Evaluation of CBSX Proteins as Regulators of the Chloroplast Thioredoxin System

The chloroplast-localized cystathionine β-synthase X (CBSX) proteins CBSX1 and CBSX2 have been proposed as modulators of thioredoxins (Trxs). In this study, the contribution of CBSX proteins to the redox regulation of thiol enzymes in the chloroplast Trx system was evaluated both in vitro and in vivo. The in vitro biochemical studies evaluated whether CBSX proteins alter the specificities of classical chloroplastic Trx f and Trx m for their target proteins. However, addition of CBSX proteins did not alter the specificities of Trx f and Trx m for disulfide bond reduction of the photosynthesis-related major thiol enzymes, FBPase, SBPase, and NADP-MDH. In vivo analysis showed that CBSX-deficient mutants grew similarly to wild type plants under continuous normal light conditions and that CBSX deficiency did not affect photo-reduction of photosynthesis-related thiol enzymes by Trx system at several light intensities. Although CBSX proteins have been suggested as modulators in the chloroplast Trx system, our results did not support this model, at least in the cases of FBPase, SBPase, and NADP-MDH in leaves. However, fresh weights of the cbsx2 mutants were decreased under short day. Since Trxs regulate many proteins participating in various metabolic reactions in the chloroplast, CBSX proteins may function to regulate other chloroplast Trx target proteins, or serve as modulators in non-photosynthetic plastids of flowers. As a next stage, further investigations are required to understand the modulation of Trx-dependent redox regulation by plastidal CBSX proteins.


