach (S. oleracea) as previously described (34) but with slight modifications.Fresh market spinach was washed thoroughly and left overnight in the dark at 4 C. Harvested leaves (approximately 10 g fresh weight) were homogenized three times for 3 s in a mixer with 200 ml of grinding buffer [50 mM Tricine-NaOH (pH 7.5), 0.4 M sucrose, 5 mM MgCl 2 , 10 mM NaCl, and 50 mM KCl].The homogenate was filtered through four layers of gauze and centrifuged at 3,000g and 4 C for 10 min.The pellet was then resuspended in the grinding buffer and centrifuged at 300g and 4 C for 1 min, after which the supernatant was collected and centrifuged at 3,000g and 4 C for 10 min.After the abovementioned washing step was repeated once, the resulting pellet was resuspended in the grinding buffer to achieve a Chl concentration of 0.5 mg/ml.The preparation was kept in the dark on ice for at least 1 h before the assay.

Monitoring the formation of proton gradients and membrane potential gradients across thylakoid membranes
We measured DpH and DJ across the thylakoid membrane using ACMA and ANS, respectively.The grinding buffer for thylakoid preparation was used as the reaction mixture, with the reaction performed at 25 C. Red light irradiation at 660 nm induced the formation of these thermodynamic gradients across the thylakoid membrane.Before use, ACMA was solubilized at 30 mg/ml in 100% ethanol and stored at −80 C.
ANS was prepared as a 10 mM solution in 10% DMSO and stored at room temperature.The emitted fluorescence of ACMA (l ex = 410 nm, l ex = 480 nm) and ANS (l ex = 330 nm, l ex = 455 nm) was measured using a FP-8500 spectrofluorometer (Jasco).
Stored ACMA or ANS solution (20 ml) and 50 ml of 200 mg Chl/ml thylakoid membrane were added to 1910 ml of the grinding buffer in a grass cuvette and left to stand for stabilization of the fluorescence signal.At 120 s after initiating the measurement, 20 ml of 200 mM 1-methoxy PMS was added to the mixture.The final concentrations in the cuvette were 5 mg Chl/ml of thylakoid membranes, 2 mM 1-methoxy PMS, and 0.3 mg/ml ACMA (in 0.1% ethanol) or 100 mM ANS (in 0.01% DMSO).Subsequently, the reaction mixture was irradiated with red light from a direction perpendicular to the cuvette using a light-emitting diode (LED) at 300 s and terminated at 900 s.The photon flux density of the red light is shown in 1.At 1200 s, 2 ml of 2 mM FCCP was added to the mixture to confirm whether DmH + was dissipated compared with the initial condition.The initial fluorescence intensity of each trace was normalized to 100% using the average of the data from approximately 30 to 90 s, i.e., when fluorescence intensity was relatively stable.

Recombinant protein preparation
The recombinant proteins used in this study, spinach Trx-f and Arabidopsis thaliana FBPase, were prepared as described previously (34,37).Protein concentrations were determined using a BCA protein assay (Pierce).

In vitro assay of Trx-mediated CF o CF reduction
For the reduction assay, the grinding buffer for thylakoid membrane preparation was used as the reaction mixture, and 50 ml of 200 mg Chl/ml thylakoid membrane were added to 1730 ml of the grinding buffer and incubated for 1 min.Next, 20 ml of 200 mM 1-methoxy PMS was added to the mixture.The final concentrations in the mixture were 5 mg Chl/ml thylakoid membranes, 2 mM 1-methoxy PMS, 100 mM DTT, and 1 mM Trx-f.This mixture was irradiated with red light at 660 nm for 5 min using an LED to initiate DmH + formation across the thylakoid membrane.The photon flux density of the red light is shown in Figure 2. Similar experiments were performed using higher thylakoid membrane concentrations (final concentration, 50 mg Chl/ml) and 1-methoxy PMS (final concentration, 100 mM) under more intense light conditions (600-650 mmol photons m −2 s −1 ).Following the in vitro assay, proteins were precipitated using 10% (w/v) trichloroacetic acid to stop the reduction reaction.

In vitro assay of Trx-mediated FBPase reduction
For FBPase reduction, a medium containing 50 mM Tris-HCl (pH 7.5) and 50 mM NaCl was used, with the reaction performed at 25 C. Protein and reducing agent concentrations as well as reaction times are described in the Figure 4 legend.

