Type-f thioredoxins have a role in the short-term activation of carbon metabolism and their loss affects growth under short-day conditions in Arabidopsis thaliana

Journal of Experimental Botany, Vol. 67, No. 6 pp. 1951–1964, 2016 doi:10.1093/jxb/erw017 Advance Access publication 2 February 2016 This paper is available online free of all access charges (see http://jxb.oxfordjournals.org/open_access.html for further details) RESEARCH PAPER Type-f thioredoxins have a role in the short-term activation of carbon metabolism and their loss affects growth under short-day conditions in Arabidopsis thaliana BelénNaranjo1, AntonioDiaz-Espejo2, MarikaLindahl1 and Francisco JavierCejudo1,* 1 Instituto de Bioquímica Vegetal y Fotosíntesis, Universidad de Sevilla and CSIC, Avda Américo Vespucio, 49, 41092-Sevilla, Spain 2 Instituto de Recursos Naturales y Agrobiología de Sevilla, CSIC, Avda Reina Mercedes, 10, 41012-Sevilla, Spain * Correspondence: [email protected] Received 21 October 2015; Accepted 5 January 2016 Editor: Christine Raines, University ofEssex Abstract Redox regulation plays a central role in the adaptation of chloroplast metabolism to light. Extensive biochemical analyses in vitro have identified f-type thioredoxins (Trxs) as the most important catalysts for light-dependent reduction and activation of the enzymes of the Calvin–Benson cycle. However, the precise function of type f Trxs in vivo and their impact on plant growth are still poorly known. To address this issue we have generated an Arabidopsis thaliana double knock-out mutant, termed trxf1f2, devoid of both f1 and f2 Trxs. Despite the essential function previously proposed for f-type Trxs, the visible phenotype of the trxf1f2 double mutant was virtually indistinguishable from the wild type when grown under a long-day photoperiod. However, the Trx f-deficient plants showed growth inhibition under a short-day photoperiod which was not rescued at high light intensity. The absence of f-type Trxs led to significantly lower photosynthetic electron transport rates and higher levels of non-photochemical energy quenching. Notably, the Trx f null mutant suffered from a shortage of photosystem Ielectron acceptors and delayed activation of carbon dioxide fixation following a dark–light transition. Two redox-regulated Calvin–Benson cycle enzymes, fructose 1,6-bisphosphatase (FBPase) and Rubisco activase, showed retarded and incomplete reduction in the double mutant upon illumination, compared with wild-type plants. These results show that the function of f-type Trxs in the rapid activation of carbon metabolism in response to light is not entirely compensated for by additional plastid redox systems, and suggest that these Trxs have an important role in the light adjustment of photosynthetic metabolism. Key words: Carbon assimilation, chloroplast, fructose 1,6 bisphosphatase, photosynthesis, redox regulation, thioredoxin. Introduction Chloroplasts are essential for plant life because these organelles perform photosynthesis, the process that allows the conversion of light energy into biomass with the concomitant production of molecular oxygen. In addition, chloroplasts act as sensors of environmental conditions, particularly light quantity and quality, thus playing an important role in harmonizing the growth of plant photosynthetic and non-photosynthetic tissues as well as the adaptation of plants to the environment (Jarvis and López- Juez, 2013). To meet these requirements, chloroplast metabolism needs to respond rapidly to external and internal signals, redox regulation being an important aspect of this adaptability. Redox regulation is a post-translational modification consisting of the dithiol-disulphide interchange of selected and well-conserved cysteine residues of proteins. It is thus a reversible mechanism that allows the rapid This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0/), which permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited. © The Author 2016. Published by Oxford University Press on behalf of the Society for Experimental Biology. 1952 | Belén Naranjo etal. regulation of metabolic pathways (Buchanan etal., 2012). Protein disulphide reductases such as thioredoxins (Trxs), small polypeptides of 12–14 kDa with a highly conserved active site (WCGPC) and a characteristic structure, the socalled Trx-fold, play a central role in redox regulation. Trxs catalyse the reduction of disulphide bridges in target proteins and, as a consequence, the two cysteine residues at the Trx active site become oxidized as a disulphide bridge. Therefore, for a new catalytic cycle this disulphide needs to be reduced, reducing power being provided by NADPH in a reaction catalysed by NADPH-dependent Trx reductase (NTR). This two-component redox system formed by NTR and Trx is universally distributed in all kinds of organisms from bacteria to fungi, animals, and plants (Jacquot etal., 2009). In contrast to heterotrophic organisms and non-photo- synthetic plant tissues, redox regulation in chloroplasts has unique features. These organelles harbour a complex set of Trxs and Trx-like proteins (Lemaire et al., 2007; Chibani etal., 2009; Meyer etal., 2012; Balsera etal., 2014 ), which are not reduced by NADPH, but by ferredoxin (Fdx) reduced by photosynthetic electron transport in a process catalysed by a Fdx-dependent