Slr0320 Is Crucial for Optimal Function of Photosystem II during High Light Acclimation in Synechocystis sp. PCC 6803

Upon exposure of photosynthetic organisms to high light (HL), several HL acclimation responses are triggered. Herein, we identified a novel gene, slr0320, critical for HL acclimation in Synechocystis sp. PCC 6803. The growth rate of the Δslr0320 mutant was similar to wild type (WT) under normal light (NL) but severely declined under HL. Net photosynthesis of the mutant was lower under HL, but maximum photosystem II (PSII) activity was higher under NL and HL. Immunodetection revealed the accumulation and assembly of PSII were similar between WT and the mutant. Chlorophyll fluorescence traces showed the stable fluorescence of the mutant under light was much higher. Kinetics of single flash-induced chlorophyll fluorescence increase and decay revealed the slower electron transfer from QA to QB in the mutant. These data indicate that, in the Δslr0320 mutant, the number of functional PSIIs was comparable to WT even under HL but the electron transfer between QA and QB was inefficient. Quantitative proteomics and real-time PCR revealed that expression profiles of psbL, psbH and psbI were significantly altered in the Δslr0320 mutant. Thus, Slr0320 protein plays critical roles in optimizing PSII activity during HL acclimation and is essential for PSII electron transfer from QA to QB.


Introduction
Light is an essential environmental factor for all photosynthetic organisms, including higher plants, algae and cyanobacteria. Photosynthetic organisms perform photosynthesis by converting absorbed sunlight energy into chemical energy. Photosystem II (PSII), as the first enzyme of photochemistry, plays essential roles in photosynthesis. The PSII is a multi-subunit protein complex embedded in the thylakoid membrane, catalyzing water oxidation, plastoquinone reduction and oxygen release [1][2][3]. The reaction center (RC) of PSII consists of heterodimeric D1 and D2 proteins that are flanked by chlorophyll-binding proteins CP43 and CP47. Upon absorbing light, light-driven charge separation occurs between the P680 and pheophytin in the RC [4,5]. The charge-separated electron could in turn transfer to two quinone acceptors, Q A and Q B . The oxygen evolution complex (OEC), harboring a manganese cluster as well as PsbO, PsbU and PsbV in cyanobacteria, is located at the lumen side of PSII and catalyzes the water-splitting, releasing oxygen. In addition to the subunits mentioned above, PSII contains 13 subunits with molecular weights less than 10 kDa, known as low molecular weight (LMW) subunits (PsbE, PsbF, PsbH, PsbI, PsbJ, PsbK, PsbL, PsbM, PsbN, PsbT, PsbX, PsbY, PsbZ) in Synechocystis sp. PCC 6803 (hereafter Synechocystis). These LMW subunits are thought to play important roles in PSII biogenesis, structure, assembly and electron transfer [6][7][8].
Light is necessary for photosynthesis, but it can cause damages and become dangerous as well. Given that the light environment for photosynthetic organisms is constantly changing, the situation becomes more complicated. The D1 protein of PSII is the main target of the light-induced damages and these unavoidable damages are positively correlated to light intensities [9,10]. When the light intensities increase and the imbalance between the PSII photodamage and PSII repair occurs, the photoinhibition become apparent [11]. Synechocystis has developed several high light (HL) acclimation strategies to alleviate photoinhibition, such as the upregulation of D1 encoding genes [12][13][14], highly efficient proteolytic degradation of damaged D1 [15][16][17], elevated levels of specific carotenoids and translation elongation factors [18,19], and specific flavodirron proteins [20]. Owning to deeper investigations of Synechocystis gene functions, more and more hypothetical and known genes have been identified as being involved in the HL acclimation and protection of PSII, such as slr0151 [21] and five small CAB-like proteins (ScpA-E), of which ScpB-E are also known as high light-induced proteins (HliA-D) [22,23].
This process is not completely clear, although many efforts and some progresses have been made in discovery of novel genes involved in HL acclimation. To investigate new genes involved in HL acclimation, a random transposon-insertional mutant library of Synechocystis was screened. A novel gene, slr0320, was identified as being involved in HL acclimation and optimal function of PSII. The possible mechanism of the slr0320 involved in HL acclimation was discussed.

Culture Conditions
The culture conditions were set according to Zhang et al. [24]. The Synechocystis sp. PCC 6803 glucose-tolerant strain (used as wild type, WT) [25], the ∆slr0320 mutant and complementary strain PpetJ::slr0320 were grown either in liquid or on solid BG-11 medium buffered with 20 mM HEPES-NaOH (pH 7.5) at 30 • C. The solid BG-11 medium was supplemented with 1.5% agar. The liquid cultures were aerated by gentle shanking at 120 rpm. The cells were illuminated under continuous white LED light (4100 K) with photon flux density of either 50 µmol photons m −2 s −1 (normal light, NL) or 220 µmol photons m −2 s −1 (high light, HL). All mutant strains were maintained in the presence of appropriate antibiotics.

