Contents lists available at ScienceDirect

BBA - Bioenergetics

journal homepage: www.elsevier.com/locate/bbabio

Global spectroscopic analysis to study the regulation of the photosyntheticproton motive force: A critical reappraisal☆

Guillaume Allorenta, Martin Byrdinb, Luca Carrarettoc, Tomas Morosinottoc, Ildiko Szaboc,⁎,Giovanni Finazzia,⁎

aUniversité Grenoble Alpes (UGA), Centre National de la Recherche Scientifique (CNRS), Commissariat à l'Energie Atomique et aux Energies Alternatives (CEA), InstitutNational de la Recherche Agronomique (INRA), Laboratoire de Physiologie Cellulaire et Végétale, UMR 5168, Institut de Biosciences et Biotechnologie de Grenoble (BIG),CEA-Grenoble, 38000 Grenoble, FrancebUniversité Grenoble Alpes (UGA), Commissariat à l'Energie Atomique et aux Energies Alternatives (CEA), Centre National de la Recherche Scientifique (CNRS), Institutde Biologie structurale (IBS), CEA-Grenoble, 38000 Grenoble, Francec Department of Biology, University of Padova, 35100 Padova, Italy

A R T I C L E I N F O

Keywords:Electrochromic shiftProton motive forceIon channelsPhotosynthesis

A B S T R A C T

In natural variable environments, plants rapidly adjust photosynthesis for optimum balance between photo-chemistry and photoprotection. These adjustments mainly occur via changes in their proton motive force (pmf).Recent studies based on time resolved analysis of the Electro Chromic Signal (ECS) bandshift of photosyntheticpigments in the model plant Arabidopsis thaliana have suggested an active role of ion fluxes across the thylakoidmembranes in the regulation of the pmf. Among the different channels and transporters possibly involved in thisphenomenon, we previously identified the TPK3 potassium channel. Plants silenced for TPK3 expression dis-played light stress signatures, with reduced Non Photochemical Quenching (NPQ) capacity and sustained an-thocyanin accumulation, even at moderate intensities. In this work we re-examined the role of this protein inpmf regulation, starting from the observation that both TPK3 knock-down (TPK3 KD) and WT plants displayenhanced anthocyanin accumulation in the light under certain growth conditions, especially in old leaves. Wethus compared the pmf features of young “green” (without anthocyanins) and old “red” (with anthocyanins)leaves in both genotypes using a global fit analysis of the ECS. We found that the differences in the ECS profilemeasured between the two genotypes reflect not only differences in TPK3 expression level, but also a modifiedphotosynthetic activity of stressed red leaves, which are present in a larger amounts in the TPK3 KD plants.

1. Introduction

In chloroplasts, photosynthesis generates a proton motive force(pmf) thanks to the coupling between electron transfer in membranes(thylakoids) embedded complexes and proton release in the inner(lumen) compartment. The pmf not only comprises a H+ gradient(ΔpH), but also an electric field (ΔΨ), negative on the stromal side ofthe thylakoids. The ΔΨ is generated by electron flow, due to theasymmetric location of the electron donors and acceptors of the dif-ferent photosynthetic complexes. All the electron donors (P680 and P700,i.e. the primary donors to Photosystem II and I, respectively, plus theQO/QP site of plastoquinol oxidation in cytochrome b6f) are facing theluminal side of the thylakoids. All the electron acceptors (the QA/QB

plastoquinones in PSII, the phylloquinones/iron-sulphur clusters of PSI

and the Qi/Qn site of the cyt b6f complex) are facing the stromal side ofthe membranes [1]. Thus, light driven photosynthesis results in theaccumulation of negative changes on the stromal side of the thylakoids,and of positive charges on the luminal side. Both components of the pmfequally contribute to ATP synthesis [2] for CO2 assimilation and othermetabolic activities, although the presence of a minimum ΔΨ is re-quired for ATP generation [3]. The ΔpH and ΔΨ also play more specificroles. The proton gradient slows down the rate of electron transfer atthe level of the cytochrome b6f complex. This phenomenon, known asthe “photosynthetic control” [4], allows keeping the primary electrondonor to P700 in the oxidized state in the light [5], possibly to protectPSI against photoinhibition [6]. The ΔpH also modulates photoprotec-tion of PSII, by enhancing thermal dissipation of light absorbed in ex-cess with respect to the CO2 assimilation capacity, in a process known

https://doi.org/10.1016/j.bbabio.2018.07.001Received 23 January 2018; Received in revised form 3 June 2018; Accepted 2 July 2018

☆ This article is part of a Special Issue entitled 20th European Bioenergetics Conference, edited by László Zimányi and László Tretter⁎ Corresponding authors.E-mail addresses: ildiko.szabo@unipd.fr (I. Szabo), giovanni.finazzi@cea.fr (G. Finazzi).

BBA - Bioenergetics 1859 (2018) 676–683

Available online 05 July 20180005-2728/ © 2018 Elsevier B.V. All rights reserved.