Trx reductase (FTR) found exclusively in plastids and cyanobacteria (Dai etal., 2007). Therefore, chloroplast redox regulation is mediated by the Fdx/FTR/Trx system and, thus, is dependent on light. Initial biochemical analyses in vitro led to the identification of two types of Trxs in chloroplasts, termed f and m, based on their ability to reduce and activate fructose-1,6-bispho- phate phosphatase (FBPase) and NADP-malic dehydrogenase (NADP-MDH), respectively (Wolosiuk et al., 1979). The availability of genome sequences from different plants has uncovered the complex set of chloroplastic Trxs, which in Arabidopsis thaliana include two isoforms of f-type, four isoforms of m-type, two isoforms of the y-type, and an x-type Trx (Lemaire etal., 2007; Meyer etal., 2012; Balsera etal., 2014). Trxs f and m were proposed to play a predominant role in the redox regulation of the central biosynthetic pathways, such as the Calvin–Benson cycle, whereas Trxs x and y show a capacity to reduce peroxiredoxins (Prxs) and so were considered to have an antioxidant function (Collin etal., 2003, 2004). In addition, a novel Trx, type-z, was identified in Arabidopsis, which is involved in the redox regulation of plastid transcription (Arsova etal., 2010; Schröter etal., 2010; Steiner etal., 2011; Wimmelbacher and Börnke, 2014). Beside these canonical Trxs, several atypical Trxs have been described in the chloroplast. This is the case of HCF164, which is localized in the thylakoid membrane facing the lumen (Motahashi and Hisabori, 2006), and the small family of atypical Trxs identified in Arabidopsis termed AtACHTs (for atypical Cys His-rich Trxs) (Dangoor etal., 2009). Finally, different Trx-like proteins were identified in chloroplasts, among which the so-called CDSP32 is the best characterized (Broin etal., 2002). In addition, chloroplasts harbour an NADPH- dependent redox system based on the activity of a bimodular enzyme consisting of an NTR with a joint Trx domain at the C-terminus, termed NTRC (Serrato et al., 2004; Kirchsteiger etal., 2012). NTRC is able to conjugate both NTR and Trx activities for the efficient reduction of 2-Cys Prxs (Pérez-Ruiz etal., 2006; Moon etal., 2006; Alkhalfioui et al., 2007), thus suggesting an antioxidant function for this enzyme (Puerto-Galán etal., 2013). Additional reports indicate a function of NTRC in the redox regulation of starch synthesis (Michalska etal., 2009; Lepistö etal., 2013) and in different reactions of the biosynthesis of tetrapyrroles (Richter etal., 2013; Pérez-Ruiz etal., 2014). The high affinity of NTRC for NADPH introduces the notion that redox regulation in chloroplasts relies not only on photosynthetically reduced Fdx but also on NADPH which can be produced during the night from sugars by the oxidative pentose phosphate pathway (Spínola etal., 2008; Cejudo etal., 2012). In parallel with the knowledge of the increasing complexity of the plastidial Trxs, the advance in proteomics has allowed the identification of a large number of putative targets of Trxs (Balmer etal., 2003; Buchanan and Balmer, 2005; Montrichard etal., 2009), which indicates that redox regulation is important for virtually any process taking place in the chloroplast. However, the question of the level of specificity or redundancy of the different Trxs in redox regulation in this organelle is still poorly understood. An Arabidopsis knockout mutant deficient in Trx z showed a severe phenotype indicating that the function of this Trx in chloroplast transcription is not redundant with other plastidial Trxs (Arsova etal., 2010). By contrast, an Arabidopsis knockout mutant lacking Trx x shows almost a wild-type phenotype, indicating that this deficiency is compensated for by other plastidial redox systems (Pulido etal., 2010). More uncertain is the specificity of Trxs with several isoforms, such as those of type m or f, which are considered to play a relevant role in the redox regulation of photosynthetic metabolism. Trx m4 is involved in alternative photosynthetic electron transport (Courteille et al., 2013), and the simultaneous deficiency of Trxs m1, m2, and m4 resulted in impaired photosystem II biogenesis (Wang etal., 2013). Proteomic and biochemical studies in vitro have shown the relevant function of type-f Trxs in the redox regulation of most of the enzymes of the Calvin–Benson cycle (Michelet et al., 2013). In addition, an Arabidopsis mutant lacking Trx f1 shows impaired light-dependent reductive activation of ADP-glucose pyrophosphorylase (AGPase) and starch turnover (Thormählen etal., 2013). Surprisingly, despite the key role proposed for type-f Trxs in the redox regulation of chloroplast metabolism, the Trx f1-deficient mutant shows a wild-type phenotype (Thormählen et al., 2013). Moreover, a double mutant devoid of both Trx f1 and Trx f2 showed a visible phenotype indistinguishable from wild-type plants (Yoshida etal., 2015), suggesting that other plastidial redox systems are able to compensate for the deficiency of f-type Trxs. Therefore, with the aim of establishing the role of type f Trxs in chloroplast performance and plant growth, we generated a double knockout mutant of Arabidopsis devoid of Trx f1 and Trx f2, and have performed the analysis of its phenotype under different growth conditions. The double mutant shows no visible phenotype, compared