Functional relationship of atypical thioredoxins with NADPH-thioredoxin reductase C and 2-Cys peroxiredoxins in Arabidopsis chloroplasts

Abstract

In addition to typical thioredoxins (TRXs) of the types m, f, x, y, and z, Arabidopsis (Arabidopsis thaliana) chloroplasts contain atypical TRXs such as cysteine- and histidine-rich TRXs (ACHTs) and TRXs-like 2 (TRXs L2), but their contribution to the redox regulation of the organelle and their functional relationship with the NADPH-TRX reductase C (NTRC)–2-Cys peroxiredoxin (PRX) redox system are poorly understood. To address these issues, we have focused on atypical TRXs ACHT3, ACHT4, and TRX L2.2. Either single or multiple Arabidopsis mutants combining deficiencies in these atypical TRXs showed wild-type-like phenotypes, but impaired light- and dark-dependent redox regulation of chloroplast enzymes. Their functional relationship with the NTRC–2-Cys PRXs redox system was tested in mutants lacking both the atypical TRXs analyzed here and NTRC or 2-Cys PRXs. Our results show the contribution of these atypical TRXs to chloroplast reductive and oxidative pathways, a function exerted via 2-Cys PRXs, which allows modulation of the redox state of chloroplast enzymes in response to light and darkness. The effects of different light regimes on the growth of the single and multiple mutants suggest that the contribution of atypical TRXs to chloroplast redox regulation might be relevant for plant acclimation to changes in light intensity.

Atypical thioredoxins ACHT3, ACHT4, and TRX L2.2 fine-tune the light- and dark-dependent redox state of chloroplast enzymes through 2-Cys peroxiredoxins.

Introduction

Redox regulation of enzyme activity based on disulfide–dithiol exchange plays a key role for fine-tuning chloroplast metabolism in response to changes in light intensity and darkness (Cejudo et al., 2019), a process in which the protein disulfide reductase activity of thioredoxins (TRXs) plays a central role. Chloroplast TRXs rely on photosynthetically reduced ferredoxin (Fdx), via a Fdx-dependent TRX reductase (FTR) (Schürmann and Buchanan, 2008) that links the regulation of chloroplast metabolism to light. Chloroplast TRXs include the so-called typical isoforms (m, f, x, y, and z) characterized by a well-defined tertiary structure, the TRX fold, and an active site formed by the conserved W-C-G/P-P-C motif (Chibani et al., 2021). TRXs of the types m and f are the most abundant, accounting for up to ∼91% of the total content of TRXs of the organelle (Okegawa and Motohashi, 2015), and their participation in the redox regulation of essential chloroplast metabolic pathways such as the Calvin–Benson cycle (CBC) is well established (Michelet et al., 2015; Okegawa and Motohashi, 2015). Moreover, the use of Arabidopsis mutants has allowed the relevant function of TRXs m in photoprotection, the control of the photosynthetic cyclic electron transport, and adaptation of photosynthesis to fluctuating light to be shown (Courteille et al., 2013; Thormählen et al., 2016; Okegawa and Motohashi, 2020; Serrato et al., 2021; Okegawa et al., 2022). In contrast, less abundant TRXs of the types x and y were shown to efficiently reduce the thiol-dependent peroxidases 2-Cys peroxiredoxins (PRXs) (Collin et al., 2003) and PRX Q (Collin et al., 2004), respectively, and thus an antioxidant function was proposed for these TRXs. Because Arabidopsis mutants lacking individual typical TRX isoforms show an almost wild-type (WT)-like growth phenotype under standard culture conditions, it is considered that these TRXs have redundant functions in chloroplast redox regulation.

In addition to the Fdx–FTR–TRXs redox system, chloroplasts harbor an NADPH-dependent TRX reductase (NTR), termed NTRC, which contains a TRX domain at the C-terminus (Serrato et al., 2004) and shows high affinity for NADPH (Bernal-Bayard et al., 2014). Biochemical analyses have shown that NTRC is a very efficient reductant of 2-Cys PRXs (Moon et al., 2006; Pérez-Ruiz et al., 2006; Alkhalfioui et al., 2007), which supported an antioxidant function for these enzymes. However, the pleiotropic phenotype of the Arabidopsis ntrc mutant suggests additional functions of NTRC. In this regard, it was shown that TRXs f are likely candidates to be reduced by NTRC (Toivola et al., 2013), which also interacts with fructose-1,6-bisphosphatase (FBPase) (Nikkanen et al., 2016). Moreover, despite the poor phenotypic effects shown by Arabidopsis mutants devoid of individual TRX isoforms, the combination of the deficiencies of NTRC with those of TRXs f (Thormählen et al., 2015; Ojeda et al., 2017), TRX x (Ojeda et al., 2017), TRXs m (Da et al., 2017; Delgado-Requerey et al., 2023), and, to a lesser extent, TRXs y (Jurado-Flores et al., 2020), aggravates the growth phenotype and photosynthetic performance shown by the ntrc mutant, indicating that NTRC activity is needed for the function of chloroplast TRXs. In contrast, Arabidopsis mutants simultaneously deficient in NTRC and 2-Cys PRXs recovered the WT-like phenotype, revealing the central function of 2-Cys PRXs in chloroplast redox regulation (Pérez-Ruiz et al., 2017). Oxidation of chloroplast enzymes in the dark is delayed in mutants deficient in 2-Cys PRXs (Ojeda et al., 2018; Vaseghi et al., 2018; Yoshida et al., 2018), and accelerated in the ntrc mutant (Ojeda et al., 2021), hence confirming the central function of the NTRC–2-Cys PRXs redox system in fine-tuning the activity of redox-regulated enzymes in response to changes in light intensity and darkness.

In addition to typical TRXs, more recent analyses identified up to 18 atypical TRXs in Arabidopsis, which have a non-canonical active site (Chibani et al., 2021). These include TRXs such as Chloroplastic Drought-induced Stress Protein of 32 kDa (CDSP32), which participates in response to stress but also in the redox regulation of ATP synthase activity (Rey et al., 2024). Other chloroplast-localized atypical TRXs include the TRX-like 2 (TRX L2) isoforms, which contain the motif W-C-R-K-C at the active site (Cain et al., 2009; Chibani et al., 2009). The Arabidopsis genome encodes two TRX L2 isoforms, termed TRX L2.1 and TRX L2.2, which have less negative redox potential than typical TRXs (Yoshida et al., 2018). Based on the finding that TRX L2.1 and TRX L2.2 are efficient reductants of 2-Cys PRXs, but poor reductants of stromal enzymes such as FBPase, their participation in oxidative thiol modulation was proposed (Yoshida et al., 2018, 2019; Sekiguchi et al., 2022). Arabidopsis chloroplasts also contain a small family of atypical Cys- and His-rich TRXs (ACHTs), comprising six isoforms (ACHT1–ACHT6) (Dangoor et al., 2009). Based on the fact that ACHT1 is oxidized by 2-Cys PRX, its function as an oxidative redox sensor relevant in the response to moderate light intensity was proposed (Dangoor et al., 2012). Moreover, it was found that ACHT4 and 2-Cys PRX participate in the oxidation of the small subunit (APS1) of ADP-glucose pyrophosphorylase (AGPase) (Eliyahu et al., 2015). Finally, the analysis of Arabidopsis mutants lacking TRX L2.1 or ACHT1 and ACHT2 further supported the function of atypical chloroplast TRXs in the oxidation of several plastid enzymes, namely the γ subunit of ATPase (γATPase), Rubisco activase (RCA), and FBPase, during light-to-dark transitions (Yokochi et al., 2021).