INTRODUCTION
Proteins belonging to the chloroplastic thioredoxin (Trx) family are reduced via ferredoxin and ferredoxin-thioredoxin reductase, using electrons from the photosynthetic electron transport pathway in light conditions (Schurmann and Buchanan, 2008;Geigenberger and Fernie, 2014;Buchanan, 2016). The reduced forms of Trx proteins regulate the activity of thiol enzymes, such as those in the Calvin-Benson cycle, by reduction of their disulfide bonds (Michelet et al., 2013). This Trx-dependent regulation of thiol enzymes is known as chloroplast redox regulation; the NADPH-dependent Trx system in the chloroplast is also known as the NTRC system. NTRC proteins uniquely contain both thioredoxin reductase and thioredoxin domains (Serrato et al., 2004;Perez-Ruiz et al., 2006). NTRC system involvement in plastidial redox-regulation processes has also been suggested (Nikkanen and Rintamaki, 2014;Buchanan, 2016;Cejudo et al., 2019).
Typical Trx family proteins in the chloroplast stroma of Arabidopsis thaliana are categorized into five types: two f-type isoforms (Trx f1 and f2); four m-type isoforms (Trx m1,m2,m3,and m4); one x-type isoform (Trx x); two y-type isoforms (Trx y1 and y2); and one z-type isoform (Trx z; Michelet et al., 2013;Serrato et al., 2013;Balsera et al., 2014). Trx f and Trx m were originally identified as regulators of thiol enzymes involved in photosynthetic carbon fixation and related processes in the chloroplast (Jacquot et al., 1978;Wolosiuk et al., 1979). Furthermore, using in vitro biochemical approaches, Trx specificities for their target thiol enzymes were well characterized (Schurmann and Buchanan, 2008;Geigenberger and Fernie, 2014). Calvin-Benson cycle enzymes, including fructose-1,6-bisphosphatase (FBPase; Wolosiuk and Buchanan, 1977;Collin et al., 2003) and sedoheptulose-1,7-bisphosphatase (SBPase; Breazeale et al., 1978;Nishizawa and Buchanan, 1981), and the γ subunit of ATP synthase (Schwarz et al., 1997) were mainly regulated by Trx f. NADP malate dehydrogenase (MDH) was characterized as an enzyme regulated by Trx m (Issakidis et al., 1992), although this enzyme was also later shown to be efficiently activated by Trx f (Collin et al., 2003;Yoshida et al., 2015). However, little analysis of the physiological role of the Trx f and Trx m proteins was available for a long time. Recently, several groups characterized Trx f-and Trx m-deficient mutants using Arabidopsis T-DNA insertion lines and RNA interference. Trx m-deficient mutants were impaired in photoreduction of photosynthesis-related enzymes including FBPase, SBPase, and MDH, and showed severe growth defects as a result (Wang et al., 2013;Okegawa and Motohashi, 2015a;Da et al., 2018). In contrast, Trx f-deficient mutants did not exhibit substantial defects (Yoshida et al., 2015;Naranjo et al., 2016;Thormahlen et al., 2017), although the in vitro biochemical analysis clearly showed that the FBPase (Wolosiuk and Buchanan, 1977) and SBPase (Breazeale et al., 1978) activities were regulated by Trx f. These results suggest that in vivo, Trx m is a more significant redox regulator than Trx f in the activation of FBPase, SBPase, and NADP-MDH. Further research is needed to explain the discrepancy between the in vitro and in vivo studies.
As a regulator of Trx proteins, thioredoxin-interacting protein (TXNIP) was first identified in mammals (Nishiyama et al., 1999). TXNIP interacts with the active site of Trx and inhibits its disulfide reduction activity (Spindel et al., 2012). In plants, CBSX proteins were identified as candidate Trx regulators (Yoo et al., 2011;Jung et al., 2013). CBSX proteins can enhance Trx disulfide reduction activity, in contrast to TXNIP (Yoo et al., 2011). In cyanobacteria, fusion proteins containing a CBS domain and CP12 component were found (Stanley et al., 2013). Plant CP12 protein forms complexes with glyceraldehyde-3-phosphate dehydrogenase (GAPDH) and phosphoribulokinase (PRK), which are also well known redox-regulated Calvin-Benson cycle enzymes. It thus functions as a regulator in the Calvin-Benson cycle (Michelet et al., 2013). The discovery of CBS-CP12 fusion proteins in cyanobacteria implies that the CBS-domain may act as a regulator in the redox regulation system (Lopez-Calcagno et al., 2014). Arabidopsis CBSX1 and CBSX2 were indeed proposed as Trx regulators in chloroplasts (Yoo et al., 2011;Jung et al., 2013).
In this study, we evaluated the possibility that CBSX1 and CBSX2 regulate Trx f and Trx m in the chloroplast redox regulation system using in vitro and in vivo experiments. Specifically, we investigated whether CBSX1 and CBSX2 modulates the target specificity of Trx f and Trx m in redox regulation of the photosynthesis-related thiol enzymes FBPase, SBPase, and NADP-MDH. However, we could not find any evidence that support this possibility. CBSX1 and CBSX2 may function not as a Trx regulator for Calvin-Benson cycle enzyme activation, but rather as a regulator for other thiol enzymes in flowers and leaves.

Construction of Expression System for
Arabidopsis CBSX1, CBSX2, FBPase, SBPase, and NADP-MDH The mature forms of Arabidopsis chloroplast proteins were predicted using TargetP 1.1 (Emanuelsson et al., 2007). The mature forms of the CBSX1 (AT4G36910), CBSX2 (AT4G34120), FBPase (AT3G54050), SBPase (AT3G55800), and NADP-MDH (AT5G58330) genes were obtained by PCR amplification from an Arabidopsis cDNA library (Yamazaki et al., 2004). Primers for cloning and vector constructs are described in Supplementary Table S1. The amplified CBSX1 and CBSX2 genes were directly inserted by seamless cloning into the NdeI and BamHI sites of pET23a and the NcoI and EcoRI sites of pET23d, respectively (vectors from Merck Millipore, Darmstadt, Germany), using seamless ligation cloning extract (SLiCE) from a laboratory Escherichia coli JM109 strain (Okegawa and Motohashi, 2015b;Motohashi, 2017b). The amplified FBPase and SBPase genes were cloned into the NdeI and EcoRI sites of pET23a, and the amplified NADP-MDH gene into the NcoI and BamHI sites of pET23d, by the conventional restriction enzyme method. The length of cloned DNA inserts was confirmed by colony PCR Motohashi, 2017a), using T7 promoter and T7 terminator primers (Okegawa and Motohashi, 2015c). Plasmids were prepared from colony PCR-positive clones and the inserted DNA sequences of plasmids were confirmed by DNA sequencing (Sanger et al., 1977). Arabidopsis NADP-MDH was purified without an affinity tag in the absence of dithiothreitol (DTT) as described previously .

