Photosynthetic characterization of transgenic Synechocystis expressing a plant thiol/disulfide-modulating protein

ABSTRACT

A previous study showed that introducing an Arabidopsis thaliana thiol/disulfide-modulating protein, Low Quantum Yield of Photosystem II 1 (LQY1), into the cyanobacterium Synechocystis sp. PCC6803 increased the efficiency of Photosystem II (PSII) photochemistry. In the present study, the authors provided additional evidence for the role of AtLQY1 in improving PSII photochemical efficiency and cell growth. Light response curve analysis showed that AtLQY1-expressing Synechocystis grown at a moderate growth light intensity (50 µmol photons m⁻² s⁻¹) had higher minimal, maximal, and variable fluorescence than the empty-vector control, under a wide range of actinic light intensities. Light induction and dark recovery curves demonstrated that AtLQY1-expressing Synechocystis grown at the moderate growth light intensity had higher effective PSII quantum yield, higher photochemical quenching, lower regulated heat dissipation (non-photochemical quenching), low amounts of reduced plastoquinone, and higher amounts of oxidized plastoquinone than the empty-vector control. Furthermore, growth curve analysis indicated that AtLQY1-expressing Synechocystis grew faster than the empty-vector control at the moderate growth light intensity. These results suggest that transgenic expression of AtLQY1 in Synechocystis significantly improves PSII photochemical efficiency and overall cell growth.

Introduction

Thiol/disulfide-modulating proteins have been found to participate in a variety of chloroplastic processes, such as oxidative protein folding, translational regulation, and assembly and repair of protein complexes.^1-3 By definition, thiol/disulfide-modulating proteins contain multiple redox-active cysteine (Cys) residues. Examples of chloroplast-targeted thiol/disulfide-modulating proteins include thioredoxins (TRXs),^4-9 protein disulfide isomerases (PDIs),^10-15 lumen thiol oxidoreductases (LTOs),^16-18 rubredoxins (RBDs),^19,20 and C₄ (four Cys residues)-type zinc-finger proteins.^21-36 Example of chloroplast-targeted C₄-type zinc-finger proteins include thylakoid lumen-targeted Photosystem I Assembly factor 2 (PSA2),^31,33 stroma-localized Bundle Sheath Defective 2 (BSD2),^21,22,27,34 HCF222 (High Chlorophyll Fluorescence 222, dual-targeted to the ER and thylakoid membranes),³⁵ thylakoid membrane-associated CYO1/SCO2 (Shiyou 1/Snowy Cotyledon 2),^23,24,28,29 and thylakoid-membrane-anchored LQY1 (Low Quantum Yield of Photosystem II 1).^25,26,32,36

LQY1 contains an N-terminal transmembrane domain and two C₄-type zinc-finger domains, each with a CXXCX₁₉CXXC motif (i.e., two tandem CXXCXGXG motifs separated by 15 amino acids). LQY1 exists in land plants (e.g., AtLQY1 in Arabidopsis thaliana) but is not found in the sequenced genomes of aquatic algae and cyanobacteria.^25,26 Loss-of-function Atlqy1 mutants were more sensitive to light stress: they had lower maximum efficiency of PSII photochemistry (F_v/F_m), lower effective PSII quantum yield (Φ_PSII), lower PSII electron transport rate (ETR_PSII), fewer PSII-Light-Harvesting Complex II (PSII-LHCII), and higher non-photochemical quenching (NPQ) and reactive oxygen species (ROS) levels than the wild type after the high light treatment.^25,26 Furthermore, AtLQY1 was found to co-migrate with the PSII core monomer and the CP43-less PSII monomer (a marker for ongoing PSII repair and reassembly) in blue native-polyacrylamide gel electrophoresis gels and co-immunoprecipitate with Cys-containing PSII core subunits CP47 and C43.²⁶ Therefore, it was proposed that LQY1 may participate in PSII repair and reassembly by forming transient disulfide bonds with Cys-containing PSII subunits and regulate redox homeostasis (e.g., reducing ROS accumulation) in land plants.³ To test whether transgenic expression of LQY1 will be beneficial to photosynthetic organisms that lack LQY1, we introduced AtLQY1 into the cyanobacterium Synechocystis sp. PCC6803 and performed a suite of physiological and biochemical assays.³⁶ When grown at a moderate growth light intensity (50 µmol photons m⁻² s⁻¹), AtLQY1-expressing Synechocystis had higher F_v/F_m, Φ_PSII, and ETR_PSII, but lower NPQ and ROS levels than the empty-vector control.³⁶ In this work, we performed additional physiological assays on AtLQY1-expressing Synechocystis and provided additional evidence for the role of AtLQY1 in improving PSII photochemical efficiency and overall cell growth.