Therefore, the analyses of chloroplast atypical TRXs performed so far that support their participation in enzyme oxidation during the night have been mainly focused on ACHT1, ACHT2, and TRX L2.1. However, Arabidopsis chloroplasts contain additional members of this family of atypical TRXs, but neither their function in the regulation of redox-sensitive proteins nor their functional relationship with the NTRC–2-Cys PRXs redox system has been established. Thus, with the aim of filling this gap, we focused on ACHT3, ACHT4, and TRX L2.2, as additional representatives of the ACHT and TRX L2 families of atypical TRXs, to analyze their contribution to the oxidative and reductive pathways of chloroplast redox regulation and their functional relationship with the NTRC–2-Cys PRXs redox system.

Materials and methods

Biological material and growth conditions

Arabidopsis thaliana (ecotype Columbia) was used throughout this work as the WT line. The single mutants acht3 (SALK_062853), acht4 (SALK_014503), and trxl2.2 (SALK_045620) were obtained from the European Arabidopsis Stock Centre (Nottingham University, UK). These single mutants were manually crossed to generate the double mutants acht3-trxl2.2, acht4-trxl2.2, and acht3-acht4, then the manual cross of acht3-trxl2.2 with acht4-trxl2.2 allowed the identification of the acht3-acht4-trxl2.2 triple mutant. To generate the double mutants ntrc-acht3, ntrc-acht4, and ntrc-trxl2.2, and the triple mutants 2cpab-acht3 and 2cpab-acht4, the acht3, acht4, and trxl2.2 single mutants were manually crossed with the ntrc (Serrato et al., 2004) and the 2cpab mutant (Ojeda et al., 2018), respectively, which had been previously generated in our laboratory. Single and multiple mutants were identified by PCR analysis of the presence of the T-DNA in genomic DNA using the oligonucleotides indicated in Supplementary Table S1. WT and mutant plants were routinely grown on soil in culture chambers under long-day (16 h light/8 h darkness) or short-day (8 h light/16 h darkness) photoperiods at 22 °C and 20 °C during the day and night periods, respectively, at a light intensity of 120 μmol m⁻² s⁻¹ and 70% relative humidity. For treatments with different light regimes, Arabidopsis WT and mutant plants were germinated under long-day conditions for 5 d and then grown under continuous light (at a light intensity of 120 μmol m⁻² s⁻¹), or a long-day photoperiod with high (525 μmol m⁻² s⁻¹) or fluctuating (357 μmol m⁻² s⁻¹, 1 min; 49 μmol m⁻² s⁻¹, 8 min) light intensity during the light period.

RNA extraction and quantitative reverse transcription–PCR analysis

Total RNA was extracted with the Trizol reagent (Invitrogen), and cDNA synthesis was performed with 2 μg of total RNA using the Maxima first-strand cDNA synthesis kit (Thermo Scientific) according to the manufacturer’s instructions. Quantitative reverse transcription–PCR (RT–qPCR) was performed using an IQ5 real-time PCR detection system (Bio-Rad). A standard thermal profile (95 °C for 3 min, followed by 40 cycles of 95 °C for 10 s and 60 °C for 30 s) was used for all reactions. After the PCR, a melting-curve analysis (55–94 °C at 0.5 °C per 30 s) was performed to confirm the specificity of the amplicon and to exclude primer dimers or non-specific amplification. Primer sequences for RT–qPCR are listed in Supplementary Table S1. Relative transcript levels were calculated using the delta-Ct method and normalized to ACTIN and UBC9 as reference genes.

Protein extraction, alkylation assays, and western blot analysis

Leaf tissues from plants grown as indicated in the figure legends were dissected and immediately frozen in liquid nitrogen. Protein extracts were prepared with extraction buffer [50 mM Tris–HCl pH 8.0, 0.15 M NaCl, 0.5% (v/v) NP-40], which was immediately added to leaf tissues that were ground under liquid nitrogen. After mixing on a vortex, extracts were centrifuged at 16 100 g at 4 °C for 20 min, and protein was quantified with the Bradford reagent (Bio-Rad). Alkylation assays were performed as previously reported (Ojeda et al., 2021) using 60 mM iodoacetamide as alkylating agent. Protein samples were subjected to SDS–PAGE under reducing conditions for the detection of NTRC or under non-reducing conditions for the detection of γATPase, RCA, FBPase, and 2-Cys PRXs, using an acrylamide gel concentration of 9.5% (γATPase, RCA, and FBPase) or 12% (NTRC and 2-Cys PRXs). Resolved proteins were transferred to nitrocellulose membranes and probed with the indicated antibody. Specific antibodies for NTRC (Serrato et al., 2004) and 2-Cys PRXs (Pérez-Ruiz et al., 2006) were previously raised in our laboratory. The anti-RCA antibody was kindly provided by Dr A.R. Portis (USDA, Urbana, USA), the γATPase antibody was purchased from Agrisera (Sweden), and the anti-FBPase antibody was provided by Dr M. Sahrawy (Estación Experimental del Zaidín, CSIC, Granada, Spain).

Determination of chlorophylls and measurements of Chl a fluorescence

Chlorophyll was measured as previously described (Pérez-Ruiz et al., 2006). Room temperature Chl a fluorescence of PSII was measured using the pulse-amplitude modulation fluorometer IMAGING-PAM M-Series (Walz). Plants that had been grown as indicated in the figure legends were dark-adapted for 30 min before measurements and induction–recovery curves were performed using actinic light at 81 μmol m⁻² s⁻¹ for 7 min. Saturating pulses of red light at 10 000 μmol m⁻² s⁻¹ intensity and 0.6 s duration were applied every 40 s, and recovery in darkness was recorded for up to 8 min. The parameters Y(II) and Y(NPQ), corresponding to the quantum yields of PSII photochemistry and non-photochemical quenching, respectively, were calculated according to reported equations (Kramer et al., 2004). The electron transport rate through PSII [ETR(II)] was measured in pre-illuminated plants applying stepwise increasing actinic light intensities up to 400 μmol m⁻² s⁻¹ every 30 s and saturation pulses at the end of each. The values were then calculated by the IMAGING-PAM software according to the manual: ETR=Yield × PAR × 0.5 × absorptivity.

Scanning electron microscopy

For each genotype, mature pollen grains were collected from flowers from different plants grown under long-day photoperiods. Pollen morphology was analyzed by SEM with a HITACHI S-4800 SEM-FEG microscope operated at 2 kV. Previously, the samples were coated with gold.