with wild-type plants, Effects of f-type thioredoxins deficiency in Arabidopsis | 1953 when grown under long-day conditions, however, it shows retarded growth under short-day conditions which is not rescued by high light intensity. Analysis of photosynthetic parameters and changes in the redox status of FBPase and Rubisco activase in response to light showed that type-f Trxs are required for the rapid reduction of the Calvin–Benson cycle enzymes in response to light, a function not compensated for by other plastidial redox systems. Materials and methods Growth conditions and plant material Arabidopsis thaliana wild-type (ecotype Columbia) and mutant plants were grown in soil in growth chambers under long-day (16/8 h light/dark) or short-day (8/16 h light/dark) conditions at 22°C during the light and 20 °C during the dark periods and a light intensity of 125μE m−2 s−1. For experiments addressing the effects of irradiance, plants were grown under short-day conditions at 125, 350, and 950μE m−2 s−1 light intensity. Ahomozygous line, GK-020E05-013161, with a T-DNA insertion in the TRX f2 gene (see Supplementary Fig. S1 at JXB online) from Arabidopsis, termed the trxf2 mutant, was selected by PCR analysis with the oligonucleotides described in Supplementary Table S1. This mutant was manually crossed with the trxf1 mutant, which was previously reported by Pérez-Ruiz et al. (2014). Seeds resulting from this cross were tested for heterozygosity of the T-DNA insertions in the TRX f1 and TRX f2 genes. Plants were then self-crossed and double homozygous lines were identified in the progeny by PCR analysis of genomic DNA using the oligonucleotides described in Supplementary Table S1. RNA extraction and qRT-PCR analysis Total RNA was extracted using the TRIsure RNA extraction reagent (BIOLINE) and cDNA synthesis was performed with 1 μg of total RNA using the Maxima first strand cDNA synthesis kit (Thermo Scientific) according to the manufacturer’s instructions. The content of Trx f1 and Trx f2 transcripts was determined by real time quantitative PCR (qRT-PCR) with RNA samples extracted from seedlings. qRT-PCR was performed with oligonucleotides shown in Supplementary Table S2 in an IQ5 real-time PCR detection system (Bio-Rad) following a standard thermal profile (95°C, 3 min, 40 cycles of 95°C for 10 s, and 60°C for 30 s). The relative level of each transcript was referred to the level of the UBIQUITIN transcript. Protein extraction, alkylation assays, and Western blot analysis For protein extraction, leaves were ground in liquid nitrogen and 10% (v/v) trichloroacetic acid (TCA) was immediately added to quench thiol oxidation. Samples were incubated on ice for 20 min and then centrifuged at 16 200 g at 4°C for 10 min. The pellets were washed with acetone, resuspended in alkylation buffer (2% SDS, 50 mM TRIS–HCl pH 7.8, 2.5% glycerol, and 4 M urea) with 10 mM methyl-maleimide polyethylene glycol (MM-PEG24) and incubated for protein thiol alkylation for 20 min at room temperature. Samples were subjected to SDS-PAGE (9.5% polyacrylamide), transferred on to nitrocellulose membranes, and probed with an anti-FBPase antibody which was kindly provided by Dr M Sahrawy, Estación Experimental del Zaidín, Granada, Spain, or with an anti-Rubisco activase antibody which was kindly provided by Dr AR Portis, USDA, Urbana, USA. The anti-Trx f antibody was raised by rabbit immunization with purified recombinant Trx f1 from Arabidopsis at the Servicio de Producción Animal, University of Seville, Spain. This antibody also detected Trx f2 although with somewhat lower efficiency (see Supplementary Fig. S2). Measurements of chlorophyll a fluorescence and P700 absorbance Room temperature chlorophyll fluorescence was measured using a pulse-amplitude modulation fluorimeter (DUAL-PAM-100, Walz, Effeltrich, Germany). The maximum PSII quantum yield, determined as variable fluorescence (Fv) to maximal fluoresence (Fm), Fv/Fm, was measured after the dark adaptation of the plants for 30 min and a single saturating pulse of red (635 nm) light at 10 000μEm−2s−1 was applied. Induction-recovery curves were performed using red (635 nm) actinic light at 75μE m−2 s−1 for 8 min. Saturating pulses of red light at 10 000μE m−2 s−1 intensity and 0.6 s duration were applied every 60 s and recovery in darkness was recorded for up to 10 min. The parameters Y(II) and Y(NPQ) corresponding to the respective quantum yields of PSII photochemistry and non-photo- chemical quenching (NPQ) were calculated by the DUAL-PAM-100 software according to the equations in Kramer etal. (2004). Relative linear electron transport rates were measured in leaves of pre-illumi- nated plants by applying stepwise increasing actinic light intensities up to 2 000μE m−2 s−1. The redox state of photosystem IP700 was monitored by following the changes in absorbance at 830 nm versus 875 nm using the DUAL-PAM-100. Plants were kept in the dark for 30 min and then, to probe the maximum extent of P700 oxidation, leaves were illuminated with far red (730 nm) light superimposed on the actinic light. Thereafter, absorbance traces were recorded during a 5 min illumination with 126μE m−2 s−1 red (635 nm) actinic light followed by 5 min darkness. Saturating pulses of red light at 10 000μE m−2 s−1 were applied every 20 s. The quantum yields of PSI