Characterization of TrxC, an Atypical Thioredoxin Exclusively Present in Cyanobacteria

antioxidants Article Characterization of TrxC, an Atypical Thioredoxin Exclusively Present in Cyanobacteria Luis López-Maury * , Luis G. Heredia-Martínez and Francisco J. Florencio Instituto de Bioquímica Vegetal y Fotosíntesis, Universidad de Sevilla-CSIC, Av/Américo Vespucio 49, E-41092 Sevilla, Spain; [email protected] (L.G.H.-M.); flor[email protected] (F.J.F.) *Correspondence: [email protected]; Tel.: +34-954489500 (ext. 909102) Received: 11 October 2018; Accepted: 11 November 2018; Published: 13 November 2018   Abstract: Cyanobacteria form a diverse group of oxygenic photosynthetic prokaryotes considered to be the antecessor of plant chloroplast. They contain four different thioredoxins isoforms, three of them corresponding to m,xand ytype present in plant chloroplast, while the fourth one (named TrxC) is exclusively found in cyanobacteria. TrxC has a modified active site (WCGLC) instead of the canonical (WCGPC) present in most thioredoxins. We have purified it and assayed its activity but surprisingly TrxC lacked all the classical activities, such as insulin precipitation or activation of the fructose-1,6-bisphosphatase. Mutants lacking trxC or over-expressing it were generated in the model cyanobacterium Synechocystis sp. PCC 6803 and their phenotypes have been analyzed. The ∆ trxC mutant grew at similar rates to WT in all conditions tested although it showed an increased carotenoid content especially under low carbon conditions. Overexpression strains showed reduced growth under the same conditions and accumulated lower amounts of carotenoids. They also showed lower oxygen evolution rates at high light but higher Fv’/Fm’ and Non-photochemical-quenching (NPQ) in dark adapted cells, suggesting a more oxidized plastoquinone pool. All these data suggest that TrxC might have a role in regulating photosynthetic adaptation to low carbon and/or high light conditions. Keywords: cyanobacteria; thioredoxin; photosynthesis 1. Introduction Thioredoxin are small (~12 kDa) evolutionary conserved proteins with redox activity that present a conserved three-dimensional structure denominated “Trx fold”, composed of five β strands surrounded by four α helices [ 1 ]. They serve as redox carriers and catalyse reduction of other proteins activating/deactivating them. Thioredoxins contain a conserved disulphide active site in the form WCGPC that undergoes oxidation-reduction cycles. Most organisms contain at least one thioredoxin gene although their numbers are expanded in photosynthetic organisms [ 2 , 3 ]. Arabidopsis contains at least twenty thioredoxin isoforms, ten are present in Chlamydomonas reinhardtii in contrast to humans or Escherichia coli that only contain two. Arabidopsis thioredoxins are classified in seven groups (Trx h, Trx o, Trx f, Trx z, Trx m, Trx xand Trx y) and all of them are also present in Chlamydomonas. Of these, Trx m, Trx xand Trx yhave a cyanobacterial origin, and together with Trx fand Trx zare located to the chloroplast [ 4 , 5 ]. In cyanobacteria, four thioredoxins classes have been found along the genome of more than 300 species, three of them corresponding to m,xand ytypes present in plant chloroplast, while the fourth one is exclusively found in cyanobacteria (TrxC) [6]. Thioredoxins are reduced by thioredoxin reductases of which at least two different families exist: NADPH-dependent thioredoxin reductases (NTR) and ferredoxin-dependent thioredoxin reductases (FTR). NTR are present from bacteria to humans while FTR is an Fe-S protein that uses ferredoxin and is present in photosynthetic organisms [ 2 ]. Thioredoxin reduction system is also diverse in cyanobacteria, most of them containing the FTR system (except Gloeobacter and Prochlorococcus Antioxidants 2018,7, 164; doi:10.3390/antiox7110164 www.mdpi.com/journal/antioxidants Antioxidants 2018,7, 164 2 of 15 groups) [ 2 , 6 ]. Some cyanobacteria also contain NTRC, a protein that contains a NTR module fused to a thioredoxin module in a single polipeptide, which seems to function as a reducing system for the 2-cys peroxiredoxin in Anabaena sp. PCC 7120 (hereafter Anabaena) [ 7 – 9 ]. They can also contain two other NTR related proteins that were initially annotated as