Germination capacity

Seeds were harvested from plants grown under long-day photoperiods. Germination assays were perfomed using four batches of seeds in independent experiments. In brief, four sets of at least 25 seeds were sown on Murashige and Skoog (MS) plates, stratified for 48 h at 4 °C in the dark, and exposed to continuous light (22 °C, light intensity of 120 μmol m⁻² s⁻¹). Seed germination, defined as radicle emergence, was scored 72 h after the start of the light period.

Statistical analysis

Statistical analyses were performed using GraphPad Prism version 6.01 (GraphPad Software, Inc., La Jolla, CA, USA) and the experimental results were analyzed with one-way ANOVA test with a Tukey’s multiple comparison test (P<0.05).

Results

Atypical TRXs ACHT3, ACHT4, and TRX L2.2 are dispensable for plant growth

Previous studies have explored the impact of atypical TRXs ACHT1, ACHT2, and TRX L2.1 on chloroplast enzyme oxidation (Yokochi et al., 2021). With the aim of analyzing the contribution of additional atypical TRXs to chloroplast redox regulation, in this work we have focused on ACHT3, ACHT4, and TRX L2.2. The amino acid sequence comparison shows the characteristic active site of atypical TRXs L2.1 and L2.2 in which the double cysteine is separated by basic amino acids arginine and lysine, whereas ACHTs 1–4 show different variations of the canonical W-C-G/P-P-C active site of typical TRXs (Supplementary Fig. S1). According to the gene expression data from the Arabidopsis ePlant browser (Waese et al., 2017), ACHT1–ACHT4 display different expression patterns. ACHT1 and ACHT2 are predominantly expressed in leaves (Supplementary Fig. S2A, B). However, ACHT3 shows the highest expression in mature pollen (Supplementary Fig. S2C) and ACHT4 in developing embryo and dry seed (Supplementary Fig. S2D). In contrast, the TRX L2.1 and TRX L2.2 genes show a similar pattern of expression in leaves, though TRX L2.2 is expressed at lower levels in rosette leaves (Supplementary Fig. S3A, B). Because no antibodies against these TRXs are available in our laboratory, the levels of expression of ACHT3, ACHT4, and TRX L2.2 in leaves were tested by RT–qPCR analysis. Supplementary Fig. S4A shows that the ACHT4 gene is expressed at a higher level than the TRX L2.2 and ACHT3 genes, yet the level of expression of these genes was much lower than those of the TRX m1 and m4 genes, encoding the most abundant TRXs of Arabidopsis chloroplasts (Okegawa and Motohashi, 2015).

As a first approach to test the impact of these atypical TRXs on plant growth and photosynthesis performance, Arabidopsis T-DNA insertion mutants in each of these genes, here termed acht3, acht4, and trxl2.2, were isolated and confirmed by PCR genotyping (Supplementary Fig. S4B, C). Supplementary Fig. S4D shows the absence of each of the transcripts of the mutated genes in the corresponding T-DNA insertion lines, hence indicating that the acht3, acht4, and trxl2.2 mutants are knockouts for ACHT3, ACHT4, and TRX L2.2, respectively. Moreover, the single mutants showed no remarkable differences in the level of expression of the other genes under analysis (Supplementary Fig. S4D), indicating that there are no expression compensatory effects between these genes in the mutants. The acht3, acht4, and trxl2.2 mutants showed growth rates similar to the WT when grown under a long-day photoperiod (Supplementary Fig. S5A), as shown by the fresh weight of the rosette leaves (Supplementary Fig. S5B) and the leaf chlorophyll content (Supplementary Fig. S5C). Despite the marked expression of the ACHT3 and ACHT4 genes in mature pollen and developing embryo, respectively (Supplementary Fig. S2C, D), no significant alterations were noticed in the pollen structure of any of the mutants (Supplementary Fig. S6A) or in their rate of germination, since 100% of the seeds of the different genotypes germinated (Supplementary Fig. S6B). The lack of effect of the deficiency of these atypical TRXs on plant growth was further tested by the analysis of the photosynthetic performance of the mutants. Y(II) (Supplementary Fig. S5D) and Y(NPQ) (Supplementary Fig. S5E) of each of the single mutants were comparable with those in the WT. In contrast, the lack of ACHT3 or ACHT4, but not of TRX L2.2, caused a slight increase in the ETR(II) at light intensities above ∼150 μmol m⁻² s⁻¹ (Supplementary Fig. S5F), further confirming that the deficiencies of these atypical TRXs have no negative effect on photosynthetic performance.

As an additional approach for investigating the contribution of atypical TRXs to chloroplast redox regulation and the functional relationship between them, double mutants acht3-trxl2.2, acht4-trxl2.2, and acht3-acht4, and the triple mutant acht3-acht4-trxl2.2 were generated as indicated in the Materials and methods and confirmed by PCR genotyping (Supplementary Fig. S7). All these mutants display WT-like phenotypes when grown under a long-day photoperiod (Fig. 1A), as shown by the rosette fresh weight (Fig. 1B) and leaf chlorophyll contents (Fig. 1C), indicating that the combined deficiencies of these atypical TRXs have no negative effect on growth. In line with their WT-like growth phenotypes, the double and triple mutants showed Y(II) (Fig. 1D) and Y(NPQ) (Fig. 1E) values like those of the WT. However, some differences were observed in the ETR, which was slightly lower in the acht4-trxl2.2 and acht3-acht4 mutants and, to a greater extent, in the acht3-acht4-trxl2.2 triple mutant, as compared with the WT (Fig. 1F). Thus, though the simultaneous absence of atypical TRXs ACHT3, ACHT4, and TRX L2.2 causes a slight impairment of PSII performance, as indicated by the lower levels of ETR in the acht3-acht4-trxl2.2 mutant, this alteration does not affect growth under standard conditions.

Fig. 1.

Growth phenotype and photosynthetic performance of double and triple mutants deficient in ACHT3, ACHT4, and TRX L2.2. Arabidopsis WT and mutant plants, as indicated, were grown under long-day conditions for 3 weeks at a light intensity of 120 μmol m⁻² s⁻¹ (A). The fresh weight (B) and chlorophyll content (C) of rosette leaves were determined from at least eight plants grown as in (A) and are represented as average values ±standard errors of the mean (SEM). Different letters above the histograms indicate significant differences by one-way ANOVA with a Tukey’s multiple comparison test (P<0.05). The absence of letters above histograms indicates no significant differences with the same test. Measurements of quantum yields of PSII photochemistry [Y(II)] (D), non-photochemical quenching [Y(NPQ)] (E), and linear photosynthetic electron transport rate [ETR(II)] at increasing photosynthetically active radiation (PAR) (F) were performed using a pulse-amplitude modulation fluorometer (IMAGING-PAM). Values are the average of at least four determinations, and SEM are represented as error bars. White and black blocks indicate periods of illumination with actinic light (81 μmol m⁻² s⁻¹) and darkness, respectively.