Preparation of Antibodies
Polyclonal antibodies were prepared from rabbits immunized with CBSX1, CBSX2, FBPase, and SBPase (Okegawa and Motohashi, 2015a), and NADP-MDH  recombinant proteins. Purified Trx f1 and Trx m1 were also used as antigens for polyclonal antibodies production. Polyclonal antibodies for PrxA (AT3G11630) and PrxB (AT5G06290) were prepared as peptide antibodies against C-terminal consensus amino acid sequence between PrxA and PrxB (Cys-SMKPDPKLSKEYFSAI). Polyclonal antibodies for PetA (ATCG00540) and PetC (AT4G03280) were prepared from rabbits immunized with partially purified inclusion bodies from recombinant expression proteins (Supplementary Table S1).

Plant Materials and Growth Conditions
Arabidopsis thaliana ecotype Columbia-0 was used as the wild type. The T-DNA insertion line GABI_050D12 (cbsx1) was obtained from the Nottingham Arabidopsis stock center (NASC, United Kingdom). The T-DNA insertion line SALK_136934C (cbsx2; Jung et al., 2013) was obtained from the Salk Institute Genomic Analysis Laboratory (SIGnAL, CA, United States). Plants were grown in soil, or in petri dishes containing Murashige Skoog (MS) medium (Murashige and Skoog, 1962) with 1.2% (w/v) agar and 1% (w/v) sucrose in the growth-chambers (50 μmol photons m −2 s −1 , continuous light, 23°C). Agar-sown seeds were first surface sterilized by soaking in 5% (v/v) sodium hypochlorite for 5 min. To generate double mutants, cbsx1 and cbsx2 T-DNA single mutants were crossed. The cbsx1 and cbsx2 double mutant was selected from F 2 plants. The presence of T-DNA insertions was confirmed by PCR (for primer, see Supplementary Table S2) and DNA agarose gel electrophoresis (Motohashi, 2019).

Analysis of mRNA Expression in cbsx Mutants by RT-PCR
The mRNA from wild type and cbsx mutants was prepared using Sepasol ® -RNA I Super G (Nakalai tesque, Kyoto, Japan), and cDNA samples were synthesized for 20 min at 42°C by ReverTra Ace ® (Toyobo, Osaka, Japan), using oligo(dT) 20 .
The cDNAs from wild type and cbsx mutants were amplified by KAPATaq EXtra (KAPA Biosystems, Wilmington, MA, United States), with the following cycle conditions: 94°C for 2 min, (94°C for 30 s, 55°C for 30 s, and 72°C for 1 min) × 25 cycles. The primers used are listed in Supplementary Table S2. Amplified samples were electrophoresed and visualized by a combination of a black light and a longpass emission-filter (SC-46, Fujifilm, Tokyo, Japan; Motohashi, 2019).