NADPH dependent enzymes. However, these two proteins have recently been shown not to use or bind NADPH, as they lack the aminoacid signature characteristic of the NADPH binding domain, and have been renamed DTR (for D eeply rooted bacterial T hioredoxin R eductase) and DDOR ( D iflavin-linked D isulfide O xido R eductase) [ 10 , 11 ]. Dithionite reduced DTR is able to reduce thioredoxins although its physiological reductant is still unknown [ 10 ]. In contrast, DDOR shows a new structure containing two to Flavin Adenine Dinucleotide (FAD) per monomer, does not reduce thioredoxins and probably functions as an oxidase [ 11 ]. These proteins are scattered distributed in cyanobacteria but strains lacking FTR usually contain a gene coding for DTR. Of these two proteins, only DDOR is present in Synechocystis sp. PCC 6803 (hereafter Synechocystis) although some cyanobacteria (such as Gloeobacter violaceus sp. PCC 7421) contains both [2,10,11]. Putative thioredoxins targets have been studied in cyanobacteria by several proteomics approaches [ 12 – 17 ]. A high degree of overlapping targets for the different isoforms have been found, highlighting that these in vitro proteomics approaches identify reactive cysteine that could be subjected to redox regulation in vivo and that overlapping and redundant roles are expected for the different thioredoxins in vivo . The role of the different thioredoxins has been also analyzed by generating mutants in their corresponding genes. Trx m(trxA) has been shown to be essential in unicellular cyanobacteria and therefore no specific phenotypes have been associated to it [ 18 , 19 ]. More recently, mutants in trxM genes have been described in Anabaena which possess two genes coding for Trx m:trxM1 and trxM2. Both genes are dispensable although trxM1 mutant showed a pleiotropic phenotype and was unable to grow in diazotrophic conditions, while trxM2 lack any appreciable phenotype [ 20 , 21 ]. In plants trxM are involved in several functions related to fluctuating light conditions, cyclic electron flow or meristem maintenance [ 4 , 5 , 22 ]. In Synechocystis trxB ´ mutant strains (lacking xtype) showed sensitivity towards HL and to the presence of DTT in the culture media [ 23 ], while trxQ ´ mutant (lacking ytype) showed sensitivity to oxidative stress induced by methyl viologen [ 23 ]. All three proteins were able to interact and reduce the different peroxiredoxins present in Synechocystis although with different efficiencies, with TrxQ being the most efficient in vitro [ 24 ]. The fourth group, TrxC, is only found in cyanobacteria. No activity or function has been ascribed to this class of Trx in Synechocystis as mutants lacking it did not show any phenotype [ 25 ], although Anabaena trxC ´ mutant showed more oxidative stress than WT in nitrate grown cultures [ 20 ]. Furthermore, Anabaena TrxC seems to be inactive in reducing OpcA or G6PDH [ 21 ]. Here we describe the characterization of Synechocystis’ TrxC protein and of mutants either lacking trxC or overexpressing it. 2. Materials and Methods 2.1. Strains and Culture Conditions Synechocystis cells were grown photoautotrophically on BG11 or BG11C [ 26 ] at 30 ˝ C under continuous illumination (50 to 180 µ Mol photons m ´2 s ´1 ) and bubbled with a stream of 1% (v/v) CO 2 in air as indicated. BG11 media was buffered with 10 mM TES-NaOH pH 7.5. For plate cultures, medium was supplemented with 1% (wt/vol) agar. All media contained the standard copper concentration (0.3 µ M) except in Figure 2 in which BG11-Cu was used. Kanamycin and nourseothricin, were added to a final concentration of 50 µ g mL ´1 .Synechocystis strains and their relevant genotypes are described in Table 1.E. coli DH5 α or BL21 cells were grown in Luria broth medium and supplemented with 100 µ g mL ´1 ampicillin, 50 µ g mL ´1 kanamycin and 50 µ g mL ´1 nourseothricin when required. Antioxidants 2018,7, 164 3 of 15 Table 1. Synechocystis strains used in this work. Strain Relevant Genotype Mutated ORFS Source WT Synechocystis sp PCC 6803 - Lab collection STXC2 ∆trxC::C.K1 sll1057 [25] WTOE glnN::PpetE:histrxC:Nat slr0288 This study STXCOE ∆trxC::C.K1 glnN::PpetE:histrxC:Nat sll1057,slr0288 This study 2.2. Western Blotting Crude extracts were prepared using glass beads and vigorous