T

as Non Photochemical Quenching (NPQ, reviewed in [7,8]). The ΔΨhas been recently suggested to modulate PSII photodamage, by con-trolling the rate of charge recombination within the complex andtherefore singlet oxygen production by PSII [9]. The presence of a ΔΨalso affects the absorption characteristics of some photosynthetic pig-ments, leading to a shift of their absorption spectrum. This phenom-enon, known as the electrochromic shift (ECS, [2]), can be used tomonitor the pmf in vivo [10].

Because of the different roles of the two pmf components, theirrelative contribution to the pmf must be actively controlled. In mi-tochondria, a consensus exists on the notion that the ΔΨ is the pre-dominant pmf component. Conversely, both the ΔΨ and the ΔpH ac-cumulate in chloroplasts in the light, and the real partitioning of thepmf between its two components is still debated [11,12]. Experimentalstudies [13] and modelling [12] has led to the notion that it is possibleto monitor the ΔpH/ΔΨ partitioning by following the time course ofECS changes during a light to dark transition. Typical kinetic of the ECSduring the onset of illumination and in steady state conditions arepresented in Fig. 1. During the dark to light transition, the ECS displaysa complex kinetic behavior, which likely reflects changes in the turn-over rate of the photosynthetic complex (i.e. in the generation of thepmf by photosynthetic electron flow) and in ion flux (i.e. its dissipationvia either H+

flux in the ATP synthase or counter ions movement [12]).After attainment of steady state, the ECS relaxes in the dark with amultiphase kinetic. After the first relaxation phase (1), a slower ECSincrease is seen (2), which has been previously interpreted as the sig-nature of charge redistribution across membranes [13]. Finally, apseudo steady state point is attained (3). There, the reversible part ofthe ECS signal (4) would correspond to the pmf stored as a ΔpH, whilethe irreversible one would correspond to the ΔΨ [13].

Recent studies have suggested that the presence of ion channels/transporters in the thylakoid membranes modulates the partitioning ofΔpH and ΔΨ under different conditions, thereby affecting the photo-synthetic efficiency (reviewed in [14,15]). They include the potassiumchannel TPK3 [16], the K+/H+ antiporter KEA3 [17–19] and trans-porters of other charged species, including Cl− [20–22], Mn2+ andCa2+ [23,24]. Concerning TPK3, its ion specificity and regulation wasassessed by electrophysiological experiments in planar lipid bilayers,revealing that it is a K+ selective channel belonging to the two-porepotassium channel family, regulated by Ca2+ and pH. Arabidopsisthaliana mutants fully silenced in the TPK3 gene displayed severalphenotypes including altered thylakoid membrane organisation, lowerefficiency of photosynthesis, higher susceptibility to photoinhibitionthan WT plants and enhanced accumulation of anthocyanins already at100 μmol photons m−2 s−1 light intensity [16]. Previous analysis of thekinetics of the ECS relaxation during a light-dark transition suggestedthat the pmf was differently partitioned in this mutant, with an increasein the ΔΨ at the expense of ΔpH [16]. TPK3 was thus proposed to play arole in counterbalancing proton influx into the lumen upon illumina-tion. The change in pmf partitioning would provide a rationale for the

different phenotypes observed in this mutant: a lower ΔpH wouldtranslate in a diminished NPQ capacity, which, along with the othermodifications of photosynthesis, would lead to stress responses like theanthocyanin accumulation [25–27]. Alternatively, the higher ΔΨ [9]could enhance PSII light sensitivity, making mutant plants morestressed even at moderate light intensities.

Recently, the commonly employed procedure to deconvolute theECS signal from other light induced spectral changes in the light, asrequired to evaluate the two components of the pmf, has been ques-tioned [11]. This procedure consists in subtracting the signal measuredeither at one reference wavelength (around 540–550) or at two re-ference wavelengths (at 505 nm and 540–550 nm) from the amplitudeof the ECS at its peak (around 520 nm) [13,16–24]. Therefore, wedecided to re-examine ECS kinetics in the light and during the light todark transition. We compared the “classic” procedure for the ECS dis-entangling with a more detailed spectroscopic study based on theanalysis of 14 different wavelengths in the 500–600 nm range in bothWT and TPK3 knock-down Arabidopsis plants.

2. Methods

Plants (WT and TPK3 antisense plants, from now on TPK3 KD) weregrown at 100 μmol photons m−2 s−1 for 5 weeks with a photoperiod of16 h light/8 h dark. New mutant lines were generated following theprotocol described in [16]. Mutant seeds were first germinated on MS1/2 plates containing kanamycin (50 μg/mL) for selection. Kanamycin-resistant plants were then transferred to soil and tested for post-transcriptional silencing of the TPK3 gene by mRNA extraction from 2-week-old leaves. Transcript analysis of the new clones indicated thatsilencing efficiency was in the range of 50–80% in the newly generatedlines (not shown). Transcript analysis on leaves from plants that wereused for ECS and NPQ measurements was performed as previouslydescribed [16].