photochemistry Y(I), donor side limitations Y(ND), and acceptor side limitations Y(NA) were based on saturating pulse analysis and calculated by the DUAL-PAM-100 software. Determination of the rate of carbon assimilation AN Net CO2 assimilation rate (AN) was measured using an open gas exchange system Li-6400 equipped with the chamber head (Li- 6400–40). All measurements were conducted in dark-adapted leaves of short-day grown plants (50 d-old) at 500μmol mol−1 of CO2, a constant leaf temperature of 20°C, and a vapour pressure deficit between leaf and air of below 1 kPa. Before the light, at an intensity of 70μE m−2 s−1, was turned on in the leaf chamber, AN was recorded for 30 min every 5 s, then recording continued until equilibrium was reached. Six leaves were measured per line. Determination of chlorophylls Leaf discs were weighed and incubated in 1 ml methanol overnight at 4°C. After extraction, chlorophyll levels were measured spectrophotometrically, as described in Porra etal. (1989), and normalized to fresh weight or leaf area. The values were compared with a Tukey Test (Anova) using a confidence interval of 99%. Determination of starch content Starch content was determined in leaves of wild-type and the Trx f-deficient mutant grown under long-day conditions harvested before flowering (22–26-d-old), essentially as described in Lin etal. (1988). For starch extraction, leaves (100–200 mg fresh weight) harvested at the end of the day were ground in liquid nitrogen and washed with 80% (v/v) ethanol in 10 mM HEPES pH 7.6, for 2 h at 80°C before being washed with the same solution at room temperature until the tissue was free of pigments. Dry pellets, after centrifugation, were resuspended in 1 ml of 0.2 N KOH and heated at 100°C for 30 min. After cooling, samples were centrifuged at 17 000 g for 10 min and the supernatant was adjusted to pH 5.0 with 1 N acetic acid. An aliquot of 200μl of this solution was used to determine the amount of starch using the enzymatic method described by Lin etal. (1988). The values were compared with a Tukey Test (Anova) using a confidence interval of 99%. 1954 | Belén Naranjo etal. Results Type-f Trxs are dispensable for plantgrowth The Arabidopsis genome encodes two isoforms of f-type Trxs, f1 and f2. It was recently reported that Arabidopsis knockout mutants lacking either Trx f1 (Thormählen et al., 2013) or both Trx f1 and Trx f2 (Yoshida etal., 2015) show a wild-type phenotype with respect to growth and pigmentation. These results suggest that the loss of f-type Trxs might be compensated for by other plastidial redox systems. To address this issue in more detail, we have generated a double knockout mutant of Arabidopsis devoid of both Trx f1 and Trx f2 and have analysed its growth under different conditions. To that end, the Arabidopsis line (GK-020E05-013161) with a T-DNA insertion at the AT5G16400.1 locus encoding Trx f2 was isolated (Supplementary Fig. S1). This line was manually crossed with the Trx f1 knockout mutant (SALK_128365.45.75.x) previously reported by our group (Pérez-Ruiz et al., 2014), which was also characterized by Thormählen et al. (2013). Plants homozygous for both T-DNAs were selected by PCR analysis of genomic DNA (Supplementary Fig. S1). The double mutant, here termed trxf1f2, was effectively devoid of both Trx f1 and Trx f2 as shown by the lack of transcripts of the two genes, based on qRT-PCR (Fig.1A), and the corresponding polypeptides, as shown by Western blots probed with an antibody raised against Trx f1 (Fig.1B), which also cross-reacted with Trx f2 (Supplementary Fig. S2). These results confirmed that Trx f1 is much more abundant than Trx f2 in wild-type plants. In addition, we tested whether the absence of type-f Trxs had any effect on the expression of other plastidial redox systems, but only minor differences in the levels of the transcripts of genes encoding NTRC, type-m and x Trxs were detected (see Supplementary Fig. S3 ). The trxf1f2 double mutant, like the single mutants trxf1 and trxf2, showed the wild-type phenotype when grown under the long-day photoperiod (Fig. 2A), as confirmed by the weight of the rosette leaves (Fig.2B) and leaf chlorophyll content (Fig.2C) of mature plants immediately before bolting. To analyse in more detail the effect of Trx f deficiency on plant growth, the different lines were grown under short-day conditions. Wild-type and Trx f-deficient mutants showed an indistinguishable growth rate up to the stage of young rosette leaves (34 d of growth), as determined by fresh weight (Fig.3A, B). However, after 53 d of growth, the double mutant displayed a significant growth inhibition, as shown by the lower weight of rosette leaves (Fig.3B), although the chlorophyll content was unaffected (Fig.3C). Data of the long growth period, 53 d, are only shown for short-day (Fig.3) and not for long-day conditions (Fig.2), because the plant developmental stages, adult plants before bolting and advanced leaf senescence, respectively, are not comparable. We then analysed the effect of different light intensities on these mutants grown under short-day conditions. While at 125 and 350μE m−2 s−1 light intensity no difference was observed between the wild type and the mutant lines after 34 d of growth, Trx f-deficient mutants grown at 950μE m−2 s−1 light intensity displayed lower rosette