vortexing using a minibead-beater. Cells (corresponding to 20 OD 750nm ) were resuspended in 300 µ L of buffer A (50 mM Tris-HCl pH 8.0, 50 mM NaCl) and subjected to 2 cycles of 1 min vortexing separated by 5 min of cooling on ice. Cell extracts were recovered from the beads by piercing a hole in bottom of the tube and samples were clarified by two sequential centrifugations: 5’ at 3000 ˆ gto eliminate cells debris and twice 15 min at 18,000 ˆ gto collect membranes. Protein concentration in cell-free extracts by the method of Lowry, using Bovine Serum Albumin as a standard and the specified amounts of proteins were separated on sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE). Gels were transferred to nitrocellulose membranes (Bio-Rad, Düsseldorf, Germany catalog #162-0115), blocked in phosphate-buffered saline (PBS) containing 0.1% Tween 20 and 5% of skimmed milk and incubated with antibodies against TrxA (1:3000), TrxB (1:3000), TrxQ (1:2000), TrxC (1:1000), DDOR (1:5000), GrxA (1:3000), GrxC (1:3000) and 2cysprx (sll0755; 1:5000). The ECL Prime immunoblotting system (GE Healthcare, Little Chalfont) was used to detect the different antigens with goat anti-rabbit conjugated to horseradish peroxidase (Sigma St Louis, USA catalog #A6154) diluted to 1:25,000. 2.3. Cloning and Purification of TrxC TrxC gene was cloned from Synechocystis’ DNA after PCR amplification with oligonucleotides Syn_TrxC_BamHI_NdeI and SynTrxC_R_SalI and cloned into NdeI-SalI digested pET28 or BamHI-SalI digested pGEX6P to generate pET_trxC or pGST_trxC respectively. To generate site directed TrxCL32P mutant a 366 pb fragment was amplified by two-step PCR using oligonucleotides TrxC_NdeI-trxC_L32P_Rv and trxC_L32P_Fw-TrxC_NotI, that introduced the desired mutation, and cloned into pGEMT generating pG_TrxCL32P. The plasmid was cut with NdeI-NotI and a 347 pb fragment was ligated to NdeI-Not digested pET28 to generate pET_TrxCL32P. Sequence of all oligonucleotides are provided in Table S1. TrxC fusion proteins were expressed in E. coli BL21. 200 mL of culture was grown in Luria broth medium to an optical density at 600 nm of 0.6, cooled to 4 ˝ C, induced with 0.5 mM isopropylβ -D-thiogalactopyranoside for 2.5 h and harvested by centrifugation. For his tagged TrxC purification, cells were resuspended in 50 mM Tris-HCl pH 8.0, 500 mM NaCl, 1 mM PMSF and disrupted (20 kHz, 75 W) on ice for 3 min (in 30-s periods) in a Branson sonicator. Lysates were clarified by centrifugation 20,000 ˆ gfor 30 min. Supernatants were supplemented with imidazole to a final concentration of 25 mM and loaded onto a 2 mL Ni-NTA agarose (IBA, Goettingen, Germany catalog #2-3201-025) column for affinity chromatography purification. Column was washed with 50 mM Tris-HCl pH 8.0, 500 mM NaCl, 25 mM imidazole until no protein was detected and bound recombinant proteins were eluted with 250 mM of imidazole in 50 mM Tris-HCl pH 8.0, 500 mM NaCl. For GST and GST-TrxC purification cell pellets were resuspended in 5 mL of PBS buffer (150 mM NaCl, 16 mM Na2HPO4, 4 mM NaH2PO4, 1 mM phenylmethylsulfonyl fluoride, 7 mM β-mercaptoethanol) supplemented with 0.1% Triton X-100. Cells extract were prepared as above, mixed with 1 mL of glutathione agarose beads (GE Healthcare, Little Chalfont catalog #17-0756-01) and incubated for 2 h at 4 ˝ C with gentle agitation. Then beads were transferred to a column and washed extensively with PBS buffer until no more protein was eluted from the column. GST and GST-TrxC were eluted in 50 mM Tris HCl pH 8.0, 10 mM GSH. TrxC was excised by digestion with PreScission Protease (GE Healthcare, Little Chalfont catalog #27084301) following manufacturer’s instructions. Antioxidants 2018,7, 164 4 of 15 2.4. Mutant Construction To generate site STXC2 mutant, a 1170 pb fragment was amplified using oligonucleotides trxCHIII and trxCXhoI and cloned into pBS digested with the same enzymes generating pSTXC1. A CK1 cassette was cloned into pSTXC1 digested with HincII (deleting 345 pb and expanding the whole ORF) generating pSTXC2. This plasmid was used to transform Synechocystis generating STXC2 strain. For the overexpression strains a XbaI-NotI fragment from pET_trxC was cloned into pN:PpetE:Nat (a plasmid containing glnN gene in which the petE regulatory region, a multiple cloning site and a Nourseothricin resistance cassette has been cloned [ 27 ]) digested in the same way generating pNPpetEtrxC. This plasmid was used to transform both WT and STXC2 generating WTOE and STXCOE strains (Table 1). 