In vivo spectroscopic analysis was performed with a JTS-10 spec-trophotometer (Biologic, France) on freshly collected leaves. The de-tection beam was provided by a white LED source filtered throughappropriate interference filters (Supplementary Fig. 1). The measureand reference photodiodes were protected from actinic light by BG 39filters. Changes in the pmf were first evaluated from ECS changesmeasured at 520–545 nm [16] to disentangle this signal from the redoxchanges associated with the cytochrome b6f complex turnover [28,29]and the 535 nm shift that is reflects a modification of the xanthophyllzeaxanthin during the onset (or the relaxation) of NPQ [30,31]. Theamplitude of the ECS signal in the different leaves was normalized tothe signal corresponding to 1 charge separation, to facilitate compar-ison. The latter was estimated as the amplitude of the ECS signalmeasured 150 μs after exposure to a saturating single turnover laserflash, under conditions where PSII is inactive (i.e. upon addition ofsaturating amounts of the PSII inhibitors DCMU and hydroxylamine). Inthese conditions, only PSI is active and therefore the ECS amplitude

Fig. 1. Kinetics of the ECS in Arabidopsis thalianaleaves exposed to 590 μmol photons m−2 s−1.(A) ECS transients during a dark to light exposure.Several phases can be seen, which likely representthe adjustments of the pmf upon illumination (seetext). (B) ECS relaxation after steady state illumina-tion. The ECS relaxation follows different phases: afast relaxation (1) is followed by a slow recovery (2)until a pseudo stationary phase level (3) is reached.This allows estimating the inverted (4) component ofthe ECS. White box: the actinic light was on; blackbox: the actinic light was switched off.

G. Allorent et al. BBA - Bioenergetics 1859 (2018) 676–683

677

corresponds 1 charge separation per photosynthetic electron transferchain [10].

Alternatively, a more complete spectral analysis was performed,measuring the kinetics of the light to dark transition at different wa-velengths from 500 to 600 nm. To this aim, the JTS-10 spectro-photometer was equipped with a rotating filter holder (JBeam Bio,France). By changing the angle between the detection light beam andthe interferential filters it was possible to modify the wavelength of thedetecting beam. Overall, we selected 14 different wavelengths (506 nm,510 nm, 516 nm, 520 nm, 527 nm, 535 nm, 540 nm, 545 nm, 550 nm,554 nm, 559 nm, 565 nm, 570 nm, 573 nm) from 7 filters centredaround 507 nm, 520 nm, 527 nm, 546 nm, 554 nm, 563 nm, 573 nm).The transmission spectra of the different filters were measured with aUSB 2000+ spectrophotometer (Ocean Optics, USA) directly pluggedto the sample holder of the JTS-10 spectrophotometer. Spectra weremeasured before or after the light had crossed green or red leaves. Thefull width at half maximum (FWHM) was variable depending on thewavelengths (Supplementary Fig. 1), and the intensity of the lighttransmitted by the leaves was differently decreased at specific wave-lengths. However, the spectral characteristics of the light transmittedby green and red leaves were similar (Supplementary Fig. 1).

In the case of the experiments presented in Fig. 4, SupplementaryFig. 2 and Supplementary Fig. 3, spectra were obtained measuring the14 wavelengths in one order and then repeating the same sequence inthe reverse temporal order. Leaves were first acclimated to light for5min, and then kinetics of the dark relaxation were measured in thedark for every wavelength. Leaves were re-exposed to light for 2minbetween two consecutive measurements. We checked that this illumi-nation regime was sufficient to restore the amplitude of the ECS signalbetween two consecutive measurements. Kinetics at all wavelengthswere decomposed into a sum of exponentials. The fit procedure imposesthat the lifetimes values must be the same for all wavelengths, whiletheir amplitude values are not fixed during the fit. At the end of thefitting procedure, the amplitudes corresponding to every lifetimes (i.e.the decay associated spectra, DAS) were plotted as a function of thewavelength, to evaluate the spectral features of the different compo-nents of the light to dark relaxation kinetics.

Fluorescence measurements were performed with a Speedzen IIfluorescence imaging setup (JBeam Bio, France, [32]). Measuringpulses were provided by blue LEDS, while actinic light and saturatingpulses were provided by red LEDs of variable intensity. NPQ wasevaluated as (Fm-Fm′)/Fm′, where Fm is the maximum fluorescenceemission measured in dark adapted leaves, and Fm′ is the maximumfluorescence emission measured in the light [33]. Fm and Fm′ wereattained by exposing leaves to short (250ms) saturating light pulses.Electron Transfer Rate was measured as (Fm′-Fs)/Fm′, where Fs is thefluorescence emission measured in the light, i.e. prior to exposure to thesaturating pulse [33].

3. Results and discussion

Data obtained using the newly generated TPK3 KD antisense plantsconfirm that these plants are more prone to develop light stress sig-natures when exposed to moderate light (100 μmol photons m−2 s−1)than their WT counterparts. This is evidenced by the enhanced accu-mulation of anthocyanins in the mutant plants (Fig. 2). On the otherhand, we observed some variability among the leaves of the same plant,with the oldest leaves -located at the periphery of the rosette- dis-playing a more intense red pigmentation. This variability is also seen inthe WT, where “red” leaves were also seen at the periphery of the ro-sette in 5-week old plants grown at 16:8 h light/dark regime, althoughto a lesser extent.