Expression Analysis of CBSX1, CBSX2, and Redox Regulation-Related Proteins by Western Blotting
To analyze protein expression in Arabidopsis, leaves and flowers of wild type and cbsx mutants were frozen and ground in liquid nitrogen; total proteins were extracted in 1 × SDS sample buffer [2% (w/v) SDS, 5% (v/v) 2-mercaptoethanol, 10% (v/v) glycerol, and 62.5 mM Tris-HCl (pH 6.8)]. Extracted proteins (30 or 90 μg) were loaded on SDS-PAGE [15% (w/v) acrylamide] and detected by specific antibodies. Immunoblot signal was detected with a LAS-3000UVmini lumino-image analyzer (Fujifilm, Tokyo, Japan). Uncropped original images were indicated in Supplementary Figure S1.

Chlorophyll Fluorescence Analysis
Chlorophyll fluorescence was measured with a Mini-pulseamplitude modulation (PAM) portable chlorophyll fluorometer (Walz, Germany). Minimum fluorescence (Fo) was obtained with open PSII centers in the dark-adapted state by a low-intensity measuring light (wavelength 650 nm, 0.05-0.1 μmol photons m −2 s −1 ). A saturating pulse of white light (800 ms, 3,000 μmol photons m −2 s −1 ) was applied to determine the maximum fluorescence with closed PSII centers in the dark-adapted state (Fm) and during actinic light (AL) illumination (Fm'). The steady-state fluorescence level (Fs) was recorded during AL illumination (25-828 μmol photons m −2 s −1 ). The quantum yield of PSII (Φ PSII ) was calculated as (Fm'−Fs)/Fm' (Genty et al., 1989). The relative rate of electron transport through PSII (ETR) was calculated as Φ PSII × light intensity (μmol photons m −2 s −1 ). Non-photochemical quenching (NPQ) was calculated as (Fm−Fm')/Fm'. For the analysis of lightintensity dependence of fluorescence parameters, AL intensity was increased in a stepwise manner every 2 min after the application of a saturating pulse.

Photo-Reduction Assay of Trx-Target Proteins in vivo
Photo-reduction of Trx target enzymes in seedlings was determined using the free thiol-specific modifying reagent, 4-acetamido-4'-maleimidylstilbene-2,2'-disulfonic acid (AMS; Thermo Fisher Scientific, Carlsbad, CA, United States), as described by (Okegawa and Motohashi, 2015a), with minor modifications. Seedlings were incubated in the dark for 3 h and then illuminated at three different light intensities for 1 h (50, 200, or 800 μmol photons m −2 s −1 ). Seedlings were collected thereafter and frozen in liquid nitrogen. Frozen samples were ground using 5 mm tungsten beads and incubated in 10 mM AMS, 10 mM EDTA, 4% (w/v) SDS, 8 M urea, and 125 mM Tris-HCl (pH 6.8) for 2 h at 25°C to complete the labeling of thiol groups with AMS. Hydrophilic proteins were separated from the lipid fraction by the addition of an equal volume of chloroform to the sample. The aqueous portion was collected by centrifugation at 15,000 g for 10 min). Reduced and oxidized proteins were separated by non-reducing SDS-PAGE [12.5% (w/v) acrylamide] and detected by western blotting. Immunoblot signal was detected with a LAS-3000UVmini lumino-image analyzer (Fujifilm, Tokyo, Japan). The reduction level was quantified using Multi Gauge 3.1 software (Fujifilm, Tokyo, Japan) and the ratio of the reduced form to total protein was determined. Uncropped original images were indicated in Supplementary Figure S1.