2.5. FBPase Activation Assay Recombinant pea FBPase was purified by Ni nickel-affinity column as previously described [ 28 ] and the eluted protein was desalted with a PD10 column in 50 mM Tris HCl pH 8.0, 150mM NaCl. 2 µ g of recombinant pea FBPase was incubated with DTT (10 mM or 0.1 mM) and 3–30 µ g of thioredoxins for 10 min in 175 µ L containing 30 mM Tris ¨ HCl pH 8.0, 7 mM MgCl 2 at 30 ˝ C. Then, 25 µ L of fructose-1,6-bisphosphate 50 mM was added and the reaction was incubated for 30 min at 30 ˝ C. Also, 1 mL LPi mix (2.5% sulfuric acid, 7.5 mM ammonium heptamolybdate, 100 mM FeSO 4 ) was added to stop the reaction and the Pi released was measured at 660 nm [ 29 ]. A calibration curve with known concentrations of Na2PO4was used to calculate Pi concentration. 2.6. Insulin Reduction Assay Thioredoxin insulin reduction assay was carried out using 3 µ g of recombinant HisTrxA or HisTrxC in 1 mL of 0.1 M potassium phosphate pH 7.0, 2 mM EDTA, 1 mM DTT and 1mg/ml insulin at 30 ˝ C. The same buffer without DTT was used as reference. Thioredoxin activity was measured by the increase of turbidity at OD 650nm due to insulin precipitation [ 30 ]. Recombinant HisTrxA was expressed and purified as described in [13,14]. 2.7. PAM A pulse amplitude modulated fluorometer DUAL-PAM-100 (Walz, Effeltrich, Germany) was used to monitor chlorophyll, a fluorescence in intact cells adjusted to OD 750nm = 1. Measurements were performed in 1 cm ˆ 1 cm cuvettes at 30 ˝ C. Red (620 nm) actinic lights was used as background light and saturating pulses (10,000 µ Mol photons m ´2 s ´1 , 635 nm, 300 ms) were applied to transiently close all PSII centers. The maximal photochemical efficiency of PSII (Fv/Fm) was measured in the presence of 3-(3,4-dichlorophenyl)-1,1-dimethylurea (DCMU). Fm’ dark is defined as the Fm of cells adapted to dark conditions for 10 min. Non-photochemical-quenching (NPQ) was calculated as Fm-Fm’dark/Fm’dark as described in [31]. 2.8. Oxygen Evolution Oxygen evolution was measured in Clark-type oxygen electrode (Hansatech Chlorolab 2) using mid-logarithmic (OD 750nm = 0.8–1) cultures adjusted to OD 750nm = 0.5 in BG11 pH 7.5 media supplemented with 20 mM NaHCO3using white LED light. 3. Results 3.1. TrxC Is an Atypical Thioredoxin Exclusively Found in Cyanobacteria Of the four thioredoxins found in cyanobacteria, only TrxC is not homologous to thioredoxins present in plant chloroplasts. Furthermore, analysis of its sequence shows an altered catalytic site WCGL/V/IC that is highly conserved in cyanobacteria (Figure 1A). Despite the N terminal part Antioxidants 2018,7, 164 5 of 15 of the sequence being similar to other Trx, most of the protein shows a clear divergence from other thioredoxin in cyanobacteria and less conservation in the C-terminal sequence of the protein (Figure 1A). There is no other thioredoxin sequence with a similar active site in any other organism. Moreover, all cyanobacterial groups except Prochlorococcus and Gloeobacteria contain at least one species that presents a trxC gene (Table S2) existing in more than 200 strains. This distribution suggests that this protein may have an important role in cyanobacterial environmental adaptation or metabolism. Antioxidants 2018, 7, x FOR PEER REVIEW 5 of 15 thioredoxin in cyanobacteria and less conservation in the C-terminal sequence of the protein (Figure 1A). There is no other thioredoxin sequence with a similar active site in any other organism. Moreover, all cyanobacterial groups except Prochlorococcus and Gloeobacteria contain at least one species that presents a trxC gene (Table S2) existing in more than 200 strains. This distribution suggests that this protein may have an important role in cyanobacterial environmental adaptation or metabolism. Figure 1. TrxC is an atypical thioredoxin. (A) Sequence logo of TrxC proteins form cyanobacteria. TrxC proteins were identified using blast at NCBI and manually curated to retain only those with a WCGL/V/IC sequence (269 sequences). This sequences were aligned using muscle and the alignment was submitted to weblogo3 to generate the consensus sequence shown. (B) Insulin reduction assay. 