Mutant Construction and Screening
The Synechocystis mutant library was constructed according to Ozaki et al. [26]. The liquid wild type cells were transformed with the transposon mutagenesis library, which contains a chloramphenicol-resistant (Cm R ) cassette. Following transformation, Cm R mutants showing slower growth rates under HL but normal growth rates under NL on solid plates were selected. Genomic DNA extracted from selected mutants were digested by HhaI and, after self-ligation, used as templates for PCR with primers (P4) complementary to two ends of the chloramphenicol-resistant cassette ( Table 1). The exact genome positions of insertions of transposons with Cm R were determined by sequencing PCR products. Table 1. Primers used in this assay.

Forward Sequence (5 -3 )
Reverse Sequence (5 -3 ) TTTAGAAAACAGCAACCAGT GGCAGGAAAAAGACCAACCT Due to disruptions by a frameshift, spkA (sll1574-sll1575) is considered as a neutral site in Synechocystis [27]. The complementary strain PpetJ::slr0320 was constructed according to Eisenhut et al. [28]. The wild type copy of slr0320 gene driven by a petJ promoter was integrated into the spkA site of genome in the ∆slr0320 mutant.

Protein Isolation, Electrophoresis, and Immunodetection
The total cell extract as well as membrane and soluble fractions were isolated according to Zhang et al. [31]. Cells from liquid cultures were harvested at 5000 rpm, 4 • C for 5 min and washed with washing buffer (50 mM HEPES-NaOH pH 7.5, 30 mM CaCl 2 ). The harvested cells were then resuspended in resuspension buffer (50 mM HEPES-NaOH pH 7.5, 30 mM CaCl 2 , 800 mM sorbitol and 1 mM 6-aminohexanoic acid) and broken in the presence of glass beads (150-212 µm, Sigma, St. Louis, MO, USA) by manually vortexing eight times at maximal speed for 1 min at 4 • C with 1 min cooling on ice between cycles. After removal of glass beads and cell debris at 2000× g for 5 min at 4 • C, the total cell extract was obtained. The total cell extract, which contains membrane and soluble fractions, could be further separated by centrifugation at 30,000× g for 30 min at 4 • C. After centrifugation, the membrane was pelleted down. Additionally, the soluble fraction remained in supernatant. The pelleted membrane was resuspended in storage buffer (50 m HEPES-NaOH pH 7.5, 30 mM CaCl 2 , 600 mM sucrose and 1 M betaine), and directly used for further analysis or frozen by liquid nitrogen in aliquots.
Protein samples were separated and immunodetected according to Zhang et al. [24]. Membrane samples for BN-PAGE (blue native-polyacrylamide gel electrophoresis) were performed as follows. The membrane samples were washed with washing buffer (50 mM Bis-Tris pH 7.0, 330 mM sorbitol) supplemented with 1 mM AEBSF (Thermo Scientific, Carlsbad, CA, USA) which is a protease inhibitor and pelleted by centrifugation at 30,000× g for 10 min at 4 • C. The pellets were then suspended in the buffer containing 25 mM Bis-Tris pH 7.0, 1 mM AEBSF, 20% glycerol (w/v) and 10 mM MgCl 2 with chlorophyll concentration at 1 µg/µL. The protein samples were further solubilized by gently adding an equal volume of 3% n-dodecyl-β-D-maltoside to the same buffer, followed by 20 min incubation on ice. Insoluble material was removed by centrifugation at 30,000× g for 15 min at 4 • C. The collected supernatant was mixed with 1/10 volume sample buffer containing 5% Serva blue G (w/v), 100 mM Bis-Tris, pH 7.0, 30% sucrose (w/v), and 0.5 M 6-aminohexanoic acid and loaded onto 5 to 12.5% acrylamide gradient gel. Electrophoresis for BN-PAGE was performed at 4 • C by increasing voltage gradually from 50 up to 200 V during the 5.5 h run. Protein samples for SDS-PAGE (sodium dodecyl sulfate-polyacrylamide gel electrophoresis) were solubilized in Laemmli SDS sample buffer [32] containing 5% β-mercaptoethanol and 6 M urea at room temperature for 1h and then separated by 12% SDS-PAGE with 6 M urea.
After electrophoresis, the proteins were electrotransferred to a polyvinylidene fluoride (PVDF) membrane (Immobilon-P, Millipore, Burlington, MA, USA) by a semidry apparatus and immunodetected with specific antibodies. NdhK and NdhH antibodies were kind gifts from Weimin Ma from Shanghai Normal University and Hualing Mi from Institute of Plant Physiology and Ecology, Chinese Academy of Sciences, respectively. FNR (ferredoxin: NADP + oxidoreductase) antibody was from Ghada Ajlani from Centre National de la Recherche. The D1, D2, CP43, CP47, PsbO, PsaA, RbcL, PetC and AtpB antibodies were purchased from Agrisera (Vennes, Sweden).