Materials and methods

Cyanobacterial strains and growth conditions

Synechocystis sp. PCC6803 cultures were transformed with pSL2035-AtLQY1¹⁻¹⁵⁴ (i.e., full-length AtLQY1) and the empty expression vector pSL2035 as described previously.³⁶ Successfully transformed Synechocystis cultures were grown in BG-11 liquid or solid medium supplemented with 25 µg/mL kanamycin.^37-39 Thirty-mL liquid cultures were grown at 28°C in 125-mL Erlenmeyer flasks with a culture depth of 1 cm on a VWR mini shaker set at 140 rpm in a Percival growth chamber. Growth light intensity was set to 25 or 50 µmol photons m⁻² s⁻¹ at the surface of the flasks. Solid cultures were grown at similar growth conditions. Liquid cultures were collected at the mid-log phase (OD₇₃₀ = 0.50 to 0.70) for downstream assays unless otherwise stated.

Chlorophyll fluorescence analysis of light response curves

Light response curves of minimal fluorescence (F_o or F_o′ for dark-adapted or illuminated cultures, respectively), maximal fluorescence (F_m or F_m′ for dark-adapted or illuminated cultures), and variable fluorescence (F_v or F_v′ for dark-adapted or illuminated cultures) in Synechocystis cultures were measured with the DUAL-PAM-100 measuring system (Walz, Germany) as described previously.³⁶ Synechocystis cultures were harvested at an OD₇₃₀ of ~0.7, dark adapted for 5 min, then were resuspended and exposed to a saturation pulse (2,000 μmol photons m⁻² s⁻¹) to determine F_o and F_m. Synechocystis cultures were then illuminated for 30 sec at the following actinic light intensities: 0, 8, 13, 20, 46, 82, 105, 161, 236, and 422 µmol photons m⁻² s⁻¹. A saturation pulse was applied at the end of each 30-s illumination to determine F₀′ and F_m′ of illuminated cultures. F_v and F_v′ were calculated using the following equations: F_v = F_m – F_o; F_v′ = F_m′ – F_o′.

Chlorophyll fluorescence analysis of light induction and dark recovery curves

Light induction and dark recovery curves of chlorophyll fluorescence parameters were measured with the DUAL-PAM-100 measuring system (Waltz) as described previously.³⁶ Synechocystis cultures were harvested at an OD₇₃₀ of ~0.7, dark adapted for 5 min, resuspended, and exposed to a saturation pulse (2,000 μmol photons m⁻² s⁻¹) to determine F_o and F_m. Synechocystis cultures were then pre-illuminated under a measuring light of 2 μmol photons m⁻² s⁻¹ for 3 min, with saturating pulses at 30-s intervals. After the 3-min pre-illumination, blue actinic light (422 μmol photons m⁻² s⁻¹) was applied for 8 min with saturating pulses at 20-s intervals. After the actinic light was turned off, fluorescence parameters were continued to be monitored under a measuring light of 2 μmol photons m⁻² s⁻¹ for 18 min, with exponentially increasing intervals between saturating pulses. NPQ was calculated using the following equation: NPQ = (F_m – F_m’)/F_m’. Coefficient of non-photochemical quenching (qN), coefficient of photochemical quenching (qP), and the redox state of the PSII acceptor side (1-qP) were calculated using the following equations: qN = (F_m – F_m′)/(F_m – F₀′); qP = (F_m′ – F)/(F_m′ – F₀′); 1-qP = 1 – (F_m′ – F)/(F_m′ – F₀′), where F is fluorescence yield. Effective PSII quantum yield (Φ_PSII), quantum yield of regulated energy dissipation (Φ_NPQ), and quantum yield of non-regulated energy dissipation (Φ_NO) were calculated using the following equations: Φ_PSII = (F_m′ – F)/F_m′; Φ_NPQ = 1 – Φ_PSII – 1/(NPQ + 1 + qP * F₀′ * (F_m/F_o – 1)/F); Φ_NO = 1/(NPQ + 1 + qP * F₀′ * (F_m/F_o – 1)/F).