Once we had established the non-significant effect of the lack of ACHT3, ACHT4, and TRX L2.2 on plant growth and photosynthesis performance, we determined their effect on the redox regulation of chloroplast enzymes. To that end, we analyzed the in vivo redox state of γATPase, RCA, and FBPase, well-known redox-regulated enzymes that show different redox behavior in response to light; that is, γATPase is rapidly reduced in the light (Yoshida et al., 2014) but more slowly oxidized in the dark than RCA and FBPase (Ojeda et al., 2021; Yokochi et al., 2021). To determine the redox changes of γATPase, RCA, and FBPase in light-to-dark transitions, plants were adapted to a light intensity of 650 μmol m⁻² s⁻¹ that provoked full or a high level of reduction of γATPase (Fig. 2A, B), RCA (Fig. 2A, C), and FBPase (Fig. 2A, D), then the rates of oxidation of these enzymes were determined after 5, 10, and 15 min of darkness. The acht3, acht4, and trxl2.2 single mutants showed slightly delayed oxidation of γATPase (Fig. 2A, B) and RCA (Fig. 2A, C), as compared with the WT, but not of the rapidly oxidized FBPase (Fig. 2A, D). Remarkably, while no significant differences in the rate of FBPase oxidation were observed between the WT and the double and triple mutants (Fig. 2A, G), oxidation of RCA (Fig. 2A, F) was more delayed in the multiple than in the single mutants, and significant differences were observed in the oxidation of γATPase. This protein was fully oxidized in the WT and all the mutants after 10 min and 15 min of darkness; however, after 5 min of darkness, the acht3-acht4 mutant showed higher oxidation than the acht3-trxl2.2 and acht4-trxl2.2 double mutants, with oxidation being slightly higher in the triple mutant (Fig. 2A, E). These results show the additive effect of the combination of atypical TRXs deficiencies on RCA and γATPase oxidation in the dark, which supports the contribution of ACHT3, ACHT4, and TRX L2.2 to the oxidizing component of chloroplast redox regulation.

Fig. 2.

In vivo redox changes of γATPase, RCA, and FBPase in WT and mutant plants in response to darkness. Arabidopsis WT and single, double, and triple mutants, as indicated, were grown under long-day conditions for 3 weeks at a light intensity of 120 μmol m⁻² s⁻¹. At the end of the night period, plants were exposed to a light intensity of 650 μmol m⁻² s⁻¹ during 30 min (L), then light was switched off and samples were harvested at 5, 10, and 15 min of darkness. The in vivo redox states of γATPase, RCA, and FBPase (A) at the indicated time points were determined by labeling of the thiol groups with the alkylating agent iodoacetamide. Molecular mass markers (kDa) are indicated on the left, and vertical lines indicate different blots. Band intensities were quantified (Gel Analyzer) and the percentage reduction, determined as the ratio between the reduced form and the sum of reduced and oxidized forms of γATPase (B, E), RCA (C, F), and FBPase (D, G), are represented as the mean ±SEM of three independent experiments. Different letters at each time point indicate significant differences by one-way ANOVA with a Tukey’s multiple comparison test (P<0.05). red, reduced; oxi, oxidized; r.i., redox insensitive isoform.

We next analyzed the redox changes of γATPase, RCA, and FBPase when the WT and mutant plants were subjected to dark-to-light transitions. When dark-adapted plants in which the three enzymes were fully oxidized (Fig. 3A, C) were illuminated for 30 min at growth light intensity (120 μmol m⁻² s⁻¹), γATPase became completely reduced in the WT and all of the single (Fig. 3A, B) or double and triple mutants (Fig. 3C, D), confirming the rapid redox response of this protein to light, which was not altered by the lack of the atypical TRXs analyzed here. In contrast, RCA and FBPase, displaying almost complete and ∼70% reduction in the WT, respectively, showed lower levels of reduction (∼70–80% for RCA and ∼40–50% for FBPase) in the single mutants (Fig. 3A, B). Remarkably, the levels of reduction of these enzymes were not further impaired in the double and triple mutants, which showed similar or slightly higher levels of RCA and FBPase reduction (Fig. 3C, D) than the single mutants (Fig. 3A, B). Overall, these results indicate that ACHT3, ACHT4, and TRX L2.2 also participate in light-dependent reduction of chloroplast enzymes.

Fig. 3.

In vivo redox changes of γATPase, RCA, FBPase, and 2-Cys PRXs in WT and mutant plants in response to light. Arabidopsis WT and mutant plants, as indicated, were grown under long-day conditions for 3 weeks at a light intensity of 120 μmol m⁻² s⁻¹. The in vivo redox states of γATPase, RCA, and FBPase in single and multiple mutants (A–D) and of 2-Cys PRXs in multiple mutants (E, F), as indicated, were determined at the end of the dark period (D) and after 30 min of illumination at 120 μmol m⁻² s⁻¹ (L) by labeling of the thiol groups with the alkylating agent iodoacetamide. Molecular mass markers (kDa) are indicated on the left. Band intensities were quantified (Gel Analyzer) and the percentage reduction, determined as the ratio between the reduced form and the sum of reduced and oxidized forms of enzymes in single mutants (B) and double and triple mutants (D, F), is represented as the mean ±SEM of three independent experiments. For each enzyme, different letters above the histograms indicate significant differences between lines by one-way ANOVA with a Tukey’s multiple comparison test (P<0.05). red, reduced; oxi, oxidized; r.i., redox insensitive isoform; mon, monomer; dim, dimer.

Finally, we analyzed the redox state of 2-Cys PRXs in response to light and darkness in the WT and the mutant lines. In WT plants, 2-Cys PRXs were more reduced in the light than in the dark, hence reflecting the redox state of the organelle, and single mutants showed only minor, non-statistically significant differences (Supplementary Fig. S8A, B). These differences were more significant in the double and triple mutants, which showed higher levels of reduction of 2-Cys PRXs in the dark and, except for the acht4-trxl2.2 mutant, lower levels of light-dependent reduction than the WT (Fig. 3E, F). Thus, these results suggest the participation of 2-Cys PRXs in the redox regulatory activity of these atypical TRXs.

The functional relationship of atypical TRXs ACHT3, ACHT4, and TRX L2.2 with the NTRC–2-Cys PRX redox system

As an additional approach to establish the contribution of atypical TRXs ACHT3, ACHT4, and TRX L2.2 to chloroplast redox regulation, we analyzed their functional relationship with the redox system formed by NTRC and 2-Cys PRXs, which has been proposed to play a central role in the control of the redox state of the organelle (Pérez-Ruiz et al., 2017). With that purpose, acht3, acht4, and trxl2.2 single mutants were manually crossed with the 2cpab (Ojeda et al., 2018) and the ntrc (Serrato et al., 2004) mutants, and mutants 2cpab-acht3, 2cpab-acht4, ntrc-acht3, ntrc-acht4, and ntrc-trxl2.2 were selected based on PCR analysis of the T-DNA insertion in genomic DNA and immunoblotting (Supplementary Fig. S9A–D). Despite repeated attempts, the 2cpab-trxl2.2 mutant could not be obtained, probably due to genetic linkage between the TRXL2.2 (At5g04260) and 2-CYS PRX B (At5g06290) genes, thus, the effect of the lack of TRX L2.2 in plants fully devoid of 2-Cys PRXs could not be analyzed.