Oxygen Evolution Measurements
The steady-state oxygen evolution rate was measured by a Clark-type oxygen electrode (Hansatech) at saturating white light intensity (2500 µmol photons m −2 s −1 ) according to Zhang et al. [33]. The cells of WT, ∆slr0320 and PpetJ::slr0320 strains grown in liquid BG-11 medium under NL and HL were harvested and resuspended in fresh BG-11 medium to chlorophyll concentration of 5 µg/mL. Net photosynthesis was measured in the presence of 10 mM NaHCO 3 . PSII activities were measured in the presence of 0.5 mM DCBQ (2,6-dichloro-p-benzoquinone) and 1 mM K 3 Fe(CN) 6 .
Flash-induced fluorescence increase and the subsequent decay of chlorophyll fluorescence yield were performed on a fluorometer FL 3500 (PSI Instruments) according to Zhang et al. [20]. The cells of WT, ∆slr0320 and PpetJ::slr0320 strains, grown under NL and HL, were harvested and resuspended to chlorophyll concentration of 5 µg/mL. After 10 min dark adaption, cells were excited with a single, short saturation flash-light. According to Cser et al. [34], a fitting function with three components was applied: Equation (1) was used for deconvolution of the measured curves, where F(t) is the variable fluorescence yield at time t, F 0 is the basic fluorescence level before the flash, A1 to A3 are the amplitudes, and T1 to T3 are the time constants. The Joliot model [35] was used to correct the nonlinear correlation between fluorescence yield and the redox state of Q A with a value of 0.5 for the energy transfer parameter between the PSII units [20,36].

Quantitative Proteomics Analysis
The protein preparation was performed according to Ge et al. [37]. The harvested cells were broken by a bead beater. The total cell extract was precipitation with ice-cold 10% trichloroacetic acid in acetone at −20 • C, washed with acetone, and resolubilized with 4% sodium dodecyl sulfate (SDS) in 0.1 M Tris-HCl, pH 7.6. The BCA protein assay kit (Thermo Scientific, Rockford, IL, USA) was used to determine the protein concentration. The filter-aided sample preparation (FASP) method [38] was used to digest total proteins to tryptic peptides. Briefly, DTT (dithiothreitol) (100 mm) and iodoacetamide (55 mm) were used for reduction and alkylation of the sample proteins and then the reduced proteins were transferred into the Microcon YM-30 centrifugal filter unites to exchange the denaturing buffer with triethylammonium bicarbonate (TEAB, 0.1 M) buffer. Proteins were then digested with trypsin (Promega, Madison, WI, USA) at 37 • C overnight, and the tryptic peptides were labeled with TMT reagents (Thermo Scientific, Rockford, IL, USA) by incubation at room temperature in dark for 2 h. The labeling process was stopped with 5% hydroxylamine and the equal amounts of each labeled samples were mixed together for next prefractionation with reversed phase (RP) high performance liquid chromatography (HPLC). After RP and HPLC, 4 fractions were collected and dried by a SpeedVac concentrator and stored at −20 • C for LC-MS/MS analysis. The LC-MS/MS analysis was performed using an Orbitrap Lumos mass spectrometer (Thermo Scientific, Rockford, IL Waltham, MA, USA) coupled online to an Easy-nLC 1000 in the data-dependent mode. The database search for the raw MS files was performed using the software MaxQuant (version 1.6.1, Max Planck Institute of Biochemistry, Martinsried, Germany) [39]. The Synechocystis proteome sequence database which contains 3672 entries was downloaded from CyanoBase. Bioinformatics and statistical analyses were performed using the software Perseus (version 1.5.4.1, Max Planck Institute of Biochemistry, Martinsried, Germany) [40], Origin2018 (version b9.5.1.195, OriginLab Corp., Northampton, MA, USA).