Growth curve analysis

Growth curve analysis of Synechocystis cultures was performed as described previously,^38,40,41 with minor modifications. Starter cultures were grown to the mid-log phase (OD₇₃₀ = 0.50 to 0.70) and then were diluted to an OD₇₃₀ equal to 0.20, with kanamycin-containing BG-11 liquid medium. Diluted starter cultures were used to inoculate cultures for growth curve analysis (four biological replicates per genotype, OD₇₃₀ = 0.05). The optical densities of the cultures for growth curve analysis were measured with a BioMate 3S spectrophotometer (Thermo Fisher) every 24 h for 14 days. Cultures for extended growth analysis were imaged after 24, 28, and 32 days of growth.

Results

AtLQY1-expressing Synechocystis grown at 50 µmol photons m⁻² s⁻¹ had higher F_o′, F_m′, and F_v′ than the empty-vector control at a wide range of actinic light intensities

We previously showed that when grown at 50 µmol photons m⁻² s⁻¹, dark-adapted AtLQY1-expressing Synechocystis had higher F_o, F_m, and F_v (minimal, maximal, and variable fluorescence of dark-adapted cultures) than the empty-vector control.³⁶ To test whether this observation holds true in illuminated cultures, we determined F_o′, F_m′, and F_v′ (minimal, maximal, and variable fluorescence of illuminated cultures) at a wide range of actinic light intensities (0, 8, 13, 20, 46, 82, 105, 161, 236, and 422 µmol photons m⁻² s⁻¹) (Figure 1). In cultures grown at 25 or 50 µmol photons m⁻² s⁻¹, the maximal values of F_o′, F_m′, and F_v′ were obtained at an actinic light intensity of 13 µmol photons m⁻² s⁻¹ (Figure 1a–f). As the actinic light intensity increased from 13 to 422 µmol photons m⁻² s⁻¹, F_o′, F_m′, and F_v′ values decreased (Figure 1a–f). Such concave actinic light dependence of fluorescence signals has been previously observed in Synechocystis.⁴² AtLQY1-expressing Synechocystis and the empty-vector control grown at 25 µmol photons m⁻² s⁻¹ had statistically similar F_o′, F_m′, and F_v′ values (p > 0.05) at each actinic light intensity (Figure 1a–c). However, AtLQY1-expressing Synechocystis grown at 50 µmol photons m⁻² s⁻¹ had significantly higher F_o′, F_m′, and F_v′ (p < 0.001) than the empty-vector control, at each actinic light intensity (Figure 1d–f). Cyanobacterial F_v arises principally from PSII; therefore, this observation reveals that PSII in AtLQY1-expressing Synechocystis grown at 50 µmol photons m⁻² s⁻¹ has a higher ability to perform primary photochemistry than PSII in the empty-vector control.^43,44

Figure 1.

Light response curves of F₀, F_m, and F_v. (a-c) Light response curves of F₀, F_m, and F_v in the empty-vector control (white squares) and AtLQY1-expressing Synechocystis (black diamonds) grown at 25 µmol photons m⁻² s⁻¹. (d-f) Light response curves of F₀, F_m, and F_v in the empty-vector control and AtLQY1-expressing Synechocystis grown at 50 µmol photons m⁻² s⁻¹. Initial cultures were started at optical density of 0.05 at 730 nm. The light response curves were determined on mid-log phase cultures after a 5-min dark adaption. During the measurement, cultures were exposed for 30 seconds at a wide range of actinic light intensities (0, 8, 13, 20, 46, 82, 105, 161, 236, and 422 µmol photons m⁻² s⁻¹). Data are presented as means ± SE (n = 4 independent biological replicates). Asterisks indicate significant differences between AtLQY1-expressing Synechocystis and the empty-vector control (Student’s t-test; *, p < 0.05; **, p < 0.01; ***, p < 0.001).

AtLQY1-expressing Synechocystis grown at 50 µmol photons m⁻² s⁻¹ had higher F_o′, F_m′, and F_v′ than the empty-vector control during light induction and dark recovery

We also monitored changes in minimal, maximal, and variable fluorescence during light induction and dark recovery (Figure 2). In cultures grown at 25 or 50 µmol photons m⁻² s⁻¹, the characteristic increases of F_o′, F_m′, and F_v′ signals were observed at the onset of blue actinic illumination (422 µmol photons m⁻² s⁻¹) (Figure 2a–f). Such characteristic rises of F_o′ and F_m′ signals are indicative of a transition from State 2 to State 1 (i.e., redistribution of phycobilisome-absorbed excitation energy from PSII to PSI).^42,45 AtLQY1-expressing Synechocystis and the empty-vector control grown at 25 µmol photons m⁻² s⁻¹ had statistically similar F_o′, F_m′, and F_v′ (p > 0.05) during the light induction stage and statistically similar F_o, F_m, and F_v (p > 0.05) during the recovery stage (Figure 2a–c). However, AtLQY1-expressing Synechocystis grown at 50 μmol photons m⁻² s⁻¹ had significantly higher F_o′, F_m′, and F_v′ (p < 0.001) than the empty-vector control during the light induction stage, and significantly higher F_o, F_m, and F_v (p < 0.001) than the empty-vector control at the recovery stage (Figure 2d–f). Taken together, the kinetics of minimal, maximal, and variable fluorescence in AtLQY1-expressing Synechocystis during the light induction and dark recovery measurements (Figure 2) is consistent with those during light response curve measurements (Figure 1).