weights than the wild-type plants (Fig.4A, B). At high light intensity the leaf chlorophyll content was decreased, but no significant differences were observed between wild-type and mutant plants (Fig.4C). Therefore, despite the central function previously attributed to type f Trxs in the redox regulation of chloroplast metabolism, these Trxs are dispensable for plant growth under the long-day photoperiod. Nevertheless, the retarded growth of the trxf1f2 double mutant under the short-day photoperiod at adult plant stages, even under high light conditions, confirms the light-dependent participation of f-type Trxs in chloroplast photosynthetic metabolism. Activation of photosynthesis upon a dark–light transition is retarded in plants lacking type fTrxs To explore the function of f-type Trxs in chloroplast metabolism, different photosynthetic parameters of plants lacking either or both of the Trx f enzymes were first examined Fig.1. The level of Trx f1 and Trx f2 transcripts and proteins in wildtype and mutant plants. (A) The content of Trx f1 and f2 transcripts was determined by qRT-PCR from total RNA, which was extracted from wild-type, trxf1, trxf2, and trxf1f2 seedlings. The pairs of oligonucleotides used for cDNA amplification are indicated in Supplementary Table S2. Transcript levels were normalized to UBIQUITIN amplification and referred to the level of TRX f1 and TRX f2 in wild-type plants. Determinations were performed three times and mean values ±SD are represented. (B) Western blot analysis of the content of type-f Trxs in wild-type and mutant plants. Proteins were extracted from leaves (100 mg fresh weight) of wild-type, trxf1, trxf2, and trxf1f2 plants grown under short-day conditions and subjected to SDS-PAGE (13% polyacrylamide) under reducing conditions, transferred to a nitrocellulose membrane, and probed with an anti-Trx f1 antibody. Ponceau staining was used as a loading control. Molecular weight markers in kDa are indicated on the left. Effects of f-type thioredoxins deficiency in Arabidopsis | 1955 using chlorophyll fluorescence. The integrity of PSII, determined as the ratio of variable to maximal fluorescence in dark-adapted leaves, was not affected in the single or double trxf1f2 mutants (Table 1). Non-photochemical quenching (NPQ) is a loss of chlorophyll fluorescence in the light which is not due to photochemistry and reflects adaptation mechanisms regulating the fraction of absorbed light that reaches the PSII reaction centre (Szabó etal., 2005; Baker, 2008). The energy-dependent quenching, qE, which involves thermal dissipation of the absorbed light energy, is the main component of NPQ in plants under moderate light intensities and depends on the proton gradient across the thylakoid membrane (Szabó et al., 2005; Baker, 2008; Ruban et al., Fig.3. Growth of Arabidopsis lines lacking type-f Trxs under a shortday photoperiod. (A) Representative plants of the wild type and mutant lines grown under short-day conditions (8/16 h light/dark, light intensity 125μE m−2 s−1) for 34 d and 53 d.(B) The weight of the rosette leaves was determined from 12 plants (34 d of growth) or seven plants (53 d of growth), except for the trxf1f2 double mutant which was determined from 14 plants in both cases. (C) Chlorophyll content was determined from leaf discs (n=6). Average values ±SD are represented. Letters indicate significant differences with the Tukey Test and a confidence interval of 99%. Fig.2. Growth of Arabidopsis lines lacking type-f Trxs under a long-day photoperiod. (A) Plants of the wild type and mutant lines grown under long-day conditions (16/8 h light/dark, light intensity 125μE m−2 s−1) for 22 d.(B) The weight of the rosette leaves was determined from nine plants, except for the trxf1f2 double mutant which was determined from 14 plants. (C) Chlorophyll content was determined from leaf discs (n=6) and average values ±SD are represented. Letters indicate significant differences with the Tukey Test and a confidence interval of 99%. 1956 | Belén Naranjo etal. 2012). Following the onset of actinic light at low or moderate intensity, a brief peak of NPQ is normally observed in wild-type plants due to transient acidification of the thylakoid lumen before activation of photosynthesis (Kalituho etal., 2007). Such an initial peak of NPQ was also found in the trxf2 mutant and a somewhat broader peak in the trxf1 mutant (Fig.5A; see Supplementary Fig. S4). By contrast, the double mutant displayed an extensive NPQ which did not relax completely even after 8 min of illumination at 75μE m−2 s−1 (Fig. 5A). As a result, the PSII effective quantum yield, Y(II), increased more slowly in the light and remained lower in the trxf1f2 double mutant (Fig.5B). In agreement with these results, the relative linear photosynthetic electron transport rates were considerably lower in the double mutant at all light intensities examined (Fig.6A). By contrast, only a small but significant reduction in photosynthetic electron transport rates was observed in the trxf1 mutant, whereas the trxf2 mutant displayed wild-type rates (Fig.6A). Notably, the yields of NPQ were higher in trxf1f2 plants, particularly at low light intensities (Fig.6B). These measurements were performed using plants grown under short-day conditions and very similar results were obtained when these photosynthetic