3 µM of recombinant TrxA (■), TrxC (●) and TrxCL32P (▲) were incubated with insulin in the presence of 1 mM of DTT. Insulin precipitation was measured as an increase in absorbance at 650 nm. Three independent purification were assayed for TrxC and two for TrxCL32P with identical results to the one shown. (C) FBPase activation assay. Oxidased pea FBPase was preincubated for 30 min with 100 µM DTT (control), 10 mM DTT or 100 µM DTT and 3 µM of TrxA, TrxC, TrxCL32P or 30 µM GST- TrxC. Data are the mean and standard error of 3 independent assays. In order to characterize TrxC function we have produced it as recombinant His-tagged version in E. coli. The protein was expressed at high levels although it was difficult to purify as it was mostly in the insoluble fraction, probably forming inclusion bodies. Only low amounts were recovered in the soluble fraction and this protein was prone to precipitation when concentrated, but we purified enough protein to assay its activity. In contrast to TrxA (m-type), TrxC did not reduce insulin in the presence of DTT (Figure 1B). To further characterize the protein, we analyzed chloroplast pea FBPase Figure 1. TrxC is an atypical thioredoxin. ( A ) Sequence logo of TrxC proteins form cyanobacteria. TrxC proteins were identified using blast at NCBI and manually curated to retain only those with a WCGL/V/IC sequence (269 sequences). This sequences were aligned using muscle and the alignment was submitted to weblogo3 to generate the consensus sequence shown. ( B ) Insulin reduction assay. 3 µ M of recombinant TrxA ( ‚ ), TrxC ( ‚ ) and TrxCL32P ( N ) were incubated with insulin in the presence of 1 mM of DTT. Insulin precipitation was measured as an increase in absorbance at 650 nm. Three independent purification were assayed for TrxC and two for TrxCL32P with identical results to the one shown. ( C ) FBPase activation assay. Oxidased pea FBPase was preincubated for 30 min with 100 µ M DTT (control), 10 mM DTT or 100 µ M DTT and 3 µ M of TrxA, TrxC, TrxCL32P or 30 µ M GST-TrxC. Data are the mean and standard error of 3 independent assays. Antioxidants 2018,7, 164 6 of 15 In order to characterize TrxC function we have produced it as recombinant His-tagged version in E. coli. The protein was expressed at high levels although it was difficult to purify as it was mostly in the insoluble fraction, probably forming inclusion bodies. Only low amounts were recovered in the soluble fraction and this protein was prone to precipitation when concentrated, but we purified enough protein to assay its activity. In contrast to TrxA (m-type), TrxC did not reduce insulin in the presence of DTT (Figure 1B). To further characterize the protein, we analyzed chloroplast pea FBPase activation by TrxC. In the same way as in the insulin reduction assay, TrxA was able to activate chloroplast FBPase, while TrxC did not activate it. As TrxC presents an unconventional active site WCGLC, this was changed to a canonical Trx active site (WGCPC) by site directed mutagenesis generating TrxCL32P. The protein was purified and used in both assays with identical results to the WT protein (Figure 1B,C). To further confirm these results we generate a recombinant version of the protein fussed to GST (GST-TrxC), which allowed us to purify higher amounts of soluble recombinant protein. Although TrxC can be excised from GST by protease digestion, the TrxC portion precipitated rapidly after digestion and therefore only GST-TrxC was used. GST and GST-TrxC were used to assay chloroplast FBPase activation at higher concentrations (30 µ M) but neither of the proteins activated FBPase. These data suggest that TrxC is not active in the classical thioredoxin assays and that this is not related to its unconventional active site. 3.2. TrxC Mutant and Overexpression Strains Characterization To investigate TrxC physiological function, a mutant lacking trxC (STXC2 mutant strain) and mutants over-expressing a His-tagged version of the protein (HisTrxC) were constructed. For overexpression, the His-tagged trxC gene was placed under control of the copper inducible petE promoter in a WT and STXC2 backgrounds, generating WTOE and STXCOE strains, respectively (Table 1). All