Identification of slr0320 under High Light
To identify genes involved in high light acclimation, a Synechocystis transposoninsertion mutant library was screened by comparison of the growth under high light (HL, 220 µmol photons m −2 s −1 ) and normal light (NL, 50 µmol photons m −2 s −1 ). By screening approximately 1000 mutants, 90 mutants were identified as photosensitive. Among identified mutants, a mutant showing significantly retarded growth under HL but normal growth under NL was selected for further analysis.
PCR sequencing data showed that the transposon insertion occurred at position 709 of slr0320 (the A of the start codon ATG was set as position 1). The slr0320 gene is 1596 bp in length and consists of two conserved domains: a B12-binding domain from position 373 to 528 and a radical S-adenosyl methionine (SAM) domain from position 724 to 1209 ( Figure 1a). (a) Organization of slr0320 in Synechocystis. The insertion position of a transposon including a chloramphenicol-resistant cassette (Cm R , green) flanked by two insertion sequences (Tn, orange) is indicated; two conserved domains of slr0320 were indicated under the slr0320 coding region (red); (b) growth of WT, ∆slr0320 and PpetJ::slr0320 strains on agar plates under normal light (NL) and high light (HL). The cells grown under NL were harvested and resuspended in fresh BG-11 medium. Then, they were diluted to 0.5, 0.05, 0.005 at OD750 and 10 µL liquid cells were placed on the agar plate. The pictures were taken after 6 days; (c) growth curves of WT, ∆slr0320 and PpetJ::slr0320 strains in liquid medium under NL and HL. The initial OD750 was 0.05. Student's t test was applied. Red asterisks mark significant differences for the ∆slr0320 mutant vs. WT and PpetJ::slr0320 strains, with a significance level of p < 0.05. The results are plotted as the means from three independent experiments ± SD.
To confirm the phenotype of the mutant was caused by the defect in slr0320 gene, a complementary strain, PpetJ::slr0320, was constructed. The WT, ∆slr0320 and PpetJ::slr0320 strains were grown on BG-11 agar plate under HL and NL. Under NL conditions, there were no apparent differences among WT, the ∆slr0320 mutant and the complementary strain PpetJ::slr0320. However, under HL condition, the ∆slr0320 mutant was grown much slower than WT and the PpetJ::slr0320 strain (Figure 1b). The growth phenotypes of the cells grown in liquid cultures were similar to that grown on agar plate. Under NL, growth rates of WT, ∆slr0320 and the PpetJ::slr0320 strain were similar. Under HL, growth rates of WT and the PpetJ::slr0320 strain were similar and increased, but the growth rate of the ∆slr0320 mutant was significantly inhibited (Student's t test, p < 0.05) (Figure 1c).
To clarify the reason for the retarded growth rate of the ∆slr0320 mutant, the steadystate oxygen evolution rates of net photosynthesis and PSII activity were monitored. When cells were grown under NL, there was no significant difference of maximal net photosynthesis among WT, ∆slr0320 and PpetJ::slr0320 strains (Student's t test, p < 0.05). However, about 23% decline in the ∆slr0320 mutant was observed when cells were grown under HL. It should be noted that the net photosynthesis rate of the ∆slr0320 mutant grown under HL was similar to that under NL, while the rates of WT and PpetJ::slr0320 strains were significantly increased in HL grown cells (Student's t test, p < 0.05), from 311 to 435 µmol O 2 mg −1 chlorophyll h −1 and 330 to 444 µmol O 2 mg −1 chlorophyll h −1 , respectively. However, when supplemented with artificial quinone acceptor 2,6-dichloro-pbenzoquinone (DCBQ), the PSII activity of the ∆slr0320 mutant was even slightly higher than WT and the PpetJ::slr0320 strain under both HL and NL (Table 2). Table 2. Steady-state oxygen evolution of WT, ∆slr0320 and PpetJ::slr0320 strains grown under NL and HL in the presence of NaHCO 3 and DCBQ. Student's t test was applied. The superscript "a" marks significant differences for WT grown under HL vs. grown under NL in the presence of NaHCO 3 ; the superscript "b" marks significant differences for the complementary PpetJ::slr0320 grown under HL vs. grown under NL in the presence of NaHCO 3 , both with a significance level of p < 0.05 The results are shown as the means from three independent experiments ± SD.

Accumulation of Major Thylakoid
Proteins and PSII Complexes in WT, ∆slr0320 and PpetJ::slr0320 Strains The lower oxygen evolution of net photosynthesis of the ∆slr0320 mutant suggested some modifications of photosynthetic machinery in the mutant. Therefore, the photosynthetic protein contents on the thylakoid membrane in WT, ∆slr0320 and PpetJ::slr0320 strains were analyzed. Western blots of representatives of the major thylakoid membrane protein complexes from the thylakoids isolated from WT, ∆slr0320 and PpetJ::slr0320 strains grown under NL and HL revealed that the differences in amounts of the thylakoid proteins tested were tiny (fold change > 1.4 and Student's t test p < 0.05), if any-probably a slight increase in PSII proteins and decrease in PSI in the ∆slr0320 mutant under HL, a slight decline of cytochrome b 6 f in the ∆slr0320 mutant under both NL and HL and almost unaltered NDH-1M complex (Figure 2a). In addition to the slightly increase in PSII proteins, it is worth noting that the amount of native PSII complexes in the ∆slr0320 mutant showed similar behaviors under both NL and HL. The BN-PAGE followed by immunodetection against D1 showed that the amounts of PSII dimers, monomers, and CP43-less monomer (CP47-RC), which is the PSII assembly intermediate containing the CP47 subunit, in the ∆slr0320 mutant were no less than in WT both under NL and HL (Figure 2b).