Figure 2.

F₀, F_m, and F_v during light induction and dark recovery. (a–f) F₀, F_m, and F_v during NPQ induction and relaxation in the empty-vector control (white squares) and AtLQY1-expressing Synechocystis (black diamonds) grown at 25 (a–c) and 50 (d–f) µmol photons m⁻² s⁻¹. Cultures were harvested at an OD₇₃₀ of ~0.7. After a 5-min dark adaption, cultures were resuspended, and a saturation pulse was applied to determine F₀ and F_m. Cultures were pre-illuminated under a measuring light of 2 μmol photons m⁻² s⁻¹ for 3 min, with saturating pulses at 30-sec intervals. After the pre-illumination, blue actinic light (422 μmol photons m⁻² s⁻¹) was applied for 8 min with saturating pulses at 20-sec intervals. Relaxation was monitored under a measuring light of 2 μmol photons m⁻² s⁻¹ for 18 min, with exponentially increasing intervals between saturating pulses. Data are presented as means ± SE (n = 4 independent biological replicates).

AtLQY1-expressing Synechocystis grown at 50 µmol photons m⁻² s⁻¹ had higher Φ_PSII and lower NPQ than the empty-vector control during light induction and dark recovery

We previously showed that AtLQY1-expressing Synechocystis grown at 50 µmol photons m⁻² s⁻¹ had higher Φ_PSII (effective PSII quantum yield) during light response curve measurements and lower steady-state NPQ (non-photochemical quenching) than the empty-vector control.³⁶ In this study, we monitored changes in Φ_PSII, Φ_NO (quantum yield of non-regulated energy dissipation), Φ_NPQ (quantum yield of regulated energy dissipation), and NPQ during light induction and dark recovery (Figure 3). At the beginning of the light induction stage (i.e., during the first 304 sec of blue actinic illumination), AtLQY1-expressing Synechocystis and the empty-vector control grown at 25 µmol photons m⁻² s⁻¹ had statistically similar Φ_PSII (p > 0.05) (Figure 3a). As the blue actinic illumination continued (i.e., after the first 304 sec of blue actinic illumination), AtLQY1-expressing Synechocystis grown at 25 µmol photons m⁻² s⁻¹ began to have statistically higher Φ_PSII (p < 0.05) than the empty-vector control (Figure 3a). These data demonstrate that AtLQY1-expressing Synechocystis grown at 25 µmol photons m⁻² s⁻¹ may have higher PSII operating efficiency (i.e., effective PSII quantum yield) soon after transition to high light intensities (e.g., blue actinic illumination at 422 µmol photons m⁻² s⁻¹), when compared to the empty-vector control. AtLQY1-expressing Synechocystis and the empty-vector control grown at 25 µmol photons m⁻² s⁻¹ had statistically similar Φ_NO (p > 0.05) throughout the light induction and dark recovery stages (Figure 3b). This suggests that non-regulated heat dissipation is similar in AtLQY1-expressing Synechocystis and the empty-vector control grown at 25 µmol photons m⁻² s⁻¹. AtLQY1-expressing Synechocystis and the empty-vector control grown at 25 µmol photons m⁻² s⁻¹ had statistically similar Φ_NPQ and NPQ (p > 0.05) throughout the light induction and dark recovery stages (Figure 3c-d). This suggests that regulated heat dissipation (i.e., NPQ) is similar in AtLQY1-expressing Synechocystis and the empty-vector control grown at 25 µmol photons m⁻² s⁻¹.

Figure 3.