Determination of the protein redox state
The protein redox state was determined by labeling free thiols with AMS and employing sodium dodecyl sulfatepolyacrylamide gel electrophoresis (for Trx-f and FBPase) or immunoblotting (for CF 1 -g), as described previously (34,37).The antibody against CF 1 -g were prepared using recombinant Arabidopsis CF 1 -g (His-tagged at the C terminus) as an antigen, and its specificity is indicated in our former paper (38).Chemiluminescence was detected using horseradish peroxidase-conjugated secondary antibodies and ECL Prime (Cytiva) and visualized on a LAS 3000 Mini Imaging System (Fuji Film).The resultant band intensities were quantified using ImageJ.The reduction level was calculated as the ratio of the reduced form to the total form.The data in Figures 2-4 were statistically analyzed using one-way analysis of variance and Tukey's honest significance differences test (p < 0.05).Statistical analyses were performed using an online calculator at iCalcu.com(https://www.icalcu.com/stat/anova-tukey-hsdcalculator.html).

H
+ translocation across the membrane concurrently generates DpH and DJ.Through the application of ionophores, such as nigericin, valinomycin, or the uncoupler FCCP, we can establish conditions where either DpH or DJ forms in the Present Address for Ken-ichi Wakabayashi: Faculty of Life Sciences, Kyoto Sangyo University, Kamigamo-motoyama, Kita-ku, Kyoto 603-8555, Japan.* For correspondence: Toru Hisabori, thisabor@res.titech.ac.jp.

Figure 1 .
Figure 1.Ionophore effects on light-induced DmH + in the thylakoid membrane.Observations of DpH and DJ across the thylakoid membrane induced by red light irradiation.(A) DpH measurement using ACMA.Thylakoid membranes (5 mg chlorophyll (Chl)/ml) were incubated with 0.3 mg/ml ACMA, and ACMA fluorescence intensity (l ex = 410 nm, l ex = 480 nm) was monitored.(B) DJ measurement using ANS.Thylakoid membranes (5 mg Chl/ml) were incubated with 100 mM ANS, and ANS fluorescence intensity (l ex = 330 nm, l ex = 455 nm) was monitored.Gray sections indicate dark conditions, whereas highlighted sections indicate red light irradiation periods.Open and closed triangles represent the addition of 2 mM 1-methoxy PMS and 1 mM FCCP, respectively.

Figure 2 .
Figure 1 (14-130 mmol photons m −2 s −1 ), we observed a light intensity-dependent reduction of CF 1 -g (Fig. 2B, Control).Thus, the CF 1 -g reduction level was controlled by the extent of DmH + across the thylakoid membrane.Next, we investigated the effects of an uncoupler and ionophores under the conditions shown in Figure 1 (2 mM FCCP, 1 mM valinomycin, or

Figure 3 .
Figure 3.In vitro CF 1 -g reduction using high-concentration thylakoid membranes.A and C, determination of the thylakoid CF 1 -g and Trx-f redox states.CF 1 -g in the thylakoid membrane (50 mg Chl/ml) was reduced by 1 mM Trx-f and 100 mM DTT in the presence of each ionophore under 600 to 650 mmol photons m −2 s −1 for 5 min.After free thiol modification with AMS, proteins underwent nonreducing SDS-PAGE, and the redox state was visualized via western blotting with anti-CF 1 -g antibodies (A) or Coomassie Brilliant Blue staining (C).Ox, oxidized form; Red, reduced form.B and D, quantification of the CF 1 -g and Trx-f reduction levels for the data shown in (A) and (C), respectively.Data represent means ± SDs (n = 3).Different letters indicate significant differences (p < 0.05; one-way ANOVA and Tukey's HSD test).

Figure 4 .
Figure 4. FBPase reduction by Trx-f in the presence of ionophores.A, FBPase (2 mM) was incubated with 1 mM Trx-f and 100 mM DTT for 30 min.After free thiol modification with AMS, proteins underwent nonreducing SDS-PAGE followed by Coomassie Brilliant Blue staining.B, quantification of the FBPase redox state for the data shown in (A).Data represent means ± SDs (n = 3).Different letters indicate significant differences (p < 0.05; one-way ANOVA and Tukey's HSD test).

Figure 5 .
Figure 5.An overview of the relationship between the redox regulation of CF 1 -g and DpH formation.The formation of DmH + when CF 1 -g is reduced by Trx under low light (A) or high light (B) conditions are illustrated.Triangles represent the extent of DpH across the thylakoid membrane, and dashed triangles represent the extent of DJ across the thylakoid membrane.