Chlorophyll Fluorescence in WT, ∆slr0320 and PpetJ::slr0320 Strains
To monitor photosynthetic electron transfer, chlorophyll fluorescence of WT, ∆slr0320 and PpetJ::slr0320 cells grown under NL and HL was monitored. When cells were grown under NL, the dark-adapted initial fluorescence, Fo, was slightly higher in the ∆slr0320 mutant. Additionally, the light-adapted fluorescence, Fs, remained higher than WT and the complementary strain PpetJ::slr0320, either illuminated with 45 or 320 µmol photons m −2 s −1 actinic light. When cells were grown under HL, the Fo of the ∆slr0320 mutant was comparable to WT and the PpetJ::slr0320 strain, while the Fs became higher than WT and the PpetJ::slr0320 strain under 45 µmol photons m −2 s −1 actinic light and the difference became a little more apparent when the actinic light was increased to 320 µmol photons m −2 s −1 . Under both NL and HL, the variable fluorescence Fv in the ∆slr0320 mutant was higher than that of WT and the PpetJ::slr0320 strain (Figure 3a,b).
To monitor Q A − reoxidation, a single flash induced an increase in chlorophyll fluorescence and the following decay was measured. The fluorescence decay consists of three components, reflecting three pathways of Q A − reoxidation. The fast phase and middle phase reveal reoxidation by bounded PQ (plastoquinone) and free PQ, respectively. The slow phase is caused by charge recombination [36]. The chlorophyll fluorescence decay showed that electron transport processes at the acceptor side of PSII was modified in the ∆slr0320 mutant (Figure 3c). Kinetics analysis demonstrated that apparent slower fluorescence decay in the ∆slr0320 mutant than WT and the complementation strain PpetJ::slr0320 was mainly due to the decrease in the fast phase amplitude and the higher time constants of the fast phase and middle phase (Table 3). In addition, the slow phase in the ∆slr0320 mutant had smaller time constant and higher amplitude than WT and the PpetJ::slr0320 strains. The phenomenon became more evident when the cells were grown under HL (Figure 3c and Table 3).   The results are shown as the means from three independent experiments ± SD.

Quantitative Proteomics of WT and the ∆slr0320 Mutant
To elucidate how Slr0320 operates in HL acclimation, quantitative proteomics was performed on WT and the ∆slr0320 mutant grown under NL and HL. For cells grown under NL, 1988 proteins were shared by WT and the ∆slr0320 mutant. When cells were grown under HL, the number of identified proteins was 1891. Combining these four samples, the number of overlapped proteins was 1785 (Figure 4a and Table S1). Hierarchical clustering analysis of the 1785 proteins using z-scored report ion intensities showed that three biological replicates of WT and the ∆slr0320 mutant, each grown under NL and HL, were clustered ( Figure 4b). Principal component analysis also revealed that each sample containing three biological replicates was successfully separated from others ( Figure 4c). These two analyses confirmed the high reproducibility of the 1785 overlapped proteins. Grouping these proteins according to the functional categories annotated by the CyanoBase [41] showed that coverages of 1785 proteins for all functional categories were almost more than 50% except for the "other categories", "unknow" and "hypothetical" (Figure 4d). Taken together, the high reproducibility and coverage ensured the proteome data were suitable for further analysis.
Before uncovering roles of slr0320 in HL acclimation, proteins involved in HL acclimation of Synechocystis were roughly investigated. The cutoffs of a fold change of 1.4 and p < 0.05 (Student's t test) were applied to filter expressed proteins with significant differences and high confidence. Compared to Synechocystis grown under NL, 551 proteins were downregulated and 389 proteins were upregulated in WT grown under HL (Figure 5a and Table S2). A Fisher's exact test was performed to enrich downregulated proteins in gene ontology (GO), KEGG (Kyoto Encyclopedia of Genes and Genomes) pathways and the functional groups annotated by CyanoBase (Figure 5c and Table S5). The result indicated that most of photosynthesis-related proteins were enriched. The majorities of PSI and phycobilisome were downregulated. Among all identified PSI subunits, PsaC (Ssl0563), PsaF (Sll0819), PsaK (Sll0629), PsaM (Smr0005) and Ycf3 (Slr0823) were significantly downregulated. Except for CpcB (Sll1577), CpcC2 (Sll1579), ApcE (Slr0335) and ApcF (Slr1459), the identified phycobilisome subunits were significantly downregulated. Unlike PSI and phycobilisome, subunits of PSII showed different behaviors. CP43 (Sll0851), PsbE (Ssr3451), Psb27 (Slr1645) and Psb28-2 (Slr1739) were significantly downregulated, whereas PsbP (Sll1418), PsbL (Smr0007), PsbY (Sml0007), PsbV (Sll0258) and Psb28 (Sll1398) were significantly upregulated (Figure 5a). Most of these changes were similar to previous transcriptomic data of Synechocystis [12][13][14]. Likewisely, most translational behaviors of identified genes in our assay were positively correlated with those in HL-treated Synechococcus sp. PCC 7002 [42]. Furthermore, our immune-detection also confirmed selective proteins (Figure 5b). Previous reports showed that some factors involved in PSII photorepair, such as FtsH2 (Slr0228) [16], EF-Tu (Sll1099) [19], Psb28 and Psb27 [43], would increase with elevated light intensity to cope with photoinhibition. Our proteomics also identified their significant upregulations in WT grown under HL (Figure 5a). Thus, our proteomic data could be considered as highly confident. Additionally and unexpectedly, the Slr0320 was almost unaltered (Figure 5a).  The cutoffs mentioned above were applied to filter proteins of WT and the ∆slr0320 mutant grown under NL and HL as well. Compared to WT under NL, 26 and 18 proteins were significantly down-and upregulated in the ∆slr0320 mutant under NL, respectively (Figure 6a and Table S3). Enrichment of those significantly upregulated proteins by Fisher's exact test revealed that Slr0320 had effects on PSII (Figure 6b and Table S6). One of the low molecular weight subunits of PSII [44], PsbL, was significantly upregulated (Figure 6a). Compared to WT under HL, there are 54 downregulated proteins and 28 upregulated proteins in the ∆slr0320 mutant under HL (Figure 6a and Table S4). Intriguingly, the significantly upregulated effect on PSII in the ∆slr0320 mutant grown under NL disappeared when grown under HL. For all identified PSII subunits, none was significantly differentially expressed (Figure 6a and Table S4). Likewisely, these proteomic data were further confirmed by our immune detections (Figure 5b).