Φ_PSII, Φ_NO, Φ_NPQ, and NPQ during light induction and dark recovery. (a–h) Φ_PSII, Φ_NO, Φ_NPQ, and NPQ during NPQ induction and relaxation in the empty-vector control (white squares) and AtLQY1-expressing Synechocystis (black diamonds) grown at 25 (a–d) and 50 (e–h) µmol photons m⁻² s⁻¹. Cultures were harvested at an OD₇₃₀ of ~0.7. After a 5-min dark adaption, cultures were resuspended, and a saturation pulse was applied to determine F_o and F_m. Cultures were pre-illuminated under a measuring light of 2 μmol photons m⁻² s⁻¹ for 3 min, with saturating pulses at 30-s intervals. After the pre-illumination, blue actinic light (422 μmol photons m⁻² s⁻¹) was applied for 8 min with saturating pulses at 20-sec intervals. Relaxation was monitored under a measuring light of 2 μmol photons m⁻² s⁻¹ for 18 min, with exponentially increasing intervals between saturating pulses. Data are presented as means ± SE (n = 4 independent biological replicates).

AtLQY1-expressing Synechocystis grown at 50 μmol photons m⁻² s⁻¹ had significantly higher Φ_PSII (p < 0.001) than the empty-vector control during both light induction and dark recovery stages (Figure 3e). This observation is consistent with the higher Φ_PSII value in AtLQY1-expressing Synechocystis grown at 50 μmol photons m⁻² s⁻¹ during light response curve measurements.³⁶ AtLQY1-expressing Synechocystis and the empty-vector control grown at 50 μmol photons m⁻² s⁻¹ had statistically similar Φ_NO (p > 0.05) during the light induction stage. However, AtLQY1-expressing Synechocystis grown at 50 μmol photons m⁻² s⁻¹ had significantly higher Φ_NO (p < 0.01 at most data points) than the empty-vector control during the recovery stage (Figure 3f). The physiological implication of these observations is not clear. AtLQY1-expressing Synechocystis grown at 50 μmol photons m⁻² s⁻¹ had significantly lower Φ_NPQ and NPQ (p < 0.001) than the empty-vector control during both light induction and dark recovery stages (Figure 3g–h). By definition, the sum of Φ_PSII, Φ_NO, and Φ_NPQ is 1. The combination of higher Φ_PSII and lower Φ_NPQ indicates that AtLQY1-expressing Synechocystis grown at 50 μmol photons m⁻² s⁻¹ devotes a higher percentage of excitation energy into photochemistry and a lower percentage of excitation energy into regulated heat dissipation, than the empty-vector control.

AtLQY1-expressing Synechocystis grown at 50 µmol photons m⁻² s⁻¹ had higher qP and lower 1-qP than the empty-vector control during light induction and dark recovery

We also monitored changes in qN (coefficient of non-photochemical quenching), qP (coefficient of photochemical quenching), and 1-qP (the redox state of the PSII acceptor side) during light induction and dark recovery (Figure 4). AtLQY1-expressing Synechocystis and the empty-vector control grown at 25 µmol photons m⁻² s⁻¹ had statistically similar qN (p > 0.05) throughout the light induction and dark recovery stages (Figure 4a). This trend is in line with the trends of Φ_NPQ and NPQ in Synechocystis grown at 25 µmol photons m⁻² s⁻¹ (Figure 3c-d); all of them indicate that AtLQY1-expressing Synechocystis and the empty-vector control have similar NPQ when grown at 25 µmol photons m⁻² s⁻¹.

Figure 4.

qN, qP, and 1-qP during light induction and dark recovery. (a–f) qN, qP, and 1-qP during NPQ induction and relaxation in the empty-vector control (white squares) and AtLQY1-expressing Synechocystis (black diamonds) grown at 25 (a–c) and 50 (d–f) µmol photons m⁻² s⁻¹. Cultures were harvested at an OD₇₃₀ of ~0.7. After a 5-min dark adaption, cultures were resuspended, and a saturation pulse was applied to determine F_o and F_m. Cultures were pre-illuminated under a measuring light of 2 μmol photons m⁻² s⁻¹ for 3 min, with saturating pulses at 30-sec intervals. After the pre-illumination, blue actinic light (422 μmol photons m⁻² s⁻¹) was applied for 8 min with saturating pulses at 20-s intervals. Relaxation was monitored under a measuring light of 2 μmol photons m⁻² s⁻¹ for 18 min, with exponentially increasing intervals between saturating pulses. Data are presented as means ± SE (n = 4 independent biological replicates).