Contribution of CBSX Proteins to Reduction of Photosynthesis-Related Thiol Enzymes, in vitro
Previous work showed that CBSX1 and CBSX2 proteins interact with Trxs (Yoo et al., 2011;Jung et al., 2013), and that CBSX1 protein directly binds with both Trx f and Trx m proteins in vitro (Yoo et al., 2011). Photosynthesis-related thiol enzymes, including several Calvin-Benson cycle enzymes, are regulated by Trx proteins in the light. Published in vitro studies indicate preferential reduction by Trx f of FBPase (Wolosiuk and Buchanan, 1977;Collin et al., 2003) and SBPase (Breazeale et al., 1978;Nishizawa and Buchanan, 1981), whereas NADP-MDH is reduced by both Trx f and Trx m proteins (Issakidis et al., 1992;Collin et al., 2003). In contrast, in vivo, levels of photo-reduction in thiol enzymes such as FBPase, SBPase, and NADP-MDH were significantly decreased in the Trx m-deficient mutants (Okegawa and Motohashi, 2015a). If chloroplast CBSX1 and CBSX2 proteins function as modulators of the chloroplast Trx system, the discrepancy between in vitro and in vivo results may be explained. To evaluate this possibility, we firstly evaluated the contribution of CBSX proteins to the reduction of thiol enzymes by Trx f and Trx m proteins, using purified recombinant proteins in vitro. Our results were consistent with previous reports described above. FBPase was efficiently reduced in vitro by Trx f1, not Trx m1 (Figure 1A, Trx-only elements), and SBPase was reduced by both Trx f1 and Trx m1, but Trx f1 could reduce more efficiently than Trx m1 (Figure 1B, Trx only elements). On the other hand, NADP-MDH was reduced to the same extent by both Trx f1 and Trx m1 (Figure 1C, Trx only elements). Given this, the effects of addition of CBSX1 or CBSX2 on Trx-dependent reduction of the thiol enzymes by Trx f1 and m1 were evaluated, in vitro. Since CBSX proteins are known to function with AMP as a ligand (Ereno-Orbea et al., 2013), 3 mM of AMP was added to each reaction medium. Under these reaction conditions, both Trx f1 and Trx m1 were reduced in a DTT concentration-dependent manner (Supplementary Figure S2). Reduction of FBPase and SBPase by Trx f1 and Trx m1 was not affected by the addition of CBSX1 (Figure 1, CBSX1 elements) or CBSX2 (Figure 1, CBSX2 elements) with AMP. Trx f1 reduced FBPase and SBPase disulfide bonds as efficiently in the presence of CBSXs and AMP as with Trx f1 alone (Figures 1A,B). The addition of CBSX1 or CBSX2 did not also affect reduction of NADP-MDH by Trx f1 and Trx m1 with AMP; NADP-MDH was equally reduced by Trx f1 and Trx m1 in the presence of CBSXs and AMP ( Figure 1C). In addition to Trx m1, Arabidopsis has abundant Trx m4 protein in its chloroplasts (Okegawa and Motohashi, 2015a). As with Trx m1, reduction patterns of the three thiol enzymes by Trx m4 in the presence of CBSX1 or CBSX2 were not affected [ Figure 1 (Trx m1) and Supplementary Figure S3]. These in vitro experiments did not indicate any contribution of CBSX proteins to Trx-dependent reduction of three photosynthesisrelated thiol enzymes under studied conditions.

Arabidopsis cbsx1 and cbsx2 Mutants Did Not Show Obvious Visible Defects
Arabidopsis has two chloroplast-localized CBSX proteins, namely CBSX1 and CBSX2. The deduced amino acid sequences of these proteins were highly conserved, showing 91% identity. As T-DNA insertion mutants of CBSXs, the cbsx1 (GABI_050D12), and cbsx2 (SALK_136934C) mutants were used to investigate the proteins' physiological significance (Figure 2A). The cbsx1 and cbsx2 mutants did not express the mRNAs of CBSX1 and CBSX2, respectively ( Figure 2B). To obtain the double mutant cbsx1 cbsx2, the cbsx1 and cbsx2 single mutants were crossed. The cbsx mutants did not show obvious phenotypic defects compared with the wild type under normal growth conditions in soil for 5 weeks (Figure 2C). We could not observe morphological abnormalities ( Figure 2C) and differences of chlorophyll content of seedlings (Figures 2D,E) between wild type and the cbsx mutants under two different growth conditions. However, the fresh weight of the cbsx mutants was gradually decreased in the order of wild type, single mutants, and double mutant in soil under short day condition ( Figure 2F). The fresh weight of the cbsx2 and cbsx double mutant grown under short day condition were decreased to ~74% and ~65% of that of wild type (Figure 2F), although there were no obvious differences in the fresh weights between mutants and wild type shoots when these were grown on MS medium agar plates for 3-weekold ( Figure 2G).