Expression of Low Molecular Weight Subunits of PSII in WT, ∆slr0320 and PpetJ::slr0320 Strains
Besides PsbLthat the only significantly upregulated PSII subunit in the ∆slr0320 mutant was a low molecular weight (LMW) subunit, it is worth analyzing the expression of other LMW subunits. Of 13 LMW PSII subunits, our proteomic study identified four of them: PsbE, PsbF (Smr0006), PsbY (Sml0007) and PsbL. Except PsbL, the other three did not show significant changes in expression between WT and the ∆slr0320 mutant (Figure 6a and Tables S3 and S4). Many attempts to separate and immuno-detect these LMW subunits by SDS-PAGE and Western blot failed, probably due to their nature of low protein mass. Therefore, transcriptional levels of the remaining nine LMW proteins detected by quantitative real-time PCR (qRT-PCR) were analyzed (Figure 7). As shown in Figure 7, the expression patterns of psbH (ssl2598) (Figure 7a) and psbI (sml0001) (Figure 7b) in the ∆slr0320 mutant were highly sensitive to HL. The transcripts of these two genes were slightly upregulated in the ∆slr0320 mutant under NL, but markedly downregulated by more than five times compared to WT under HL. Compared to WT and the PpetJ::slr0320 strain grown under HL, the psbM (sml0003) (Figure 7c) and psbK (sml0005) (Figure 7f) genes in the ∆slr0320 mutant grown under HL were only slightly downregulated-i.e., by less than four times. Unlike the downregulations of psbH, psbI, psbK and psbM in the ∆slr0320 mutant grown under HL, the psbN (smr0009) (Figure 7d), psbJ (smr0008) (Figure 7e), psbT (smr0001) (Figure 7g), psbX (sml0002) (Figure 7h) and psbZ (sll1281) (Figure 7i) were upregulated in the ∆slr0320 mutant compared to WT and PpetJ::slr0320 strains grown under HL.

Slr0320 Is Crucial for Cyanobacteria during HL Acclimation
Slr0320 protein is conserved among sequenced cyanobacteria. Although Slr0320 was nonresponsive to HL in Synechocystis WT (Figure 5a), inactivation of slr0320 led to a significant defect in the photosynthetic growth under HL (Figure 1b,c). The growth difference between WT and the ∆slr0320 mutant under NL was not significant (Student's t test, p < 0.05), suggesting a more important role of Slr0320 under the HL condition. The fact that reconstitution of the slr0320 gene at a neutral site of the genome in the ∆slr0320 mutant restored the growth of the insertional mutant under HL further confirmed the growth phenotype by inactivation of slr0320 (Figure 1b,c). That the net photosynthesis rate measured in the presence of bicarbonate was significantly decreased (Student's t test, p < 0.05) in the ∆slr0320 mutant grown under HL ( Table 2) may explain the retarded growth. The standard condition commonly used in most laboratories for growing Synechocystis is under 50 µmol photons m −2 s −1 , whereas here we used NL conditions. Elevation of light intensity moderately to a sub-saturating condition increased the growth rate of WT due to more light energy. This HL condition, however, decreased the growth rate of the ∆slr0320 mutant (Figure 1b,c). These data suggest that moderately high light could not be used as additional light substrate for photosynthesis but was harmful to the cells when Slr0320 was missing. This phenomenon promoted us to figure out how Slr0320 affects photosynthesis during HL acclimation.