mutants were verified by PCR analysis and shown to be completely segregated (Figure 2A,B). All strains presented similar growth rates to the WT in BG11C plates and liquid media under our standard growth conditions ([ 25 ]; data not shown). TrxC protein levels were analyzed in the mutants by western blot using TrxC antibodies in whole cell extracts. TrxC was detected in WT and WTOE strains but not in STXC2 and STXCOE, while a band corresponding to the recombinant HisTrxC was detected in both WTOE and STXCOE strains. Furthermore, HisTrxC expression was regulated by the presence of copper in these strains, the amount of HisTrxC increased after copper addition in both strains. Surprisingly HisTrxC expression levels were higher than endogenous TrxC levels in the WT strain even in the absence of copper in the media in both WTOE and STXCOE strains and was further elevated in the presence of copper (Figure 2C). To study the TrxC impact on redox regulation in Synechocystis, we have analyzed the expression levels of other redox related proteins in WT and trxC mutant strains. All strains were grown to exponential phase in copper containing media and levels of different proteins were analyzed by western blot. Levels of the other thioredoxins (TrxA, TrxB and TrxQ), glutaredoxins (GrxA and GrxC) and 2-cys peroxiredoxin did not change in the mutants (Figure 3). In contrast, levels of DDOR (slr0600) increased in both the WTOE y STXCOE strains (Figure 3) in which TrxC levels were also increased (Figure 2C). These data showed that TrxC did not affect levels of other thioredoxins and glutaredoxins. The changes in DDOR protein levels suggest that these two proteins could be functionally related. Antioxidants 2018,7, 164 7 of 15 Antioxidants 2018, 7, x FOR PEER REVIEW 7 of 15 Figure 2. Construction of WT, WT_OE, STXC and STXCOE strains. (A) Schematic representation of the trxC and glnN loci in the WT and mutant strains. (B) PCR analysis of the mutant strains using the oligonucleotides indicated in (A). (C) Western-blot analysis of TrxC protein levels in the WT, WTOE, STXC and STXCOE strains. Cells were grown in BG11 supplemented with 1% CO 2 to mid-log growth phase and 1 OD 750nm was collected before and after 2 h of 0.3 µM Cu addition. The pellet was resuspended in 1× Laemmli buffer and boiled for 10 min, then 20 µL of the boiled cell suspension were loaded on and gel and analyzed by western blot. Figure 2. Construction of WT, WT_OE, STXC and STXCOE strains. ( A ) Schematic representation of the trxC and glnN loci in the WT and mutant strains. ( B ) PCR analysis of the mutant strains using the oligonucleotides indicated in (A). ( C ) Western-blot analysis of TrxC protein levels in the WT, WTOE, STXC and STXCOE strains. Cells were grown in BG11 supplemented with 1% CO 2 to mid-log growth phase and 1 OD 750nm was collected before and after 2 h of 0.3 µ M Cu addition. The pellet was resuspended in 1 ˆ Laemmli buffer and boiled for 10 min, then 20 µ L of the boiled cell suspension were loaded on and gel and analyzed by western blot. Antioxidants 2018,7, 164 8 of 15 Antioxidants 2018, 7, x FOR PEER REVIEW 8 of 15 Figure 3. Redox related proteins in WT, WTOE, STXC2 and STXCOE strains. Western-blot analysis of TrxA, TrxB, TrxQ, GrxA, GrxC, DDOR and 2-cysprx in the WT, WTOE, STXC2 and STXCOE strains. Cells were grown in BG11 supplemented with 1% CO 2 to mid-log growth phase and collected. Cells were broken and 20 µg of total protein from soluble extracts were separated by SDS PAGE and analysed by western blot to detect the different proteins using specific antibodies. Experiments were repeated at least two (for TrxQ antibody) or three (all other antibodies) times with biological independent samples. SDS PAGE: sodium dodecyl sulfate polyacrylamide gel electrophoresis. 