The functional relationship of ACHT3 and ACHT4 with 2-Cys PRXs was addressed by analyzing the 2cpab-acht3 and 2cpab-acht4 triple mutants. When grown under a standard long-day photoperiod, no significant differences were observed between the 2cpab-acht3 and 2cpab-acht4 mutants, which displayed phenotypes like the 2cpab mutant, as shown by the fresh weight (Fig. 4A, B) and the chlorophyll contents, though the 2cpab-acht4 mutant showed a slightly, but statistically significant, lower content of chlorophyll of the rosette leaves (Fig. 4C). In line with their similar growth phenotypes, the photosynthetic parameters Y(II) (Fig. 4D), Y(NPQ) (Fig. 4E), and ETR (Fig. 4F) were indistinguishable in the 2cpab, 2cpab-acht3, and 2cpab-acht4 mutants, all of them being altered as compared with the WT. Therefore, the lack of ACHT3 or ACHT4 has no additional effects in plants devoid of 2-Cys PRXs.

Fig. 4.

Growth phenotype and photosynthetic performance of mutants combining the deficiencies of 2-Cys PRXs and ACHT3 or ACHT4. Arabidopsis WT and mutants 2cpab, 2cpab-acht3, and 2cpab-acht4 were grown under long-day conditions for 4 weeks at a light intensity of 120 μmol m⁻² s⁻¹ (A). The fresh weight (B) and chlorophyll content (C) of rosette leaves were determined from at least 10 plants grown as in (A) and represented as average values ±SEM. Different letters above the histograms indicate significant differences by one-way ANOVA with a Tukey’s multiple comparison test (P<0.05). Measurements of quantum yields of PSII photochemistry [Y(II)] (D), non-photochemical quenching [Y(NPQ)] (E), and linear photosynthetic electron transport rate [ETR(II)] at increasing photosynthetic active radiation (PAR) (F) were performed using a pulse-amplitude modulation fluorometer (IMAGING-PAM). Values are the average of at least four determinations, and SEMs are represented as error bars. White and black blocks indicate periods of illumination with actinic light (81 μmol m⁻² s⁻¹) and darkness, respectively.

In contrast, the combined deficiencies of NTRC and the atypical TRXs under study caused different growth phenotypes. When grown under a standard long-day photoperiod in which the ntrc mutant showed the expected growth retardation (Fig. 5A, B), the ntrc-acht3 mutant was like the ntrc mutant; however, the ntrc-acht4 mutant showed a more aggravated growth inhibition than the ntrc mutant (Fig. 5A, B), but with a similar chlorophyll content, which was slightly increased in the ntrc-acht3 mutant (Fig. 5C). In agreement with previous results (Naranjo et al., 2016a), the ntrc mutant showed impaired photosynthetic performance, which was further affected in the ntrc-acht3 and ntrc-acht4 mutants, as revealed by the lower levels of Y(II) (Fig. 5D), higher levels of Y(NPQ) (Fig. 5E), and lower ETR(II) (Fig. 5F). No significant differences were observed between the ntrc and ntrc-trxl2.2 mutants when grown under a long-day photoperiod (Supplementary Fig. S10A–C). However, when grown under short-day photoperiods, conditions that aggravate the growth inhibition phenotype of the ntrc mutant (Pérez-Ruiz et al., 2006), the ntrc-trxl2.2 double mutant showed a partial, but significant, suppression of the ntrc growth phenotype (Supplementary Fig. S10D, E) without affecting chlorophyll content (Supplementary Fig. S10F), whereas the trxl2.2 mutant showed slightly better performance than the WT (Supplementary Fig. S10D–F). Nevertheless, the depletion of TRX L2.2 in the ntrc background led to minor effects on photosynthetic parameters, with the ntrc-trxl2.2 double mutant exhibiting slightly reduced levels of Y(II) (Supplementary Fig. S10G), moderately increased Y(NPQ) (Supplementary Fig. S10H), and decreased ETR (Supplementary Fig. S10I) compared with the ntrc mutant. Overall, the partial suppression of the ntrc phenotype shown by the ntrc-trxl2.2 mutant resembles the recently reported suppressed phenotype of the ntrc-trxl2.1 mutant (Yokochi et al., 2021), suggesting that TRX L2.1 and TRX L2.2, which share high similarity at the amino acid level (Supplementary Fig. S1), have redundant functions. Thus, the ntrc-trxl2.2 mutant was not further analyzed.

Fig. 5.

Growth phenotype and photosynthetic performance of mutants combining the deficiencies of NTRC and ACHT3 or ACHT4. Arabidopsis WT and mutants ntrc, ntrc-acht3, and ntrc-acht4 were grown under long-day conditions for 4 weeks at a light intensity of 120 μmol m⁻² s⁻¹ (A). The fresh weight (B) and chlorophyll content (C) of rosette leaves were determined from at least 10 plants grown as in (A) and represented as average values ±SEM. Different letters above histograms indicate significant differences by one-way ANOVA with a Tukey’s multiple comparison test (P<0.05). Measurements of quantum yields of PSII photochemistry [Y(II)] (D), non-photochemical quenching [Y(NPQ)] (E), and linear photosynthetic electron transport rate [ETR(II)] at increasing photosynthetic active radiation (PAR) (F) were performed using a pulse-amplitude modulation fluorometer (IMAGING-PAM). Values are the average of at least four determinations, and SEMs are represented as error bars. White and black blocks indicate periods of illumination with actinic light (81 μmol m⁻² s⁻¹) and darkness, respectively.

To test the effects of the lack of atypical TRXs in the 2cpab and ntrc genetic backgrounds, the redox states of γATPase, RCA, and FBPase in response to darkness and light were determined in the combined mutants. In light-to-dark transitions, the 2cpab mutant showed the expected delay of γATPase oxidation (Fig. 6A, B), which was lower for RCA (Fig. 6A, C), and much lower for FBPase, the enzyme that shows the most rapid rate of oxidation (Fig. 6A, D); however, no additional effects were observed in any of the triple mutants in the 2cpab background (Fig. 6A–D). When dark-dependent oxidation was analyzed in the ntrc backgrounds, no significant differences between the WT and the mutants were observed in the rates of oxidation of γATPase (Fig. 6A, E) and FBPase (Fig. 6A, G). However, it should be noted that the latter was reduced to varying degrees after light adaptation at high light intensity (650 μmol m⁻² s⁻¹), being lower in the ntrc (∼75%), ntrc-acht3 (∼60%), and ntrc-acht4 (∼50%) mutants than in the WT (∼90%) (Fig. 6A, G time 0). In contrast to γATPase and FBPase, the oxidation of RCA in the dark was accelerated in ntrc, ntrc-acht3, and ntrc-acht4 mutants (Fig. 6A, F).

Fig. 6.