Slr0320 Is Involved in Optimal Function of PSII
Despite poor growth of the ∆slr0320 mutant under HL (Figure 1b,c), the accumulation of major photosynthetic proteins of thylakoid membranes had little change among three strains (Figure 2a). Similar to the protein amounts, impacts of the interruption of slr0320 on the activities of PSI and the Calvin-Benson cycle may not dramatically change. The measurements of P700 oxidation-reduction detected by absorbance changes in WT and the ∆slr0320 mutant showed that, illuminated by saturated far-red light (725 nm, 1400 µmol photons m −2 s −1 ), which solely activates PSI and in the presence of methyl viologen which accepts electrons directly from acceptor side of PSI [45], the oxidation and subsequent reduction traces of the ∆slr0320 mutant were very similar to those of WT grown under NL and HL ( Figure S1), indicating that the electron transfer within PSI was normal in the ∆slr0320 mutant. Previous reports showed that the activity of the Calvin-Benson cycle was positively correlated to protein amounts of RuBisCo, SBPase, aldolase and transketolase in plants and Synechocystis [46,47]. Our proteomic data show that these proteins were not significantly differentially regulated in the mutant (Tables S3 and S4), suggesting the activity of the Calvin-Benson cycle in the ∆slr0320 mutant could be similar to WT. In addition to the similar PSII accumulation (Figure 2a), the assembly of PSII complexes in the ∆slr0320 mutant was not affected either (Figure 2b). The HL acclimation responses of Synechocystis cause efficient PSII photorepair. Previous reports revealed that this process involves FtsH proteases [48], EF-Tu, EF-G [19], Slr1051 [21], CrtO [18], Psb28 and Psb27 [43] and the efficiency of PSII photorepair under increasing light intensity is strongly associated with the amounts of these proteins. Our proteomics of WT grown under NL and HL revealed that these proteins were significantly differentially regulated under HL, further supporting that the efficient PSII photorepair is closely associated with their amounts (Figure 5a). Our proteomic data of the ∆slr0320 mutant grown under NL and HL show that the amounts of these PSII photorepair related proteins were not significantly changed in the ∆slr0320 mutant (Figure 6a and Tables S3 and S4), indicating the photorepair and assembly of PSII complex under HL were unlikely the major reasons for the HL phenotype of the ∆slr0320 mutant. Chlorophyll fluorescence traces showed that the ∆slr0320 mutant had much higher stable fluorescence Fs than that of the WT but almost normal Fo (Figure 3a,b), suggesting higher portion of closed PSII centers under light in the mutant. These data suggest a defect in electron transfer processes at the acceptor side of PSII in the ∆slr0320 mutant. To confirm this, flash-induced chlorophyll fluorescence measurements were performed (Figure 3c).
Illuminations of Synechocystis by a single turn over flash could result in an increase in the chlorophyll fluorescence yield simultaneously [36]. The subsequent decay of fluorescence yield in the dark reflects three electron transfer pathways of Q A − reoxidation. First, the fast electron transfer from Q A − to Q B is accomplished by a plastoquinone molecule already bound to the Q B pocket at the moment of the flash. Second, the middle phase resembles the Q A − oxidized by free PQ from the PQ pool in the thylakoid membrane diffusing to the Q B pocket. Third, the slow phase resembles the Q A − reoxidized by the S 2 state of water-oxidizing complex. The fluorescence decay was much slower in the ∆slr0320 mutant, especially when the cells were grown under HL (Figure 3c and Table 3), indicating that, in the absence of Slr0320 protein, the Q A − reoxidation process is slowed down. The major reasons for the slow Q A − reoxidation were smaller amplitude of fast phase and bigger time constant of fast phase and middle phase. These data suggest that the acceptor side of PSII was modified in the ∆slr0320 mutant. Further oxygen evolution measurements in the presence of DCBQ (Table 2) suggest that, when enough quinone acceptor was provided, the PSII in the ∆slr0320 mutant were as functional as those in the WT. Therefore, we assumed that there was probably a defect in the quinone-binding of PSII due to the lack of Slr0320. Despite this, we identified effects of the interruption of slr0320 on the PSII acceptor side and the activities of PSI and the Calvin-Benson cycle were unchanged, so it would be unwise to exclude the possibility that Slr0320 still has an impact on other photosynthetic components downstream of PSII, such as cytochrome b 6 f. When the downstream of photosynthetic electron transport is inhibited, the PQ pool could stay reduced and thus the Fs increases [49], as shown in Figure 3a,b. The proteomics data and Western blots (Figures 5b and 6a; Tables S3 and S4) showed that the expression of subunits of the cytochrome b 6 f complex was not altered in the ∆slr0320 mutant under both NL and HL, suggesting that it may not the major reason for the obvious phenotype.