The light induction and dark recovery curve of qP (Figure 4b) is very much similar to that of Φ_PSII (Figure 3a). At the beginning of the light induction stage (i.e., during the first 266 sec of blue actinic illumination), AtLQY1-expressing Synechocystis and the empty-vector control grown at 25 µmol photons m⁻² s⁻¹ had statistically similar qP (p > 0.05) (Figure 4b). As the blue actinic illumination continued (i.e., after the first 266 sec of blue actinic illumination), AtLQY1-expressing Synechocystis grown at 25 µmol photons m⁻² s⁻¹ started to have statistically higher qP (p < 0.05) than the empty-vector control (Figure 4b). During the recovery stage, AtLQY1-expressing Synechocystis grown at 25 µmol photons m⁻² s⁻¹ continued to have higher qP (p < 0.05 at most data points) than the empty-vector control (Figure 4b). These results demonstrate that AtLQY1-expressing Synechocystis grown at 25 µmol photons m⁻² s⁻¹ may have higher photochemical quenching than the empty-vector control, soon after transition to high light intensities (e.g., blue actinic illumination at 422 µmol photons m⁻² s⁻¹).

qP and 1-qP are inversely related. Consequently, the light induction and dark recovery curves of 1-qP displayed an opposite trend as qP. At the beginning of the light induction stage (i.e., during the first 266 sec of blue actinic illumination), AtLQY1-expressing Synechocystis and the empty-vector control grown at 25 µmol photons m⁻² s⁻¹ had statistically similar 1-qP (p > 0.05) (Figure 4c). As the blue actinic illumination continued (i.e., after the first 266 sec of blue actinic illumination), AtLQY1-expressing Synechocystis grown at 25 µmol photons m⁻² s⁻¹ started to have statistically lower 1-qP (p < 0.05) than the empty-vector control (Figure 4c). During the recovery stage, AtLQY1-expressing Synechocystis grown at 25 µmol photons m⁻² s⁻¹ continued to have lower 1-qP (p < 0.05 at most data points) than the empty-vector control (Figure 4c). These observations suggest that AtLQY1-expressing Synechocystis grown at 25 µmol photons m⁻² s⁻¹ may have a larger pool of oxidized plastoquinone (i.e., a smaller pool of reduced plastoquinone) than the empty-vector control soon after transition to high light intensities.

In line with the trends of Φ_NPQ and NPQ, AtLQY1-expressing Synechocystis grown at 50 μmol photons m⁻² s⁻¹ had significantly lower qN (p < 0.001) than the empty-vector control during both light induction and dark recovery stages (Figure 4d). These results consistently reveal that AtLQY1-expressing Synechocystis has lower non-photochemical quenching than the empty-vector control, when grown at 50 µmol photons m⁻² s⁻¹. AtLQY1-expressing Synechocystis grown at 50 μmol photons m⁻² s⁻¹ had significantly higher qP (p < 0.001 at most data points) than the empty-vector control during both light induction and dark recovery stages (Figure 4e). This observation shows that AtLQY1-expressing Synechocystis has higher photochemical quenching than the empty-vector control, when grown at 50 µmol photons m⁻² s⁻¹. As mentioned previously, qP and 1-qP are inversely related. Consequently, AtLQY1-expressing Synechocystis grown at 50 μmol photons m⁻² s⁻¹ had significantly lower 1-qP (p < 0.001 at most data points) than the empty-vector control during both light induction and dark recovery stages (Figure 4f). These data suggest that, when grown at 50 µmol photons m⁻² s⁻¹, AtLQY1-expressing Synechocystis may have a larger pool of oxidized plastoquinone (i.e., a smaller pool of reduced plastoquinone) than the empty-vector control.

AtLQY1-expressing Synechocystis grew faster than the empty-vector control at 50 µmol photons m⁻² s⁻¹

To investigate whether AtLQY1 expression in Synechocystis improves growth, we performed growth curve analysis by monitoring OD₇₃₀ every 24 h for 14 days (Figure 5). We plotted the growth curves on a Log₂ scale (Figure 5a–b), so that the growth phases are more obvious. At a growth light of 25 µmol photons m⁻² s⁻¹, the empty-vector control and AtLQY1-expressing Synechocystis had no statistical difference in OD₇₃₀ at all the time points tested (Figure 5a). At a growth light of 50 µmol photons m⁻² s⁻¹, OD₇₃₀ in AtLQY1-expressing Synechocystis was significantly higher (p < 0.05) than that in the empty-vector control at multiple time points, especially after the mid-log phase (Figure 5b). For example, after 336 h (i.e., 14 days) of growth, OD₇₃₀ in AtLQY1-expressing Synechocystis was 43% higher than that in the empty-vector control. These data suggest that AtLQY1-expressing Synechocystis grows faster than the empty-vector control at 50 µmol photons m⁻² s⁻¹.

Figure 5.