Effect of the combined deficiencies of 2-Cys PRXs, NTRC, and atypical TRXs ACHT3 and ACHT4 on the dark-dependent redox state of γATPase, RCA, and FBPase. Arabidopsis WT, 2cpab, 2cpab-acht3, 2cpab-acht4, ntrc-acht3, and ntrc-acht4 mutants were grown under long-day conditions for 4 weeks at a light intensity of 120 μmol m⁻² s⁻¹. (A) The redox state of γATPase, RCA, and FBPase, as indicated, were determined in dark-adapted plants that were exposed to a light intensity of 650 μmol m⁻² s⁻¹ during 30 min (L) and then subjected to 5, 10, and 15 min of darkness. Molecular mass markers (kDa) are indicated on the left and vertical lines indicate different blots. For γATPase (B, E), RCA (C, F), and FBPase (D, G), bands corresponding to the reduced and oxidized forms were quantified (Gel Analyzer) and the percentage reduction was calculated as the ratio between the reduced form and the sum of reduced and oxidized forms. The percentage reduction is represented as the mean ±SEM of three independent experiments. Different letters indicate significant differences by one-way ANOVA with a Tukey’s multiple comparison test (P<0.05). red, reduced; oxi, oxidized; r.i., redox insensitive isoform.

Next, the effect of the lack of atypical TRXs ACHT3 and ACHT4 in the 2cpab and ntrc backgrounds on the light-dependent reduction of γATPase, RCA, and FBPase was analyzed. As expected, the 2cpab mutant showed higher levels of light-dependent reduction than the WT for the three enzymes, an effect which was less evident in the case of the γATPase subunit than in RCA and FBPase (Fig. 7A, B) due to the rapid response to light of the former. However, no remarkable differences were observed in the level of reduction of these enzymes in 2cpab-acht3 and 2cpab-acht4 plants when compared with the 2cpab mutant (Fig. 7A, B). In contrast, the levels of light-dependent reduction of RCA and FBPase enzymes were significantly decreased in the ntrc mutant, but no additional differences were observed in the ntrc-acht3 and ntrc-acht4 mutants (Fig. 7C, D).

Fig. 7.

Effect of the combined deficiencies of 2-Cys PRXs, NTRC, and atypical TRXs ACHT3 and ACHT4 on the light-dependent redox state of γATPase, RCA, and FBPase. Arabidopsis WT and mutants 2cpab, 2cpab-acht3, 2cpab-acht4, ntrc, ntrc-acht3, and ntrc-acht4 were grown under long-day conditions for 4 weeks at a light intensity of 120 μmol m⁻² s⁻¹. The light-dependent redox states of γATPase, RCA, and FPBase (A, C) were determined at the end of the dark period (D), and after 30 min of illumination at 120 μmol m⁻² s⁻¹ (L) by labeling of the thiol groups with the alkylating agent iodoacetamide. (B, D) For each enzyme, as indicated, bands corresponding to the reduced and oxidized forms were quantified (Gel Analyzer) and the percentage reduction was calculated as the ratio between the reduced form and the sum of reduced and oxidized forms. The percentage reduction is represented as the mean ±SEM of three independent experiments. Molecular mass markers (kDa) are indicated on the left. Letters indicate significant differences by one-way ANOVA with a Tukey’s multiple comparison test (P<0.05). red, reduced; oxi, oxidized; r.i., redox insensitive isoform.

The contribution of atypical TRXs ACHT3, ACHT4, and TRX L2.2 to plant acclimation to light conditions

The finding of altered redox regulation of chloroplast enzymes in the mutants under analysis prompted us to explore whether these atypical TRXs might be relevant for plant acclimation to light conditions. To test this possibility, single and multiple mutants were grown under different light regimes and the fresh weight of the rosette leaves was determined as a readout of plant growth, taking the long-day photoperiod (Fig. 8A) as standard conditions for comparison. No negative effect on growth in any of the mutants was observed when plants were grown under a short-day photoperiod, which showed a significant beneficial effect for the acht3, acht4, acht4-trxl2.2, and acht3-acht4 mutants (Fig. 8B). Similarly, no negative effects were observed when plants were grown under continuous light (120 μmol m⁻² s⁻¹) (Fig. 8C). However, conditions of a long-day photoperiod with higher light intensity (525 μmol m⁻² s⁻¹) were slightly detrimental, more significantly for the acht3-acht4-trxl2.2 mutant lacking the three atypical TRXs under analysis (Fig. 8D). Remarkably, fluctuating light (357 μmol m⁻² s⁻¹, 1 min; 49 μmol m⁻² s⁻¹, 8 min) during the light period of the long-day photoperiod was beneficial for all the mutants, but more significantly for the acht3-trxl2.2 and acht4-trxl2.2 mutants (Fig. 8E).

Fig. 8.

Growth phenotype of single and multiple mutants in different light conditions. Arabidopsis WT and mutant plants were grown under long-day conditions for 5 d at a light intensity of 120 μmol m⁻² s⁻¹. Then, plants were grown under different photoperiods and light intensities: long-day, 120 μmol m⁻² s⁻¹ (A); short-day, 120 μmol m⁻² s⁻¹ (B); continuous light, 120 μmol m⁻² s⁻¹ (C); long-day, 525 μmol m⁻² s⁻¹ (D); and fluctuating light, 357 μmol m⁻² s⁻¹ for 1 min and 49 μmol m⁻² s⁻¹ for 8 min (E). The fresh weight of rosette leaves, determined from at least 10 plants grown for 12 d (A, C–E) or 20 d (B), is represented as average values ±SEM. Different letters above histograms indicate significant differences by one-way ANOVA with a Tukey’s multiple comparison test (P<0.05).

Discussion

The high content of TRXs in plant chloroplasts, including both typical and atypical isoforms (Chibani et al., 2021), raises the question of why these organelles have such a complex redox regulatory network. Previous reports suggested the participation of atypical TRXs ACHT1, ACHT2, and TRX L2.1 in oxidation of plastid enzymes (Dangoor et al., 2012; Yokochi et al., 2021). However, chloroplasts contain additional atypical TRXs such as ACHT3, ACHT4, and TRX L2.2, which have not been previously analyzed in the context of the redox network of the organelle. Thus, with the aim of establishing the contribution of atypical TRXs to chloroplast redox regulation, in this work we have analyzed Arabidopsis mutants combining the deficiencies in these TRXs and their functional relationship with the NTRC–2-Cys PRXs redox system.

Atypical TRXs ACHT3, ACHT4, and TRX L2.2 contribute to chloroplast enzyme oxidation through 2-Cys PRXs

The finding that single mutants acht3, acht4, and trxl2.2 (Supplementary Fig. S5A–C) and double and triple mutants combining these deficiencies (Fig. 1A–C) showed WT-like growth phenotypes and minor alterations of photosynthetic performance (Supplementary Fig. S5D–F; Fig. 1D–F) indicates the dispensable role of these TRXs for plant growth, unless under standard growth conditions. These results are in line with the previously reported WT-like phenotypes of Arabidopsis mutants lacking atypical TRXs ACHT1, ACHT2 (Dangoor et al., 2012; Yokochi et al., 2021), and TRX L2.1 (Yokochi et al., 2021), as well as mutants lacking typical TRXs m (Serrato et al., 2021), f (Yoshida et al., 2015; Naranjo et al., 2016b), x (Pulido et al., 2010), and y (Laugier et al., 2013; Jurado-Flores et al., 2020). However, it should be noted that despite the low effect on growth, the lack of any of these TRXs affects chloroplast performance. This is the case of mutants lacking TRXs f, which show impaired light-dependent reduction of CBC enzymes (Yoshida et al., 2015; Naranjo et al., 2016b), TRX m4, which show defective photosynthetic cyclic electron flow (Courteille et al., 2013; Okegawa and Motohashi, 2020), or TRXs x and y, having abnormal function of PSI (Okegawa et al., 2023). Altogether, these results suggest the concerted action of different TRXs so that the lack of a single TRX has a minor impact on chloroplast performance and plant growth. Supporting this notion, the participation of typical TRXs f and m (Geigenberger et al., 2005), atypical TRX ACHT4 (Eliyahu et al., 2015), and NTRC (Michalska et al., 2009; Lepistö et al., 2013) in the redox regulation of starch synthesis has been shown.