3.3. Growth of TrxC Mutant Strains under Low CO 2 Conditions. In order to investigate the physiological role of TrxC, trxC mutant strains were analyzed under several growth conditions. No difference between strains was detected in BG11C (containing NaHCO 3 ). We have only detected differences in growth and/or pigmentation under moderate light intensities and low carbon conditions (BG11 pH 7.5 bubble with air + 1% CO 2 or BG11 pH 7.5 bubbled with air). We have selected an intermediate light intensity (180 µmol m −2 s −1 ) and air bubbled cultures (low carbon availability) as these were the most consistent growth conditions in which we were able to detect differences. At higher light intensities (500 µmol m −2 s −1 ), even the WT strain showed impaired growth and in some cases died. All strains were previously adapted to BG11 pH 7.5 and high carbon (bubbled with air + 1% CO 2 ) in low light (50 µmol m −2 s −1 ) until they reached late exponential phase (OD 750nm = 1–2) and then were diluted to OD 750nm = 0.2 and shifted 180 µmol m −2 s −1 in either high carbon (air + 1% CO 2 ) or low carbon conditions (air); this adaptation step was necessary for the strains to show consistent growth under this condition. In cultures bubbled with air + 1% CO 2 all strains grew at similar rates but differences in color were observed (Figure 4A). These are clearly visible in whole cell absorption spectra in which STXC2 showed a higher absorption at 485 nm, which corresponds to carotenoids absorption maxima, than the WT (Figure 4B). Moreover, both overexpression strains showed lower carotenoid contents (Figure 4B) making them to appear bluish when compared to WT or STXC2 strains. These phenotypes were exacerbated in air bubbled cultures in which carbon availability is further reduced. Under these conditions, all strains showed a reduced growth rate (Figure 4C), although growth of WTOE and STXCOE strains was reduced more than that of WT and STXC2 strains (Figure 4C). Despite the STXC2 strain growth rate being very similar to the Figure 3. Redox related proteins in WT, WTOE, STXC2 and STXCOE strains. Western-blot analysis of TrxA, TrxB, TrxQ, GrxA, GrxC, DDOR and 2-cysprx in the WT, WTOE, STXC2 and STXCOE strains. Cells were grown in BG11 supplemented with 1% CO 2 to mid-log growth phase and collected. Cells were broken and 20 µ g of total protein from soluble extracts were separated by SDS PAGE and analysed by western blot to detect the different proteins using specific antibodies. Experiments were repeated at least two (for TrxQ antibody) or three (all other antibodies) times with biological independent samples. SDS PAGE: sodium dodecyl sulfate polyacrylamide gel electrophoresis. 3.3. Growth of TrxC Mutant Strains under Low CO2Conditions In order to investigate the physiological role of TrxC, trxC mutant strains were analyzed under several growth conditions. No difference between strains was detected in BG11C (containing NaHCO 3 ). We have only detected differences in growth and/or pigmentation under moderate light intensities and low carbon conditions (BG11 pH 7.5 bubble with air + 1% CO 2 or BG11 pH 7.5 bubbled with air). We have selected an intermediate light intensity (180 µ Mol m ´2 s ´1 ) and air bubbled cultures (low carbon availability) as these were the most consistent growth conditions in which we were able to detect differences. At higher light intensities (500 µ Mol m ´2 s ´1 ), even the WT strain showed impaired growth and in some cases died. All strains were previously adapted to BG11 pH 7.5 and high carbon (bubbled with air + 1% CO 2 ) in low light (50 µ Mol m ´2 s ´1 ) until they reached late exponential phase (OD 750nm = 1–2) and then were diluted to OD 750nm = 0.2 and shifted 180 µ Mol m ´2 s ´1 in either high carbon (air + 1% CO 2 ) or low carbon conditions (air); this adaptation step was necessary for the strains to show consistent growth under this condition. In cultures bubbled with air + 1% CO 2 all strains grew at similar rates but differences in color were observed (Figure 4A). These are clearly visible in whole cell absorption spectra in which STXC2 showed a higher absorption at 485 nm, which corresponds to carotenoids absorption maxima, than the WT (Figure 4B). Moreover, both overexpression strains showed lower carotenoid contents (Figure 4B) making them to appear bluish when compared to WT or STXC2 strains. These phenotypes were exacerbated in air bubbled cultures in which carbon availability is further reduced. Under these conditions, all strains showed Antioxidants 2018,7, 164 9 of 15 a reduced growth rate (Figure 4C), although growth of WTOE and STXCOE strains was reduced more than that of WT and STXC2 strains (Figure 4C). Despite the STXC2 strain growth rate being very similar to the WT, it contained