Arabidopsis CBSX1 and CBSX2 Proteins Are Present in Leaves and Flowers
According to the Arabidopsis eFP browser 2.0 database for gene expression in Arabidopsis (Winter et al., 2007), the mRNA of CBSX genes is expressed in leaves, flowers, and seeds. To detect proteins in Arabidopsis tissues, we generated specific antibodies for CBSX1 and CBSX2 proteins. The mature forms of the CBSX1 and CBSX2 proteins were predicted to be ~18 kDa proteins, using ChloroP 1.1 (Emanuelsson et al., 1999), and were detected as ~20 kDa proteins in wild type Arabidopsis, using antibodies obtained as described above.
The specificities of the antibodies for CBSX1 and CBSX2 were confirmed by loss of signal in each deficient mutant (Figure 3). The CBSX1 and CBSX2 proteins were present in leaves (Figure 3A, WT) and flowers (Figure 3B, WT), as suggested by the Arabidopsis eFP browser. Western blots detected two bands for CBSX2 protein. We evaluated the two bands as resulting from distinct processing of chloroplast transit sequence elements, not as distinct redox forms, because the CBSX1 and CBSX2 proteins had no cysteine residue in the mature forms. In the cbsx deficient mutants, chloroplast Trxs (Trx f1 and m1), Trx-regulated thiol enzymes (FBPase, SBPase, and NADP-MDH), and PrxA and PrxB that were supplied reducing equivalents by Trxs, were normally detected in leaves (Figure 3A; Supplementary Figure S4). These results indicated that the disappearance of CBSX proteins did not affect the abundance of Trx-related proteins in Arabidopsis photosynthetic leaves.
The cbsx Single and Double Mutants Showed Normal Electron Transport Activities in Leaves activity (Yoo et al., 2011;Lopez-Calcagno et al., 2014). Several Calvin cycle enzymes are Trx-regulated proteins. If CBSX proteins are functional regulators for Trxs, photosynthetic activity in leaves may be affected in cbsx mutants. To characterize photosynthetic electron transport activity in the mutants, their chlorophyll fluorescence parameters were analyzed (Figure 4). The maximum quantum yield of PSII (Fv/Fm) reflects the intactness of PSII. The three cbsx mutants exhibited the same level of Fv/Fm as the wild type ( Figure 4A). The light-intensity dependence of the electron transport rate (ETR) is an indicator of the relative flow rate of electrons through PSII during steadystate photosynthesis, and the NPQ of chlorophyll fluorescence mainly reflects the level of thermal dissipation of excess light energy. No differences in ETR and NPQ were observed between the wild type and the cbsx mutants (Figures 4B,C). These results showed that a deficiency of CBSX did not affect these photosynthetic pathways.