Previous analysis had pinpointed changes in the ECS profile ofmutant plants [16]. Indeed a diminished amplitude of the recoveryphase (phase “2” according to the nomenclature used in Fig. 1) wasfound in mutant leaves when compared to the WT leaves. This finding

suggests that the ΔpH could be diminished in the TPK3 KD mutants.Although located in the same respective positions in the rosettes of thetwo genotypes, mutant leaves were essentially red, while WT leaveswere green in that study [16]. On the other hand, the larger hetero-geneity in the leaf pigmentations observed here (with green leaves inthe centre, red leaves at the periphery of the rosette in both genotypesalthough to different extent), allows to re-evaluate this hypothesis byextending the comparative analysis between the two genotypes to theirdifferent types of leaves.

We found (Fig. 3) that “green”, i.e. young leaves from both WT andTPK3 KD plants display similar ECS profiles. In both genotypes, therelaxation phase of the ECS was followed by a substantial recovery.According to previous interpretation of these kinetics, a substantial ΔpHis thus present in green leaves of both genotypes. On the other hand, theECS recovery was reduced in “red” leaves, especially in the TPK3 KDplants, suggesting that the generation of the ΔpH might be impaired inthese leaves. To better understand this phenomenon, we decided toperform a more detailed spectroscopic analysis. Besides the ECS, otheroptical signal are induced in the light in the same spectral region as theECS. These signals include, for example, the 535 nm shift that is asso-ciated to the presence of zeaxanthin during the development of NPQ[30,31] and/or the redox-induced absorption changes in the550–555 nm region due to the turnover of the cytochrome f moietywithin the cyt b6f complex [28,29]. Since both the ECS and the 535 shiftreflect a modification in the absorption spectrum of specific pigments,changes in the pigment composition could modify their relative re-sponses. As a consequence, the overall dark relaxation kinetic could bemodified, because the two signals do not have necessarily the sameonset and relaxation dynamics.

To test this possibility, we re-evaluated the spectral features of thelight to dark relaxation kinetic, measuring the time course at 14 wa-velengths in the 500–600 nm region (Supplementary Fig. 1 and Fig. 4).We checked that despite their different pigment composition, the twotypes of leaves similarly transmitted the different wavelengths chosenfor this spectral analysis (Supplementary Fig. 1), as required for acorrect comparison. Spectral changes measured during the light to darktransition were described using a global fit procedure to determine thespectral features of the different phases of the signal relaxation. Wefound that four exponentials correctly reproduce the decay profile inthe 0–60 s time range in “green” leaves, while only three exponentialswere needed to fully reproduce the spectroscopic features of “red”leaves.

In green leaves, the first component (i.e. the fast relaxation phase inFig. 4) has a 1/e decay time (τ) of 10–20ms, and, based on its peak

Fig. 2. Representative features of 5 weeks old WT (A) and TPK3-(B) silencedplants grown at 100 μmol photons m−2 s−1 with a 18 h light and a 6 h darkphotoperiod.This light regime induced accumulation of anthocyanin- rich leaves (redleaves), which are more abundant in mutants plants than in their WT coun-terparts.

G. Allorent et al. BBA - Bioenergetics 1859 (2018) 676–683

678

position (around 520 nm, Fig. 5), it could represent the relaxation of theECS signal. Therefore, this phase could reflect H+

flux though the ATPsynthase, as proposed earlier [2,34,35]. The second phase (τ ~300ms)also displays a peak around 520 nm, and should represents a slowerdecay of the ECS signal. It could reflect a slowing down of the turnoverrate of the ATP synthase, which is known to occur at low values of thepmf [36]. The third phase (τ ~5 s) also displayed a main peak at520 nm, and could therefore represent the inversion of the ECS iden-tified previously [13]. This phase could stem from the slow

equilibration of ions other than H+ to restore their distribution acrossthe thylakoids in the light, due to the building of the pmf [12]. Alter-natively, it could reflect a H+ influx into the lumen, mediated by ATPsynthase complexes, which would work in the direction of ATP hy-drolysis, until an equilibrium between the pmf and the phosphorylatingpotential (∆ = ∆ ′ + ∗

∗G G RT lnp oADP Pi

ATP[ ] [ ]

[ ] ) is reached, as proposed ear-lier [13,37].

We found that an additional decay phase was needed in both WT

Fig. 3. In vivo kinetics of the relaxation of the ECS signal after illumination.Light intensity was 590 μmol photons m−2 s−1. White box: the actinic light was on; black box: the actinic light was switched off. N= 10 ± S.E.M form 4 biologicalreplicates. See figure panels for symbols explanation.

Fig. 4. Time course of the light (590 μmol photons m−2 s−1) to dark relaxation measured at 14 different wavelengths in the 500–600 nm range.Datapoints (black circles) are presented along with fitting traces (surfaces). Samples (WT and TPK3 KD, green and red leaves) are indicated above the panels.N=5 ± S.E.M. See also Supplementary Fig. 2 for a different view.