Growth analysis. (a-b) Growth curves of the empty-vector control (white squares) and AtLQY1-expressing Synechocystis (black diamonds) at 25 (a) or 50 (b) µmol photons m⁻² s⁻¹ on a log₂ scale. Initial cultures were started at optical density of 0.05 at 730 nm. The optical density of the cultures was measured at 730 nm every 24 hours for 15 days. Data are presented as mean ± SE (n = 4). The asterisks indicate significant differences between AtLQY1-expressing Synechocystis and the empty-vector control (Student’s t-test; *, p < 0.05; **, p < 0.01). (c) Images of AtLQY1-expressing Synechocystis and the empty-vector control cultures grown at 50 µmol photons m⁻² s⁻¹ for 24, 28, and 32 days. Initial cultures were started at optical density of 0.05 at 730 nm.

AtLQY1-expressing Synechocystis grown at 50 µmol photons m⁻² s⁻¹ showed delayed yellowing after extended growth

After extended growth at 50 µmol photons m⁻² s⁻¹ for 24, 28, and 32 days, AtLQY1-expressing Synechocystis stayed green, while the empty-vector control progressively turned yellow (Figure 5c). These observations indicate that compared to the empty-vector control, AtLQY1-expressing Synechocystis grown at 50 µmol photons m⁻² s⁻¹ may have delayed discoloration during extended growth.

Discussion

Taken together, chlorophyll fluorescence analysis of light response curves and light induction and dark recovery curves as well as cell growth curve analysis consistently showed that transgenic expression of AtLQY1, a thiol/disulfide-modulating protein from the higher plant A. thaliana, significantly improved PSII photochemical efficiency and overall cell growth in the cyanobacterium Synechocystis. Light response curves and light induction and dark recovery curves both showed that AtLQY1-expressing Synechocystis grown at a moderate growth light intensity (50 µmol photons m⁻² s⁻¹) had higher minimal, maximal, and variable fluorescence than the empty-vector control (Figures 1 and 2). Light induction and dark recovery curves also demonstrated that AtLQY1-expressing Synechocystis grown at 50 µmol photons m⁻² s⁻¹ had higher effective PSII quantum yield, higher photochemical quenching, lower regulated heat dissipation (i.e., non-photochemical quenching), low amounts of reduced plastoquinone, and higher amounts of oxidized plastoquinone than the empty-vector control (Figures 3 and 4). The decreases in reduced plastoquinone and increases in oxidized plastoquinone (Figure 4)^46,47 are consistent with the higher PSII electron transport efficiency in AtLQY1-expressing Synechocystis grown at 50 µmol photons m⁻² s⁻¹. Furthermore, growth curve analysis showed that compared to the empty-vector control, AtLQY1-expressing Synechocystis grew faster at 50 µmol photons m⁻² s⁻¹ (Figure 5a–b). These results suggest that transgenic expression of AtLQY1 in Synechocystis improves PSII photochemical efficiency and overall cell growth.

Cyanobacteria and land plants have similar PSII core subunits but differ in the composition of light harvesting complexes (LHCs), the composition of oxygenic evolving complexes, and the mechanisms of NPQ.^48-50 Similar to their counterparts in land plants, a number of PSII core subunits in cyanobacteria contain Cys residues.²³ Phycobiliproteins, which constitute phycobilisomes (PSII LHCs in cyanobacteria) also contain Cys residues.⁵¹ In phycobilisomes, chromophores are covalently attached to phycobiliproteins, via thioether bonds to Cys residues.⁵² Furthermore, the orange carotenoid protein (OCP), which mediates NPQ in cyanobacteria, contains conserved Cys residues.^53,54 Therefore, AtLQY1 may benefit photosynthesis in Synechocystis by: (1) taking part in the folding, assembly, and repair of Cys-containing PSII subunits; (2) regulating phycobilisome assembly and associations between phycobiliproteins and their chromophores; and (3) reducing NPQ by modulating the redox status of Cys residues in OCP.³⁶ Additional studies are needed to identify the causal mechanism(s).