Normal Photo-Reduction of Photosynthesis-Related Thiol Enzymes in cbsx Mutants
We next investigated the contributions of CBSX1 and CBSX2 to Trx-dependent redox regulation in chloroplasts, using the cbsx1, cbsx2, and cbsx1 cbsx2 mutants. Photosynthesis-related thiol enzymes, including several Calvin-Benson cycle enzymes, are reduced by Trx, using electrons from photosynthetic electron transport, and the enzymes are thus activated in light conditions (Hisabori et al., 2007;Michelet et al., 2013;Geigenberger and Fernie, 2014;Buchanan, 2016). If CBSX proteins modulate the activity of photosynthesis-related thiol enzymes via a Trx-dependent system, the photo-reduction levels of thiol enzymes should be altered in the cbsx mutants. FBPase, SBPase, and NADP-MDH disulfide bonds were reduced by Trx in a light intensity-dependent manner in the wild type (Figure 5, WT). The photo-reduction of FBPase, SBPase, and NADP-MDH in the cbsx mutants was compared to that in the wild type (Figure 5, upper elements). At all light intensities studied, the photo-reduction levels of FBPase, SBPase, and NADP-MDH did not differ between the wild type and cbsx mutants (Figure 5, lower elements). Additionally, Trxs (Trx f and Trx m) and Trx-related proteins [FBPase, SBPase, NADP-MDH, 2-Cys peroxiredoxin A (PrxA), and PrxB] were normally present in leaves of the three cbsx mutants (Figure 3A). This in vivo analysis thus showed no observable contribution of CBSXs to redox regulation of FBPase, SBPase, and NADP-MDH under the studied conditions.