G. Allorent et al. BBA - Bioenergetics 1859 (2018) 676–683

679

and TPK3 KD green leaves to properly reproduce the relaxation kinetic.While the τ of this phase could not be correctly resolved by our globalanalysis, its spectral features were once again suggestive of a change ofthe ECS signal. We tentatively interpret this phase as the signature of aslow relaxation of the pmf after attainment of the equilibrium with thephosphorylating potential, presumably due to a gradual cellular ATPconsumption. This interpretation is in line with earlier suggestions ofthe existence of a “long-lasting” pmf in chloroplasts [13,37,38]. Itsexistence is also corroborated by the finding that NPQ relaxation inplants not only includes a fast, pH sensitive phase (qE), but also com-prises a slower decay, which likely represents a slow relaxation of qE[39], due to a slow decrease of the proton gradient in the dark [40].

We note that the spectral features of the four components are notentirely identical. In particular, a contribution of cyt f redox changescould be seen (peak at ~555 nm) in the first component. Other possiblecontribution could be observed in the other DAS. However, due to somesuperposition of the measuring wavelengths (Supplementary Fig. 1), itwas not possible to correctly disentangle these signals, which areprobably of lower amplitude, from the ECS contribution.

Different kinetics were seen in the red leaves from both genotypes.In these samples, three exponential decays were sufficient to describethe light to dark transition kinetics (Fig. 5). The spectral features andlifetime of the first two phases were similar to those observed in greenleaves, but the second phase was smaller. On the other hand, the thirdrecovery phase (characterised by a maximum at 520 nm in greenleaves) was replaced by a decay signal, which we tentatively assign tothe 535 nm bandshift based on its peak position. This suggests that inthese anthocyanin-rich leaves the slow dynamics of the ECS signal aremodified, and therefore other spectroscopic signals (otherwise masked

by the ECS) become visible.Different hypotheses can be proposed to explain the different ECS

profiles observed in green and red leaves. The changes in the ECSprofile could reflect changes in the physiological state of the leaves. Inparticular, stress/increased age could reduce their photosynthetic per-formances, and affect the partitioning of the pmf. Consistent with thisidea, earlier findings have pinpointed a change in the ECS partitioningin leaves where photosynthesis was decreased by lowering CO2 avail-ability [41]. To test this hypothesis, we measured the photosyntheticperformances and light stress responses of green and red leaves byimaging chlorophyll fluorescence responses in whole plants. Indeed,one of the main signature of the TPK3 KD plants was their reducedphotosynthesis and diminished NPQ capacity [16]. We found (Fig. 6)that old leaves (i.e. the ones at the outermost part of the rosette) had adecreased photosynthetic capacity, measured as a lower electrontransfer rate (ETR) than their younger counterparts. This lower activityparalleled with a diminished NPQ response, likely reflecting the de-creased ability to generate a ΔpH in the light.

Red leaves could also present altered expression/activity of ionchannels, leading to a modification in counter ion fluxes. To test thispossibility, we measured the transcript level of TPK3 in green and redleaves of both genotypes, in particular in the same leaves of 5-week oldplants that were analysed also by ECS and for NPQ. We found that TPK3KD plants expressed substantially less TPK3 with respect to WT or-ganisms. Moreover, TPK3 expression in red leaves was reproduciblylower than in the green leaves in both genotypes (Fig. 7A, B). A com-parison of the TPK3 expression levels for the four types of leaves withthe ECS and NPQ values is reported in Fig. 7C and D, respectively.

Fig. 5. Time-resolved ECS relaxation in WT and mutant leaves.The Decay Associated Spectra (DAS) and their relative lifetimes were calculated from fitting of the dark relaxation kinetics shown in Fig. 4. Positive amplitudesrepresent decay phases, while negative amplitudes are representative of recovery phases. Error bars represents the standard error of the fit. Samples (WT and TPK3KD, green and red leaves) are indicated above the panels. See methods and text for further explanation.

G. Allorent et al. BBA - Bioenergetics 1859 (2018) 676–683

680

4. Conclusions

Using a global spectral analysis, our study allows disentangling theECS signal from other optical changes in the light. Thus, we can confirmthat several ECS kinetic phases exist during a light to dark transition.One of these phases (phase “2” in Fig. 1) reflects an increase of the ECSin the dark, and could therefore be indicative of an equilibration of ionsdifferent from H+, as proposed earlier [13]. Based on the comparison ofthe ECS spectral and kinetic profiles (and in particular of the phase “2”)in the four types of leaves analysed in this work (green and red leavesfrom WT and TPK3 KD plants), several conclusions can be drawn. First,i. differences in the ECS recovery (phase “2”) are substantial betweenWT green and TPK3 red leaves, where the lower amplitudes of phase“2” are paralleled by diminished NPQ values. These data are in linewith our previous conclusion that TPK3 participates in the regulation ofplant photosynthesis [16]. On the other hand, ii. the relationship be-tween changes in the above mentioned photosynthetic parameters and

the expression levels TPK3 is complex. In both WT and TPK3 KD plants,the red leaves express significantly less TPK3 than green leaves, and thismight contribute to the observed decrease of photosynthetic parameterseven in WT leaves when they become red. WT and TPK3 green leavesbehave similarly, although they display different expression levels ofTPK3 (Fig. 7A). Moreover, TPK3 KD green leaves express lower levels ofTPK3 than WT red leaves, but the ECS recovery and NPQ are larger inthe TPK3 KD green leaves than in the WT red ones (Fig. 7). WT greenleaves obtained from plants derived from the same batch of seeds dis-play a slightly different expression level of TPK3 grown in two differentlaboratories (not shown), suggesting that growth conditions are influ-encing the expression of the channel. Overall, the intricate link betweenphotosynthetic responses, light stress and the absolute TPK3 levelssuggests that while this channel could modulate leaf responses via adirect role in ion equilibration across the thylakoid levels, it could playa more indirect role, by affecting photosynthesis via its consequenceson plant stress, which in turn affects the pmf. Finally, iii. our previous