even higher carotenoids which gave a yellow appearance to the cultures (Figure 4D,E). Overexpression strains showed similar carotenoids contents that were lower than WT levels (Figure 4D). Furthermore, the STXC2 strain showed higher chlorophyll contents than other strains in both 1% CO 2 ( 4.1 ˘0.3 µg chl OD750nm ´1 for STXC2 vs 3.5 ˘ 0.5, 3.5 ˘ 0.2 and 3.4 ˘ 0.3 µ g chl OD 750nm ´1 for WT, WTOE and STXCOE, respectively) and air bubbled conditions (3.83 ˘ 0.5 µ g chl OD 750nm ´1 for STXC2 vs. 3.1 ˘ 0.3, 3.5 ˘ 0.3 and 3.5 ˘ 0.2 µ g chl OD 750nm ´1 for WT, WTOE and STXCOE, respectively). All these results suggest that TrxC is involved in adaptation to low carbon conditions and that it could be mediated by regulating pigment content in Synechocystis. Antioxidants 2018, 7, x FOR PEER REVIEW 9 of 15 WT, it contained even higher carotenoids which gave a yellow appearance to the cultures (Figure 4D,E). Overexpression strains showed similar carotenoids contents that were lower than WT levels (Figure 4D). Furthermore, the STXC2 strain showed higher chlorophyll contents than other strains in both 1% CO 2 (4.1 ± 0.3 µg chl OD 750nm−1 for STXC2 vs 3.5 ± 0.5, 3.5 ± 0.2 and 3.4 ± 0.3 µg chl OD 750nm−1 for WT, WTOE and STXCOE, respectively) and air bubbled conditions (3.83 ± 0.5 µg chl OD 750nm−1 for STXC2 vs. 3.1 ± 0.3, 3.5 ± 0.3 and 3.5 ± 0.2 µg chl OD 750nm−1 for WT, WTOE and STXCOE, respectively). All these results suggest that TrxC is involved in adaptation to low carbon conditions and that it could be mediated by regulating pigment content in Synechocystis. Figure 4. Overexpression of trxC slows growth in low carbon conditions. (A) Growth of trxC mutant strains in high carbon conditions. WT (□), STXC2 (○), WTOE (■) and STXCOE (●) were grown in BG11 pH 7.5 under low light until the exponential phase, diluted to 0.2 OD 750nm and shifted to 180 µmol m −2 s −1 light intensity bubbled with air + 1% CO 2 . Growth was monitored by measuring OD 750nm . Data represented are the mean and standard error of 3–4 (depending on the time point) biological independent cultures. (B) Whole cell spectra of WT (blue solid line), STXC2 (green solid line), WTOE (blue dashed line) and STXCOE (green dashed line) grown as in (A). (C) Growth of trxC mutant strains in low carbon conditions. WT (□), STXC2 (○), WTOE (■) and STXCOE (●) were grown in BG11 pH 7.5 under low light until the exponential phase, diluted to 0.2 OD 750nm and shifted to 180 µmol m −2 s −1 light intensity and bubbled with air. Growth was monitored by measuring OD 750nm . Data represented are the mean and standard error of 3–4 (depending on the time point) biological independent cultures. (D) Whole cell spectra of WT (blue solid line), STXC2 (green solid line), WTOE (blue dashed line) and STXCOE (green dashed line) grown as in (C). (E) Photograph of WT, STXC2, WTOE and STXCOE cultures grown in BG11 pH 7.5 bubbled with air + 1% CO 2 or air. Figure 4. Overexpression of trxC slows growth in low carbon conditions. ( A ) Growth of trxC mutant strains in high carbon conditions. WT ( ˝ ), STXC2 ( # ), WTOE ( ‚ ) and STXCOE ( ‚ ) were grown in BG11 pH 7.5 under low light until the exponential phase, diluted to 0.2 OD 750nm and shifted to 180 µ Mol m ´2 s ´1 light intensity bubbled with air + 1% CO 2 . Growth was monitored by measuring OD 750nm . Data represented are the mean and standard error of 3–4 (depending on the time point) biological independent cultures. ( B ) Whole cell spectra of WT (blue solid line), STXC2 (green solid line), WTOE (blue dashed line) and STXCOE (green dashed line) grown as in (A). ( C ) Growth of trxC mutant strains in low carbon conditions. WT ( ˝ ), STXC2 ( # ), WTOE ( ‚ ) and STXCOE ( ‚ ) were grown in BG11 pH 7.5 under low light until the exponential phase, diluted to 0.2 OD 750nm and shifted to 180 µ Mol m ´2 s ´1 light intensity and bubbled with air. Growth was monitored by measuring OD 750nm . Data represented are the mean and standard error of 3–4 (depending on the time point) biological independent cultures. ( D ) Whole cell spectra of WT (blue solid line), STXC2 (green solid line), WTOE (blue dashed line) and STXCOE (green dashed line) grown as in (C). ( E ) Photograph of WT, STXC2, WTOE and STXCOE cultures grown in BG11 pH 7.5 bubbled with air + 1% CO2or air.