parameters were analysed in plants grown under the long-day photoperiod (see Supplementary Figs S5 and S6). The effect of the type f Trxs on the control of energy quenching and photosynthetic yield might be through direct interaction with the photosynthetic apparatus or an indirect consequence of limited biosynthesis and demand for ATP, leading to lower luminal pH. The latter case should be revealed by a lower rate of consumption of electrons from photosynthetic electron transport. Therefore, we measured the PSI quantum yield, Y(I), based on P700 absorbance changes, in mutant and wild-type plants (Fig. 7; see Supplementary Fig.S7). Indeed, the double mutant had a lower PSI activity due to prolonged limitations on the acceptor side, Y(NA), after turning on the light (Fig.7D). This indicates retardation of the activation of the biosynthetic processes which consume reducing equivalents derived from photosynthetic electron transport. The response of the rate of CO2 fixation (AN) to illumination was more retarded in the trxf1 than in the trxf2 single mutant while the trxf1f2 double mutant displayed even slower Table1. Fv/Fm and kinetics of response of net CO2 assimilation rate to light Genotype WT trxf1 trxf2 trxf1f2 Fv/Fm0.81 ± 0.01 0.82 ± 0.01 0.81 ± 0.01 0.82 ± 0.01 t1/2 (s) 106.0 ± 2.3 a 187.9 ± 7.5 b 134.3 ± 9.5 a 237.2 ± 10.1 c The maximum PSII quantum yield was determined as variable fluorescence (Fv) to maximal fluorescence (Fm), Fv/Fm, in dark-adapted leaves of plants grown under short-day conditions. The Fv/Fm values (±SD) are the average of 12 measurements. t1/2 represents the time to achieve 50% of the final rate of CO2 assimilation, as shown in the data of Supplementary Fig. S5. Data presented are the means (±SE; n=6). The differences between mutants and the wild type, when significant, are indicated by different letters (P <0.01; Anova, Tukey test). Fig.4. Effect of light intensity on growth of Arabidopsis lines lacking type-f Trxs under a short-day photoperiod. (A) Representative plants of the wild type and mutant lines grown under short-day conditions (8/16 h light/dark) for 34 d at increased light intensities, as indicated. (B) The weight of the rosette leaves from all lines was determined from 12 plants. (C) Chlorophyll content was determined from leaf discs (n=6). Average values ±SD are represented. Letters indicate significant differences with the Tukey Test and a confidence interval of 99%. Effects of f-type thioredoxins deficiency in Arabidopsis | 1957 kinetics (Table1; see Supplementary Fig. S8). The half-time for reaching the maximal CO2 fixation rate after turning on the light was more than twice as long in the double mutant compared with wild-type plants (Table1). These results demonstrate the function of f-type Trxs in the rapid response of the chloroplast assimilatory metabolism to light. Light-dependent reduction of FBPase and Rubisco activase is impaired in the trxf1f2 doublemutant There is extensive evidence in vitro supporting the relevant function of f-type Trxs in light-dependent reductive activation of different chloroplast enzymes including those of the Calvin–Benson cycle (recently reviewed by Michelet et al., 2013). To analyse further the function of f-type Trxs in vivo, we examined the change of the redox status of two well-known in vitro targets of type f Trxs, FBPase and Rubisco activase, in response to light in the mutant lines. To this end, samples were taken at the end of the night period from plants that had been grown at a light intensity of 125 μE m−2 s−1 and then subjected to illumination with the same or higher (500μE m−2 s−1) light intensity. Short-term changes of the redox status of these enzymes were analysed with the aid of the thiol-alkylating agent methyl maleimide-polyethylene glycol24 (MM-PEG24), which adds 1.24 kDa per thiol group, thus producing a shift in the electrophoretic mobility of the labelled proteins that reflects their redox status. In dark-adapted plants FBPase was detected as a single band indicating that, as expected, the enzyme was fully oxidized under these conditions (Fig.8). In wild-type plants, FBPase becomes rapidly reduced in response to light within seconds after illumination (Fig.8). The level of reduction of FBPase proved to be dependent on light intensity since the enzyme became fully reduced after 10 min at 500μE m−2 s−1, whereas the growth light intensity of 125μE m−2 s−1 only promoted a partial reduction, approximately 50%, of the enzyme (Fig.8). While the deficiency of Trx f2 exerted almost no effect on FBPase reduction in response to light, the deficiency of Trx f1 significantly impaired the level of reduction of the enzyme at both light intensities (Fig.8). The trxf1f2 double mutant showed not only a pronounced delay of FBPase reduction in response to light but also a lower level of reduction of the enzyme after 10 min of illumination, even at the higher light intensity tested here (Fig.8). Asimilar pattern of reduction in response to light was observed for another prominent redox-regulated enzyme of the Calvin–Benson cycle, Rubisco activase (Fig.9). In dark-adapted samples Rubisco activase was detected as a double band corresponding to isoforms of 46 and 43 kDa, respectively, due to alternative splicing. The long isoform has a C-terminal extension that includes two cysteines which