It has been reported that some low molecular weight (LMW) subunits of PSII may affect assembly and quinone-binding [6,8]. Indeed, one of those LMW subunits, PsbL, was found to be significantly upregulated in the ∆slr0320 mutant from our proteomic data (Figure 6a and Table S3). Previous reports have demonstrated that PsbL is a key factor for stabilization of PSII complexes and, by potential indirect interaction with D1 protein, influences PSII structure and the Q A activity [43,49,50]. A detailed investigation on tobacco PsbL revealed that additional PsbL protein could optimize the binding of plastoquinone with the PSII complex and subsequent electron transfer in vitro [50]. Given that PsbL is highly conserved between Synechocystis and plants [8], the role of PsbL could be similar in Synechocystis. The enhanced expression of PsbL in the ∆slr0320 mutant under NL may function in optimizing the binding of PQ to PSII and further partially restore the PSII electron transfer. This could be the reason that the impaired electron transfer from Q A to Q B in the ∆slr0320 mutant was less evident under NL than HL (Figure 3c and Table 3). Among 13 LMW subunits of Synechocystis, PsbE, PsbF, PsbL and PsbY were also detected but no significant difference was shown between WT and the ∆slr0320 mutant by proteomic analysis under HL (Figure 6a and Tables S3 and S4). Due to failure in the proteomic identification and difficulties in protein-level detection, we applied qRT-PCR to examine the expression of remaining nine LMW subunits. The data showed that psbH and psbI were downregulated by more than five times and could possibly have effects on their protein levels in the ∆slr0320 mutant under HL (Figure 7). Though psbM and psbK were downregulated as well, we thought their transcriptional downregulations might be minor. Due to the retarded photoautotrophic growth rates, highly sensitive to high light and slower electron transfer within PSII in psbH and psbI inactivation mutants, these two genes were considered to regulate electron transfer from Q A to Q B [51][52][53][54][55][56]. Furthermore, multiple reports of the crystal structure of PSII suggested that these two genes products were highly related to the acceptor-side structure within PSII [1][2][3]. Therefore, we thought that the much more evident impaired electron transfer from Q A to Q B caused by the interruption of slr0320 under HL was possibly due to decreases in psbH and psbI.
The B12-binding and the radical SAM domains of Slr0320 resemble the class B radical SAM methyltransferases. Previous reports have demonstrated that the radical SAM methyltransferases use SAM and other cofactors, such as B12 (cobalamin), to catalyze methylations. Members of the radical SAM methyltransferases were more than 113,000 and identified to methylate a wide variety of natural products, playing very important roles [57][58][59][60]. RlmE from various microorganisms, such as Escherichia coli [61,62] and Saccharomyces cerevisiae [63,64], and RlmN from Escherichia coli [65] revealed that the radical SAM methyltransferases were responsible for tRNA methylation and the subsequent ribosome assembly. The Fom3 from Streptomyces wedmorensis was identified to catalyze methylations in the biosynthesis of antibiotic fosfomycin [66]. In addition to the substrates mentioned above, many reports have shown that some members of this superfamily from various microorganisms, such as MenA and MenG from Escherichia coli and Streptomyces coelicolor, MenK from Adlercreutzia equolifaciens, Collinsella tanakaei and Ferrimonas marina, were involved in the methylation of menaquinone and composition of the PQ pool, and this adenosylmethionine-dependent methylation had important roles on the biosynthesis and function of quinone [67][68][69][70]. Considering that the impaired electron transfer from Q A to Q B (Figure 3c) might be due to the poor binding of PQ to PSII in the ∆slr0320 mutant, it cannot be excluded that the Slr0320 is a member of the radical SAM methyltransferases, catalyzing the methylation of quinone, and the interruption of slr0320 may result in defects of methylation of quinone, affecting its binding to PSII and impairing electron transfer between Q A and Q B . Yet, more proofs still need to be discovered.
Combining all the data above, a conclusion is made. Other than PSII photorepair and assembly, Slr0320 is involved in the electron transfer within PSII during HL acclimation of Synechocystis. The interruption of slr0320 leads to an impaired electron transfer from Q A to Q B within PSII. This impaired electron transfer could be partially restored by upregulation of PsbL under NL, possibly enhancing the binding of plastoquinone and consequently maintaining the mutant to grow similar to WT. However, the impaired electron transfer was further inhibited under HL, probably due to downregulations of psbH and psbI and defects in the structure of the acceptor side. Based on the conserved domains, there is a hint that Slr0320 might be involved in methylation of plastoquinone, further affecting the binding to PSII, and electron transfer from Q A to Q B cannot be excluded; more proofs are needed.
Supplementary Materials: The following are available online at https://www.mdpi.com/article/10 .3390/life11040279/s1, Figure S1: P700 oxidation-reduction in WT and the ∆slr0320 mutant grown under NL and HL; Figure S2: The mutant segregation by genomic PCR; Table S1: Reported ion intensities of quantitative proteomics; Table S2: Names of significantly differential expressed proteins involved in HL acclimation of Synechocystis; Table S3: Names of significantly differential expressed proteins of WT and ∆slr0320 grown under NL; Table S4: Names of significantly differential expressed proteins of WT and ∆slr0320 grown under HL; Table S5: All Fisher's exact test results of significantly differential expressed proteins involved in HL acclimation of Synechocystis; Table S6: All Fisher's exact test results of significantly differential expressed proteins of WT and ∆slr0320 grown under NL.
Author Contributions: P.Z. designed and supervised this work; P.Z. and Y.W. designed the quantitative proteomics which were performed by H.G.; H.Z. and H.G. analyzed the quantitative proteomics; H.Z. performed other experiments with some assistances of Y.Z.; H.Z. and P.Z. analyzed the data and wrote the manuscript. All authors have read and agreed to the published version of the manuscript.
Funding: This research was funded by Elite Youth program of Chinese Academy of Agricultural Sciences.

Data Availability Statement:
The reported ion intensities of quantitative proteomics presented in this study are available in Table S1.