Fig. 6. Imaging of photosynthetic responses via chlorophyll fluorescence parameters in Arabidopsis thaliana plants.Top line: plant visible phenotype and false colour representation of Non Photochemical Quenching (NPQ) values measured after 5 min of illumination in a wholerosette of a 5-weeks old plant. Brighter colours indicate higher NPQ values. Middle line: Electron Transfer Rate (ETR) measured in plants exposed to different lightintensities. Bottom line: NPQ values measured in plants exposed to different light intensities. Green: kinetics measured in “green” central leaves. Blue: kineticsmeasured in “red” peripheral leaves. Left column: WT; right column: TPK3 KD. N=5 ± S.E.M.

G. Allorent et al. BBA - Bioenergetics 1859 (2018) 676–683

681

comparison of 4-week old WT leaves without anthocyan with stressed,anthocyan-rich TPK3-less leaves in the same respective position in therosettes likely resulted in an overestimation of the difference of pho-tosynthetic parameters among the two genotypes [16]. Indeed, thedifferences between WT and TPK3 KD observed under the present ex-perimental conditions are closer to values previously reported for otherchannels/transporters (e.g. VCCN, ClC-e, KEA3 and PAM71, wherechanges in the range of ≤25% with respect to WT organisms werereported [14]).

Altogether, we propose that global ECS measurements (to properlydistinguish this signal form other coexisting ones) in different geno-types, performed in parallel with expression analysis in the single ex-amined leaves under relevant environmental conditions should providea correct evaluation of the pmf partitioning between the proton gra-dient and the electric field (as proposed in reference [13]) and helpassessing the role of the numerous existing channels/transporters in theregulation of the proton motive force in photosynthetic systems.

Transparency document

The Transparency document associated with this article can befound, in online version.

Acknowledgements

This work was supported by a grant from the HFSP (to G.A, L.C., I. S.and G. F.). Funds from the Labex Gral (ANR-10-Labx-49-01) to G.A. andG.F. are also acknowledged. We would like to thanks Olivier Bastien(Grenoble) for help in generating the 3D plots of Fig. 4 and

Supplementary Fig. 2.

Appendix A. Supplementary data

Supplementary data to this article can be found online at https://doi.org/10.1016/j.bbabio.2018.07.001.

References

[1] S. Eberhard, G. Finazzi, F.-A. Wollman, The dynamics of photosynthesis, Annu. Rev.Genet. 42 (2008) 463–515.

[2] H.T. Witt, Energy conversion in the functional membrane of photosynthesis.Analysis by light pulse and electric pulse methods: the central role of the electricfield, Biochim. Biophys. Acta 505 (1979) 355–427.

[3] G. Kaim, P. Dimroth, ATP synthesis by F-type ATP synthase is obligatorily depen-dent on the transmembrane voltage, EMBO J. 18 (1999) 4118–4127.

[4] W.S. Chow, A.B. Hope, The electrochromic signal, redox reactions in the cyto-chrome bf complex and photosystem functionality in photoinhibited tobacco leafsegments, Funct. Plant Biol. 25 (1998) 775–784.

[5] P. Joliot, G.N. Johnson, Regulation of cyclic and linear electron flow in higherplants, Proc. Natl. Acad. Sci. U. S. A. 108 (2011) 13317–13322.

[6] K. Sonoike, Photoinhibition of photosystem I, Physiol. Plant. 142 (2011) 56–64.[7] S. de Bianchi, M. Ballottari, L. Dall'osto, R. Bassi, Regulation of plant light har-

vesting by thermal dissipation of excess energy, Biochem. Soc. Trans. 38 (2010)651–660.

[8] K.K. Niyogi, T.B. Truong, Evolution of flexible non-photochemical quenching me-chanisms that regulate light harvesting in oxygenic photosynthesis, Curr. Opin.Plant Biol. 16 (2013) 307–314.

[9] G.A. Davis, A. Kanazawa, M.A. Schöttler, K. Kohzuma, J.E. Froehlich,A.W. Rutherford, M. Satoh-Cruz, D. Minhas, S. Tietz, A. Dhingra, D.M. Kramer,Limitations to photosynthesis by proton motive force-induced photosystem IIphotodamage, elife 5 (2016), https://doi.org/10.7554/eLife.16921 pii: e16921.

[10] B. Bailleul, P. Cardol, C. Breyton, G. Finazzi, Electrochromism: a useful probe tostudy algal photosynthesis, Photosynth. Res. 106 (2010) 179–189.