may form an intramolecular disulphide that renders the enzyme inactive (Zhang and Portis, 1999). Quantification of the reduction of FBPase and Rubisco activase from a series of experiments (such as those presented in Figs 8 and 9), revealed that the levels of reduction of both enzymes remained at 50% of the wild-type level in the trxf1f2 double mutant after 10 min illumination at the growth light intensity (see Supplementary Fig. S9A, B). Furthermore, we analysed the change of the redox status of FBPase following a light–dark transition. Re-oxidation of FBPase in the dark was faster in all three Trx f deficient mutants than in wildtype plants (Fig.10). However, after 5 min darkness, FBPase was also nearly completely oxidized in the wild type (Fig.10). Finally, we compared the starch content in leaves which was significantly lower in the trxf1f2 double mutant than in the wild type and the single mutants (Fig.11). Discussion Following the demonstration of the function of Trxs in the light-dependent regulation of chloroplast photosynthetic B A WT trxf1 trxf2 trxf1f2 0 200 400 600 800 1000 0.0 0.2 0.4 0.6 0.8 1.0 Y(NPQ) Time (s) 0 200 400 600 800 1000 0.0 0.2 0.4 0.6 0.8 1.0 Y(II) Time (s) ***** ** *** *** ** ****** *** *** *** ** *** *** *** *** ** ** * *** *** *** *** ** Fig.5. Photosystem II activity in wild type and Trx f-deficient mutants. Chlorophyll fluorescence of photosystem II (P680) was measured using a pulse-amplitude modulation fluorimeter. After incubation of the plants for 30 min in darkness, an induction–recovery curve was performed to determine the quantum yields of non-photochemical quenching, Y(NPQ) (A) and photosystem II, Y(II) (B). During the 8 min induction period, a red (653 nm) actinic light at 75μE m−2 s−1 intensity was applied. Thereafter, the actinic light was switched off and measurements were continued for another 10 min in the dark. Saturation pulses (10 000μE m−2 s−1, 0.6 s) every 60 s were applied. Values are the mean ±SD of 6 plants grown for 53 d under short-day conditions. Light and dark periods are indicated by the white and black bars. Statistical significance (*P <0.05; **P <0.01; ***P <0.001) was determined with Student’s t test comparing values for each of the mutant lines with the wild type. 1958 | Belén Naranjo etal. metabolism (Wolosiuk and Buchanan, 1976), extensive biochemical and proteomic work has shown the central role of f-type Trxs in the reductive activation of most of the Calvin–Benson cycle enzymes. In this regard, the findings that overexpression of Trx f, but not Trx m, enhanced starch accumulation and increased the content of sugars in tobacco leaves (Sanz-Barrio etal., 2013) lend further support to the idea that, among the complex set of plastidial Trxs, those of type f play a key role in the redox regulation of carbon metabolism. However, Arabidopsis knockout mutants for Trx f1, which accounts for up to 95% of the total f-type Trxs in this plant, showed impairment of AGPase redox regulation and leaf diurnal starch turnover alterations, yet the visible phenotype of these mutants was indistinguishable from the wild type (Thormählen etal., 2013). Moreover, double mutants lacking both Trx f1 and f2 were also indistinguishable from wild type plants (Yoshida etal., 2015). Thus, the lack of f-type Trxs seems not to affect plant growth despite the relevant function proposed for these Trxs in the redox regulation of chloroplast metabolism based on biochemical analyses. These results suggest that additional redox systems may compensate for the function of f-type Trxs being sufficient to support the redox regulation that allows plant growth. To address this issue, in the present work we have generated a double mutant of Arabidopsis, here termed trxf1f2, lacking both Trx f1 and Trx f2. Both qRT-PCR and Western blot analyses (Fig. 1A, B) confirmed that the trxf1f2 double mutant is a null mutant devoid of f-type Trxs. Western blot analysis of leaf extracts from wild-type plants detected a double band, which might correspond to f1 and f2 Trxs. However, this possibility was ruled out because the double band was detected in the trxf2 mutant but not in the trxf1 mutant. Thus, the double band is most probably indicative of a post-translational modification, the nature of which is still unknown. Under long-day conditions the growth of the single mutants, trxf1 and trxf2, was indistinguishable from that of wild-type plants, consistent with previous studies on Trx f1-deficient mutants (Thormählen etal., 2013). Similarly, the double mutant displayed the wild-type phenotype when grown under long-day conditions, as shown for type-f null mutants recently reported (Yoshida et al., 2015). However, under short-day conditions, single mutants trxf1 and trxf2 showed slightly retarded growth (Fig.3B), in agreement with the behaviour of the Trx f1-deficient mutant, which shows retarded growth under short-days, but not under the longday photoperiod (Thormälen et al., 2015). Adult plants of the trxf1f2 double mutant showed retarded growth (Fig.3) WT trxf1 trxf2 trxf1f2 0 500 1000 1500 2000 0 5 10 15 20 25 30 ETR(II) PAR (µE m-2 s-1) WT trxf1 trxf2 trxf1f2 0 500 1000 1500 2000 0.0 0.2 0.4 0.6 0.8 1.0 Y (NPQ) PAR (µE m-2 s-1) 050 100 150 200 250 300 350 0 5 10 15 20 25 30 ETR(II) PAR (µE m-2 s-1 ) 050 100 150 200 250 300 350 0.0 0.1 0.2 0.3 0.4 