Despite the low impact on growth, the lack of atypical TRXs ACHT3, ACHT4, and TRX L2.2 affects redox regulation of chloroplast enzymes. By analogy with mutants lacking 2-Cys PRXs, which show delayed enzyme oxidation in the dark (Ojeda et al., 2018; Vaseghi et al., 2018; Yoshida et al., 2018), the delayed oxidation of RCA (Fig. 2A, C) and γATPase (Fig. 2A, B) in the acht3, acht4, and trxl2.2 single mutants, which were increased in the combined mutants (Fig. 2A, E, F), indicates the participation of ACHT3, ACHT4, and TRX L2.2 in chloroplast enzyme oxidation. These results agree with the oxidizing function previously proposed for ACHT1 and ACHT2 (Dangoor et al., 2012; Yokochi et al., 2021), ACHT4 (Eliyahu et al., 2015), and TRX L2.1 (Yoshida et al., 2019; Yokochi et al., 2021; Sekiguchi et al., 2022), in line with their less negative redox potential (Yoshida et al., 2018; Yokochi et al., 2019). Furthermore, the finding that 2cpab-acht3 and 2cpab-acht4 mutants show phenotypes like the 2cpab mutant in terms of growth (Fig. 4A–C), photosynthetic performance (Fig. 4D–F), and dark- (Fig. 6A–D) and light-dependent (Fig. 7A, B) redox changes of γATPase, RCA, and FBPase suggests that the oxidizing function of ACHT3 and ACHT4 is exerted through 2-Cys PRXs, in agreement with previous reports of ACHT4 function (Dangoor et al., 2012; Eliyahu et al., 2015).

NTRC affects the redox regulatory activity of ACHT3, ACHT4, and TRX L2.2

Based on the oxidizing function proposed for ACHT3, ACHT4, and TRX L2.2, higher levels of light-dependent enzyme reduction in mutants devoid of these TRXs would be expected, as occurs in mutants lacking 2-Cys PRXs (Fig. 7A, B; Ojeda et al., 2021). On the contrary, though no differences were found in light-dependent γATPase reduction (Fig. 3A, B), probably due to the rapid response of this enzyme to light, the lower light-dependent reduction of RCA and FBPase in the acht3, acht4, and trxl2.2 mutants (Fig. 3A, B) shows the contribution of ACHT3, ACHT4, and TRX L2.2 to enzyme reduction in the light, a notion further supported by the impaired light-dependent reduction of 2-Cys PRXs in the single (Supplementary Fig. S8A, B) and combined mutants (Fig. 3E, F), as previously reported for the mutant lacking ACHTs 1 and 2 (Yokochi et al., 2021). Thus, the question arising is how atypical TRXs affect both dark-dependent enzyme oxidation and light-dependent enzyme reduction. By analogy with the ntrc mutant, which shows impaired light-dependent reduction of chloroplast enzymes (Fig. 7C, D), as previously reported (Thormählen et al., 2015; Ojeda et al., 2017, 2021), we propose that the lack of ACHT3, ACHT4, or TRX L2.2 increases drainage of reducing equivalents from typical TRXs m and f via 2-Cys PRXs, thereby impairing RCA and FBPase reduction in the light. Notably, the ntrc mutant showed further decreased levels of reduced RCA and FBPase compared with the acht3, acht4, and trxl2.2 mutants, whereas no additional drop in the level of reduction of these enzymes was observed in the ntrc-acht3 and ntrc-acht4 mutants (Fig. 7C, D). Therefore, ACHT3, ACHT4, and TRX L2.2 contribute to the maintenance of the redox balance of 2-Cys PRXs during the daytime, albeit to a lesser extent than NTRC. Intriguingly, the levels of reduction of RCA and FBPase were not further impaired in double and triple mutants combining ACHT3, ACHT4, and TRX L2.2 deficiencies, which essentially behaved like single mutants (Fig. 3C, D). A possibility to explain these results is that in the absence of atypical TRXs, the draining of reducing power from typical TRXs toward 2-Cys PRXs during the day is limited by NTRC.

The combined deficiencies of ACHT3, ACHT4, or TRX L2.2 in the ntrc background allowed the contribution of these TRXs to chloroplast redox regulation to be more precisely established. The partial suppression of the ntrc phenotype shown by the ntrc-trxl2.2 mutant when grown under a short-day photoperiod (Supplementary Fig. S10D, E) resembles the suppressor phenotype of the ntrc-trxl2.1 mutant (Yokochi et al., 2021), indicating equivalent functions of TRXs L2.1 and L2.2 in enzyme oxidation, in line with their in vitro activity (Yoshida et al., 2018). Moreover, the fact that the TRXL2.1 gene is expressed at higher levels than the TRXL2.2 gene (Supplementary Fig. S3A, B) may explain the different degrees of phenotypic suppression observed in ntrc-trxl2.2 (Supplementary Fig. S10D, E) and in ntrc-trxl2.1 (Yokochi et al., 2021). Notably, the Fdx/FTR pathway is unable to reduce TRX L2.1 and TRX L2.2 (Yokochi et al., 2021), but it is unknown whether ACHTs are reduced by this pathway. The similar phenotypes of the ntrc-acht3 and the ntrc mutants (Fig. 5A–C) are probably due to the low level of expression of the ACHT3 gene in leaves (Supplementary Fig. S4A). In contrast, the aggravated phenotype of the ntrc-acht4 mutant, as compared with the ntrc mutant (Fig. 5A–C), resembles the phenotypes of mutants combining the lack of NTRC and typical TRXs, hence suggesting a more relevant impact of ACHT4 on the reducing component of chloroplast redox regulation. In this regard, the ntrc-acht4 mutant resembles the ntrc-trxx mutant, which also shows a more aggravated phenotype than the ntrc mutant (Ojeda et al., 2021), despite the low abundance of TRX x in Arabidopsis chloroplasts (Okegawa and Motohashi, 2015). TRX x and ACHT4 have in common that they are efficient reductants of 2-Cys PRXs (Collin et al., 2003; Dangoor et al., 2009), suggesting that the simultaneous lack of NTRC and an efficient reductant of 2-Cys PRXs leads to severe effects, probably by affecting the reducing capacity of the rest of the chloroplast TRXs.