[11] M.P. Johnson, A.V. Ruban, Rethinking the existence of a steady-state Δψ component

Fig. 7. Comparison of transcription level of TPK3 and of photosynthetic parameters in green and red leaves from WT and TPK3 plants.(A) Representative RT-PCR results showing different levels of TPK3 expression in leaves from the indicated plants. Samples were processed together and are from thesame gel. (B) quantification based on measurements on 6 independent leaves. Mean normalized values ± SEM obtained on the basis of densitometry (using actin asan internal standard) are shown. Normalization was obtained on values measured on WT green leaves. (C): Results of experiments from Fig. 3 redrawn with the samecolour code as in (B) for comparison. N=5 ± S.E.M. (D): Results of experiments from Fig. 6 redrawn with the same colour code as in (B) for comparison;N=10 ± S.E.M. a: statistically significant differences (p < 0.05) between WT (green leaves) and TPK3 KD (green leaves); b: statistically significant differences(p < 0.05) between WT (green leaves) and TPK3 KD (red leaves); c: statistically significant differences (p < 0.05) between WT (red leaves) and TPK3 KD (greenleaves); d: statistically significant differences (p < 0.05) between WT (red leaves) and TPK3 KD (red leaves); e: statistically significant differences (p < 0.05)between WT (green leaves) and WT (red leaves); f: statistically significant differences (p < 0.05) between TPK3 KD (green leaves) and TPK3 KD (red leaves).

G. Allorent et al. BBA - Bioenergetics 1859 (2018) 676–683

682

of the proton motive force across plant thylakoid membranes, Photosynth. Res. 119(2014) 233–242.

[12] G.A. Davis, A.W. Rutherford, D.M. Kramer, Hacking the thylakoid proton motiveforce for improved photosynthesis: modulating ion flux rates that control protonmotive force partitioning into Δψ and ΔpH, Philos. Trans. R. Soc. Lond. Ser. B Biol.Sci. 372 (2017) 20160381.

[13] J.A. Cruz, C.A. Sacksteder, A. Kanazawa, D.M. Kramer, Contribution of electric field(Delta psi) to steady-state transthylakoid proton motive force (pmf) in vitro and invivo. Control of pmf parsing into Delta psi and Delta pH by ionic strength,Biochemistry 40 (2001) 1226–1237.

[14] C. Spetea, A. Herdean, G. Allorent, L. Carraretto, G. Finazzi, I. Szabo, An update onthe regulation of photosynthesis by thylakoid ion channels and transporters inArabidopsis, Physiol. Plant. 161 (2017) 16–27.

[15] V. Checchetto, E. Teardo, L. Carraretto, L. Leanza, I. Szabo, Physiology of in-tracellular potassium channels: a unifying role as mediators of counterion fluxes?Biochim. Biophys. Acta 1857 (2016) 1258–1266.

[16] L. Carraretto, E. Formentin, E. Teardo, V. Checchetto, M. Tomizioli, T. Morosinotto,G.M. Giacometti, G. Finazzi, I. Szabo, A thylakoid-located two-pore K+ channelcontrols photosynthetic light utilization in plants, Science 342 (2013) 114–118.

[17] H. Kunz, M. Gierth, A. Herdean, M. Satoh-Cruz, D.M. Kramer, C. Spetea,J.I. Schroeder, Plastidial transporters KEA1, −2, and −3 are essential for chlor-oplast osmoregulation, integrity, and pH regulation in Arabidopsis, Proc. Natl.Acad. Sci. U. S. A. 111 (2014) 7480–7485.

[18] U. Armbruster, L.R. Carrillo, K. Venema, L. Pavlovic, E. Schmidtmann, A. Kornfeld,P. Jahns, J.A. Berry, D.M. Kramer, M.C. Joniknas, Ion antiport accelerates photo-synthetic acclimation in fluctuating light environments, Nat. Commun. 5 (2014)5439.

[19] C. Wang, H. Yamamoto, F. Narumiya, Y.N. Munekage, G. Finazzi, I. Szabo,T. Shikanai, Fine-tuned regulation of the K(+)/H(+) antiporter KEA3 is required tooptimize photosynthesis during induction, Plant J. 89 (2017) 540–553.

[20] A. Herdean, H. Nziengui, O. Zsiros, K. Solymosi, G. Garab, B. Lundin, C. Spetea, TheArabidopsis thylakoid chloride channel AtCLCe functions in chloride homeostasisand regulation of photosynthetic electron transport, Front. Plant Sci. 7 (2016) 115.

[21] A. Herdean, E. Teardo, A.K. Nilsson, B.E. Pfeil, O.N. Johansson, R. Ünnep, G. Nagy,O. Zsiros, S. Dana, K. Solymosi, G. Garab, I. Szabo, C. Spetea, B. Lundin, A voltage-dependent chloride channel fine-tunes photosynthesis in plants, Nat. Commun. 7(2016) 11654.

[22] Z. Duan, F. Kong, L. Zhang, W. Li, J. Zhang, L. Peng, A bestrophin-like proteinmodulates the proton motive force across the thylakoid membrane in Arabidopsis,J. Integr. Plant Biol. 58 (2016) 848–858.