0.5 0.6 Y (NPQ) PAR (µE m-2 s-1 ) BA * ** ** * ** ****** *** ****** ***** ** ** ** *** *** *** *** ** * ** ** * Fig.6. Light-dependent linear photosynthetic electron transport and NPQ in the wild type and Trx f-deficient mutants. (A) Relative linear electron transport rates of photosystem II, ETR (II) was measured in pre-illuminated attached leaves of plants grown at 125μE m−2 s−1 under short-day conditions. Chlorophyll fluorescence of photosystem II was determined using a pulse-amplitude modulation fluorimeter. Each data point is the mean ±SD of the ETR (II) from eight plants under short-day conditions. PAR (photosynthetically active radiation). (B) Quantum yields of NPQ from the corresponding lighttitration curves in (A). Statistical significance (*P <0.05; **P <0.01; ***P <0.001) was determined with Student’s t test comparing values for each of the mutant lines with the wild type. Effects of f-type thioredoxins deficiency in Arabidopsis | 1959 which was observed only at later stages of development, indicating that the effect of the deficiency of f-type Trxs reported here (lower efficiency of photosynthetic parameters, impaired reduction of Calvin–Benson enzymes, lower starch content) requires time to affect plant growth. Most likely, the delayed growth of the trx f-deficient mutant under short-day conditions is not attributable to short-term kinetics of activation, but rather to the lower final level of reduction of enzymes, such as FBPase and Rubisco activase, which can be seen after 10 min of illumination (Figs 8, 9)and also at the end of the day (Fig.10). Moreover, high light intensity (950μE m−2 s−1), which caused the decrease in chlorophyll content of both the wild type and mutant lines (Fig. 4C), resulted in retarded growth not only of the double trxf1f2 mutant but also of the single mutants after 34 d of growth (Fig.4B). Therefore, deficiency of f-type Trxs affects the plant’s response to high light suggesting the participation of these Trxs in plant adaptation to light intensities that might produce oxidative stress. Therefore, despite the central function attributed to type f Trxs in chloroplast redox regulation based on biochemical in vitro analyses, the in vivo approach reported here shows that these Trxs are dispensable for plant growth at least under the standard long-day conditions performed in this study. It should be noted, however, that conditions that may cause light limitation, such as growth under a short-day photoperiod, have a negative effect on the growth rate of Trx f-deficient mutants. Moreover, higher light intensity failed to stimulate the growth of mutant plants to the extent that it was stimulated in wildtype plants (Fig.4), indicating the relevant function of type-f Trxs is for plant adaptation to varying light conditions. The genes encoding Trx f1 and Trx f2 isoforms are subject to different regulation in Arabidopsis. The expression of the TRX f1 gene responds to light, whereas the TRX f2 gene is under circadian control (Barajas-López etal., 2011). The differential pattern of expression of these genes might indicate different functions for the respective Trx f isoforms. However, all the photosynthetic parameters analysed in this study, such as photosynthetic electron transport and response to illumination of the rate of CO2 fixation (AN), were more affected in the trxf1 than in the trxf2 mutant, while the double mutant showed an additive effect. Most probably, these results reflect the higher content of Trx f1 in wild-type plants (Fig.1B) and support a redundant function for both f-type Trxs in Arabidopsis. A relevant question concerning chloroplast redox regulation is the level of redundancy or specificity among the 0 100 200 300 400 500 600 0.0 0.2 0.4 0.6 0.8 1.0 trxf2 Time (s) 0 100 200 300 400 500 600 0.0 0.2 0.4 0.6 0.8 1.0 trxf1f2 Time (s) 0 100 200 300 400 500 600 0.0 0.2 0.4 0.6 0.8 1.0 WT Time (s) 0 100 200 300 400 500 600 0.0 0.2 0.4 0.6 0.8 1.0 trxf1 Time (s) A C B D Y (I) Y (ND) Y (NA) Y (I) Y (ND) Y (NA) Y (I) Y (ND) Y (NA) Y (I) Y (ND) Y (NA) Relative quantum yield Relative quantum yield Relative quantum yield Relative quantum yield *** ** ** * *** *** *** ** *** *** *** *** *** *** * ** *** *** ** *** ** ***** *** *** * Fig.7. Comparative analysis of photosystem Iactivity in the wild type and Trx f-deficient mutants. The absorbance of oxidized phtotosystem I(P700) at 830 nm was measured using a pulse-amplitude modulation fluorimeter. The redox state of photosystem I(P700) was inferred from changes in this absorbance and monitored for 10 min. After 30 min dark adaptation (maximum P700 reduced), leaves were illuminated with far-red light superimposed on actinic red light (126μE m−2 s−1, 635 nm) to achieve the maximum oxidation of P700 and, subsequently, the actinic light was kept on for 5 min followed by another 5 min of darkness. White light saturation pulses (10 000μE m−2 s−1, 0.6 s) were applied every 20 s. The quantum yields of PSI Y(I), donor side limitations Y(ND), and acceptor side limitations Y(NA) are based on saturating pulse analysis. Data were collected from six plants grown under short-day conditions and mean ±SD are represented. Statistical significance (*P <0.05; **P <0.01; ***P <0.001) was determined with Student’s t test comparing values for each of the mutant lines (B, C, D) with the wild type (A).