DISCUSSION
Arabidopsis CBSX proteins are proposed as modulators of Trx activity in previous works (Yoo et al., 2011;Jung et al., 2013). Although direct interactions between Trxs and CBSXs have been reported (Yoo et al., 2011), their regulatory mechanism is still unknown. CP12 protein forms the PRK/GAPDH/CP12 regulatory complex in Trx-dependent redox regulation. In cyanobacteria, a fused CP12-CBSX gene was also found (Stanley et al., 2013). These fusion genes imply that CBSX proteins may function as a modulator that cooperates with CP12 in Trx-dependent redox regulation. Lopez-Calcagno et al. (2014) suggested that CBSX proteins are modulators in the redox regulation of Trx-dependent thiol-enzymes in Arabidopsis and cyanobacteria (Lopez-Calcagno et al., 2014). However, few biochemical and physiological analyses of CBSX proteins in Trx-dependent redox regulation have been performed. In particular, their contribution to photosynthesisrelated Calvin cycle thiol enzyme reduction in leaves has not been tested. In this study, we evaluated the contributions of CBSX1 and CBSX2 proteins to the redox regulation of photosynthesis-related thiol enzymes by chloroplast Trx f and Trx m, in vitro and in vivo. In vitro, we evaluated whether CBSX proteins can alter the target specificity of Trx f and Trx m, using photosynthesis-related thiol enzymes. We could not demonstrate that CBSX1 and CBSX2 proteins modulated the Trx f-or Trx m-dependent reduction of thiol enzymes such as FBPase, SBPase, and NADP-MDH. CBSX proteins may contribute to the activation of thiol enzymes other than these, because the chloroplast Trx system does function in various redox regulation pathways (Motohashi et al., 2001;Buchanan and Balmer, 2005). Alternatively, other ligands for CBS-domain proteins may be required. CBS-domain proteins are known in wide variety of organisms, and require AMP as a ligand for activation Samples were collected at each of these light intensities and modified with 4-acetamido-4'-maleimidylstilbene-2,2'-disulfonic acid (AMS). Ninety micrograms of protein samples were loaded on non-reducing SDS-PAGE. Reduced (red) and oxidized (ox) forms were detected by western blotting (upper element of each panel), and reduction levels of thiol enzymes were determined as described in Materials and Methods (lower element of each panel). Each value is the mean ± SD (independent-biological replicates, FBPase: n = 9; SBPase: n = 8; and NADP-MDH: n = 7).
Frontiers in Plant Science | www.frontiersin.org 10 February 2021 | Volume 12 | Article 530376 (Baykov et al., 2011). However, several other ligands, including ADP, ATP, and NADH, have been also suggested for CBSXs (Ereno-Orbea et al., 2013). In this study, we used AMP, as the best-characterized ligand. Under these conditions, CBSX proteins had no observable effect on Trx f or Trx m-dependent activation of photosynthesis-related proteins, but other compounds may be bound as ligands to Arabidopsis CBSX1 and CBSX2 proteins in chloroplast redox regulation. Arabidopsis has also Trx m2 protein, in addition to Trx m1 and m4, as a major m-type Trx. Trx m2 is closely related to Trx m1 (81% identity in their deduced amino acid sequences, Supplementary Figure S5) and has the same redox potential value (−335 mV) and similar enzyme kinetic parameters (Yoshida and Hisabori, 2017), although Trx m4 has lower similarity to Trx m1 (60% identity of their deduced amino acid sequences, Supplementary Figure S5) and a different redox potential value (−312 mV). From these facts, we concluded that Trx m2 has similar properties to Trx m1 proteins and is unlikely to be changed due to specificity by CBSX proteins.
In vivo, the cbsx1, cbsx2, and cbsx1 cbsx2 mutants did not exhibit obvious morphological defects compared to the wild type ( Figure 2C). Consistent with their phenotype, the photosynthetic parameters of the three cbsx mutants did not differ from those of the wild type (Figure 4). Deficiencies of the CBSX proteins did not affect photosynthetic activity. Moreover, we analyzed photoreduction of three thiol enzymes, FBPase, SBPase, and NADP-MDH, in three cbsx mutants. The photo-reduction rates of these enzymes in the cbsx mutants were similar to those of the wild type ( Figure 5). For these photosynthesis-related enzymes, the CBSX proteins may not be required for regulation of their activities by Trx f or Trx m under the light conditions used in this study. However, fresh weight of the cbsx single and double mutants was gradually decreased in soil under short day condition ( Figure 2F). Particularly, fresh weight of the cbsx double mutant was decreased to ~65% of that of WT during the condition (Figure 2F), although obvious difference in growth of the cbsx mutants could not be observed on MS plates under continuous light (Figures 2E,G). Interestingly, regarding other redox-related gene deficient mutants, Lepistö et al. reported that the ntrc mutants displayed no phenotype when grown under continuous light, but exhibited a strong growth suppression under short day condition (Lepisto et al., 2009). The ntrc mutant grown under short day condition had smaller mesophyll cells and contained fewer chloroplasts than wild type leaves. Similar mechanism may also function in the cbsx mutants.
From the in vitro and in vivo analyses, we concluded that CBSX1 and CBSX2 proteins did not function as modulators of Trx f and Trx m in redox regulation of the three major thiol enzymes FBPase, SBPase, and NADP-MDH in leaves, at least under the studied conditions. In the last 20 years, a large number of Trx target proteins have been newly identified (Geigenberger et al., 2017). More recently, new target proteins have also been identified (Okegawa and Motohashi, 2020;Yoshida et al., 2020). Although Arabidopsis CBSX proteins are suggested to modulate Trx-activity (Yoo et al., 2011;Jung et al., 2013), it is unclear which Trx-dependent pathways do CBSX proteins regulate. There are several possibilities other than the possibility of Trx-related pathways, including Calvin cycle enzymes in the leaves. CBSX proteins may modulate other unknown redox-related pathways, particularly concern to the growth delay of the cbsx mutants grown in soil under short day condition. Expression of CBSX mRNA was seen in leaves, flowers, and seeds in the Arabidopsis eFP Browser 2.0 (Winter et al., 2007) and we detected accumulation of CBSX proteins in leaves and flowers by western blotting, using anti-CBSX1 and anti-CBSX2. Given this, CBSX proteins may function as modulators for thiol enzymes other than those involved in photosynthesis in leaves and flowers. In particular, redox regulation by Trx proteins in flowers remains largely uninvestigated. To elucidate the functions of CBSX proteins, the effects of Trx proteins including Trx f, Trx m, and other chloroplast Trxs on thiol enzymes in leaves and flowers should form the next stage of investigation.

DATA AVAILABILITY STATEMENT
The datasets generated for this study are available on request to the corresponding author.

ACKNOWLEDGMENTS
We thank Akira Kawabe (Kyoto Sangyo University) for the discussion on statistical analysis.