Atypical TRXs contribute to plant acclimation to light conditions

The WT-like phenotype of the single and multiple mutants analyzed in this work, when grown under a standard long-day photoperiod (Supplementary Fig. S5A–C; Fig. 1A–C), raises the question of the actual impact of atypical TRXs ACHT3, ACHT4, and TRX L2.2 on plant growth. The slight negative effect of high light intensity, most significantly in the case of the acht3-acht4-trxl2.2 triple mutant (Fig. 8D), which shows impaired ETR(II) as light intensity increases (Fig. 1F), suggests that these atypical TRXs might be relevant for acclimation to different light environments. Further supporting this notion, most of the mutants outperform WT plants under fluctuating light consisting of cycles of moderately high (357 μmol m⁻² s⁻¹, 1 min) and low (49 μmol m⁻² s⁻¹, 8 min) light intensities (Fig. 8E). This suggests that under low light, the lack of the oxidizing contribution caused by the absence of these atypical TRXs allows a higher level of reduction, hence activation, of biosynthetic enzymes, improving growth under these conditions. Remarkably, the Arabidopsis mutant devoid of 2-Cys PRXs, a central component of chloroplast oxidation, shows improved growth under fluctuating light (Vaseghi et al., 2018).

Conclusion

In summary, the results presented in this work reveal the contribution of atypical TRXs ACHT3, ACHT4, and TRX L2.2 to fine-tuning the redox state of chloroplast enzymes in response to changes in light intensity and darkness, as depicted in the scheme shown in Fig. 9. In the dark, atypical TRXs contribute to oxidation by directly transferring reducing equivalents from enzyme thiols to 2-Cys PRXs, which reduces hydrogen peroxide to water (Fig. 9). Though it was reported that TRXs f can mediate enzyme oxidation in vitro (Ojeda et al., 2018; Yokochi et al., 2019) and in vivo (Yokochi et al., 2021), the physiological relevance of typical TRXs in enzyme oxidation remains to be determined. In the light, photosynthetically reduced Fdx fuels reducing equivalents to the abundant TRXs m and f via FTR (Yoshida and Hisabori, 2017), maintaining reduced, hence active, redox-regulated chloroplast biosynthetic enzymes (Fig. 9). TRX L2.2 is not reduced by FTR (Yokochi et al., 2021), but it is unknown whether ACHTs are FTR targets. NTRC, which uses NADPH generated by Fdx-NADP reductase (FNR) to reduce 2-Cys PRXs, avoids drainage of reducing power from typical TRXs. Since atypical TRXs are more efficient reductants of 2-Cys PRXs than typical TRXs f and m, the lack of atypical TRXs would increase drainage of electrons of typical TRXs via 2-Cys PRXs, hence impairing reduction of their targets (Fig. 9). While it is assumed that the high efficiency of 2-Cys PRXs to reduce hydrogen peroxide allows the rapid oxidation of chloroplast enzymes in the dark, the mechanism that limits this activity in the light, hence avoiding complete enzyme oxidation during the day, remains unsolved. Finally, though atypical TRXs ACHT3, ACHT4, and TRX L2.2 are dispensable for growth under a standard long-day photoperiod, their function may become relevant for fine-tuning photosynthetic metabolism in response to changes of light intensity in a natural environment. Therefore, the results presented in this report add a new layer of complexity to the redox regulatory machinery of the chloroplast, which is the result of the contribution of multiple TRXs with functions which are to some extent redundant. Because the balance between reduction and oxidation allows the readjustment of enzyme activities, it is likely that the complex set of chloroplast TRXs might be an adaptative advantage to cope with the continuous changes of environmental conditions that plants, as sessile organisms, must face in nature.

Fig. 9.

A proposed model of the contribution of atypical TRXs to chloroplast redox regulation. In the light, the photosynthetic electron transport chain produces reduced Fdx (Fdx_RED), which reduces typical TRXs (m and f types being the most abundant) via FTR. It remains unknown whether atypical TRXs are reduced via FTR. The efficient reduction of 2-Cys PRXs by atypical TRXs avoids excess draining of reducing power from typical TRXs. NADPH generated by the action of FNR is used by NTRC to reduce 2-Cys PRXs, hence counterbalancing drainage of reducing power from the pool of TRXs, thereby maintaining reduced, and active, chloroplast biosynthetic enzymes. In the dark, atypical TRXs mediate the oxidation of chloroplast enzymes via 2-Cys PRXs, which reduce hydrogen peroxide to water. More work is needed to establish the contribution of typical TRXs to enzyme oxidation in the dark.

Abbreviations

ACHT

atypical cysteine- and histidine-rich thioredoxin

AGPase

ADP-glucose pyrophosphorylase

γATPase

γ subunit of ATPase

FBPase

fructose-1,6- bisphosphate phosphatase

Fdx

ferredoxin

FNR

Fdx-NADP reductase

NTRC

NADPH-thioredoxin reductase C

PRX

peroxiredoxin

RCA

Rubisco activase

TRX

thioredoxin

Supplementary data

The following supplementary data are available at JXB  online.

Fig. S1. Amino acid sequence comparison of ACHT1, ACHT2, ACHT3, ACHT4, TRX L2.1, and TRX L2.2 from Arabidopsis.

Fig. S2. Expression pattern of ACHT1–ACHT4 genes during Arabidopsis development.

Fig. S3. Expression pattern of TRX L2.1 and TRX L2.2 genes during Arabidopsis development.

Fig. S4. Expression analysis and isolation of Arabidopsis T-DNA insertion mutants of ACHT3, ACHT4, and TRXL2.2 genes.

Fig. S5. Growth phenotype and photosynthetic performance of the mutants acht3, acht4, and trxl2.2.

Fig. S6. Pollen development and seed germination capacity in single and multiple mutants combining the deficiencies of ACHT3, ACHT4, and TRX L2.2.

Fig. S7. Genotyping of T-DNA insertions in double and triple mutants deficient in ACHT3, ACHT4, and TRX L2.2.

Fig. S8. Effect of light and darkness on the redox state of 2-Cys PRXs in WT and single mutants.

Fig. S9. Genotyping and western blot analysis of multiple mutants combining the deficiencies of 2-Cys PRXs or NTRC and ACHT3, ACHT4, or TRX L2.2.

Fig. S10. Growth phenotype and photosynthetic performance of mutants deficient in NTRC and TRX L2.2.

Table S1. Oligonucleotides used in this study.

Author contributions

JMP-R and FJC: conceptualization, investigation, resources, supervision, and funding acquisition; JJ-L, AC, and AG-R: methodology; JJ-L and AC: data curation; FJC: writing—original draft; JJ-L, AC, JMP-R, and FJC: formal analysis and writing—review & editing.

Funding

This work was supported by grant PID2023-146573NB-I00, funded by Ministerio de Ciencia, Innovación y Universidades (MICIU)/Agencia Estatal de Investigación (AEI)/10.13039/501100011033 and by European Regional Development Fund (ERDF)/European Union (EU). AC was supported by a Formación de Profesorado Universitario (FPU) pre-doctoral contract (FPU18-03035) from Ministerio de Universidades (Spain).

Data availability

The primary data supporting this study were not made publicly available at the time of publication. The underlying data are available from the corresponding author, Francisco J. Cejudo, on request.