[23] A. Schneider, I. Steinberger, A. Herdean, C. Gandini, M. Eisenhut, S. Kurz,A. Morper, N. Hoecker, T. Rühle, M. Labs, U.I. Flügge, S. Geimer, S.B. Schmidt,S. Husted, A.P. Weber, C. Spetea, D. Leister, The evolutionarily conserved proteinPHOTOSYNTHESIS AFFECTED MUTANT71 is required for efficient manganeseuptake at the thylakoid membrane in Arabidopsis, Plant Cell 28 (2016) 892–910.

[24] C. Wang, W. Xu, H. Jin, T. Zhang, J. Lai, X. Zhou, S. Zhang, S. Liu, X. Duan,H. Wang, C. Peng, C. Yang, C.A. Putative, Chloroplast-localized Ca(2+)/H(+)antiporter CCHA1 is involved in calcium and pH homeostasis and required for PSII

function in Arabidopsis, Mol. Plant (2016) 1183–1196.[25] Z. Xu, K. Mahmood, S.J. Rothstein, ROS induces anthocyanin production via late

biosynthetic genes and anthocyanin deficiency confers the hypersensitivity to ROS-generating stresses in Arabidopsis, Plant Cell Physiol. 58 (2017) 1364–1377.

[26] T. Tohge, Y. Nishiyama, M.Y. Hirai, M. Yano, J. Nakajima, M. Awazuhara, E. Inoue,H. Takahashi, D.B. Goodenowe, M. Kitayama, M. Noji, M. Yamazaki, K. Saito,Functional genomics by integrated analysis of metabolome and transcriptome ofArabidopsis plants over-expressing an MYB transcription factor, Plant J. 42 (2005)218–235.

[27] L. Chalker-Scott, Environmental significance of anthocyanins in plant stress re-sponses, Photochem. Photobiol. 70 (1999) 1–9.

[28] D.C. Fork, N. Murata, N. Sato, Effect of growth temperature on the lipid and fattyacid composition, and the dependence on temperature of light-induced redox re-actions of cytochrome f and of light energy redistribution in the thermophilic blue-green alga Synechococcus lividus, Plant Physiol. 63 (1979) 524–530.

[29] P. Joliot, A. Joliot, Mechanism of electron transfer in the cytochrome b/f complex ofalgae: evidence for a semiquinone cycle, Proc. Natl. Acad. Sci. U. S. A. 91 (1994)1034–1038.

[30] U. Heber, Conformational changes of chloroplasts induced by illumination of leavesin vivo, Biochim. Biophys. Acta 180 (1969) 302–319.

[31] A.V. Ruban, A.A. Pascal, B. Robert, P. Horton, Activation of zeaxanthin is an ob-ligatory event in the regulation of photosynthetic light harvesting, J. Biol. Chem.277 (2002) 7785–7789.

[32] X. Johnson, G. Vandystadt, S. Bujaldon, F.A. Wollman, R. Dubois, P. Roussel,J. Alric, D. Béal, A new setup for in vivo fluorescence imaging of photosyntheticactivity, Photosynth. Res. 102 (2009) 85–93.

[33] K. Maxwell, G.N. Johnson, Chlorophyll fluorescence-a practical guide, J. Exp. Bot.51 (2000) 659–668.

[34] D.M. Kramer, J.A. Cruz, A. Kanazawa, Balancing the central roles of the thylakoidproton gradient, Trends Plant Sci. 8 (2003) 27–32.

[35] P. Joliot, R. Delosme, Flash-induced 519 nm absorption change in green algae,Biochim. Biophys. Acta 357 (1974) 267–284.

[36] U. Junesch, P. Gräber, The rate of ATP synthesis as a function of ΔpH in normal anddithiothreitol-modified chloroplasts, Biochim. Biophys. Acta 809 (1985) 429–434.

[37] T.J. Avenson, A. Kanazawa, J.A. Cruz, K. Takizawa, W.E. Ettinger, D.M. Kramer,Integrating the proton circuit into photosynthesis: progress and challenges, PlantCell Environ. 28 (2005) 97–109.

[38] P. Cardol, R. De Paepe, F. Franck, G. Forti, G. Finazzi, The onset of NPQ and ΔμH+upon illumination of tobacco plants studied through the influence of mitochondrialelectron transport, Biochim. Biophys. Acta 1797 (2010) 177–188.

[39] P. Joliot, G. Finazzi, Proton equilibration in the chloroplast modulates multiphasickinetics of nonphotochemical quenching of fluorescence in plants, Proc. Natl. Acad.Sci. U. S. A. 107 (2010) 12728–12733.

[40] P. Joliot, A. Joliot, Quantification of the electrochemical proton gradient and ac-tivation of ATP synthase in leaves, Biochim. Biophys. Acta 1777 (2008) 676–683.

[41] T.J. Avenson, J.A. Cruz, D.M. Kramer, Modulation of energy-dependent quenchingof excitons in antennae of higher plants, Proc. Natl. Acad. Sci. U. S. A. 101 (2004)5530–5535.

G. Allorent et al. BBA - Bioenergetics 1859 (2018) 676–683

683