Loss-of-function mutants of many chloroplastic thiol/disulfide-modulating proteins with roles in the assembly and repair of photosynthetic complexes have a phenotype opposite from AtLQY1-expressing Synechocystis. TRX-Ms were proposed to assist the assembly of CP47 into PSII.⁹ Triple silencing of TRX-M1, TRX-M2, and TRX-M4 genes in A. thaliana resulted in reduced PSII stability, decreased F_v/F_m and Φ_PSII values, higher NPQ and 1-qP, and increased ROS levels.⁹ AtLTO1 (A. thaliana LTO1) was proposed to participate in PSII assembly, redox regulation, and ROS homeostasis.^17,18 The loss-of-function mutation in the AtLTO1 gene caused reduced PSII assembly and F_v/F_m, higher NPQ, and increased photoinhibition and ROS accumulation in A. thaliana.¹⁷ RBD1 was proposed to function in PSII assembly and repair.^19,20 Loss-of-function rbd1 mutants of Synechocystis, C. reinhardtii, and A. thaliana had substantially reduced PSII accumulation and F_v/F_m.¹⁹ PSA2 was proposed to participate in PSI biogenesis.³¹ Loss-of-function mutations in the PSA2 gene resulted in reduced PSI activity, lower F_v/F_m, and abnormal thylakoid assembly in A. thaliana and maize.^31,33 Lastly, LQY1 was proposed to participate in PSII repair and regulate redox homeostasis.³ Loss-of-function mutations in the AtLQY1 gene resulted in lower F_v/F_m, Φ_PSII, and ETR_PSII and higher NPQ and ROS levels in A. thaliana, especially under high light conditions.^25,26 These studies collectively demonstrated that chloroplastic thiol/disulfide-modulating proteins are involved in the assembly and repair of photosynthetic complexes and that manipulating the expression of these chloroplastic thiol/disulfide-modulating proteins may result in changes in photosynthetic efficiency and overall fitness.

The functions of these thiol/disulfide-modulating proteins are consistent with their locations in the cell. BSD2 is located in the chloroplast stroma and was found to participate in post-translational assembly of rubisco, a soluble enzyme in the chloroplast stroma.^21,22,27,34 TRX-M1, TRX-M2, and TRX-M4 are present in the chloroplast stroma and could also be associated to stroma-exposed thylakoid membranes.^8,55,56 These three TRX-Ms are capable of activating NADH-malate dehydrogenase in the chloroplast stroma,^4,5 and assisting the assembly of PSII core complexes in thylakoid membranes.⁹ Both CrPDI1/RB60 and AtPDI6/PDIL1-2 are triple-targeted to the chloroplast stroma, thylakoid membranes, and the ER.^10,12,13 Chloroplastic CrPDI1/RB60 and AtPDI6/PDIL1-2 were found to regulate the synthesis of D1 protein in thylakoid membranes.^10-13,15 ER-targeted CrPDI1/RB60 was proposed to catalyze disulfide bond formation/isomerization of ER-imported proteins.¹³ HCF222 is dual-targeted to the ER and chloroplast thylakoid membranes, and chloroplastic HCF222 was found to function in the maturation and/or assembly of the cytochrome b₆f complex in thylakoid membranes.³⁵ LTO1 is targeted to thylakoid membranes with the two redox-active CXXC motifs exposed to the thylakoid lumen.⁵⁷ In line with this topology, AtLTO1 was found to regulate PSII assembly, redox status, and ROS homeostasis, by catalyzing disulfide bond formation in lumenal proteins and lumen-exposed proteins.^16-18,58 RBD1 and LQY1 are targeted to thylakoid membranes and were found to function in assembly and repair of PSII in thylakoid membranes.^{19,20,25,26,32} CYO1/SCO2 is targeted to thylakoid membranes in cotyledons and was found to participate in chloroplast and thylakoid biogenesis, as well as vesicular transport of photosynthetic proteins from the inner chloroplast envelope to developing thylakoids in cotyledons.^23,24,28,29 PSA2 is located in the thylakoid lumen and was found to participate in PSI biogenesis, via interaction with the PsaG-containing complex in the thylakoid lumen.^31,33

As discussed above, chloroplast-targeted thiol/disulfide-modulating proteins participate in essential chloroplast processes, such as (1) synthesis, folding, and transport of photosynthetic proteins; (2) assembly, stability, and repair of photosynthetic complexes; (3) activation of target proteins; (4) redox regulation; and (5) chloroplast and thylakoid biogenesis. Loss-of-function mutations in these genes often resulted in reduced photosynthetic efficiency and light tolerance, and increased photoinhibition and ROS accumulation. On the contrary, transgenic expression of AtLQY1, an A. thaliana thiol/disulfide-modulating protein, in the cyanobacterium Synechocystis sp. PCC6803 resulted in improvements in photosynthetic efficiency, light tolerance, and cell growth. Among these thiol/disulfide-modulating proteins, TRX-Ms, LTOs, and RBDs exist in oxygenic photosynthetic organisms, including cyanobacteria, algae, and land plants. PDIs and PSA2 exist in photosynthetic eukaryotes (algae and land plants), while BSD2, HCF222, CYO1/SCO2, and LQY1 exist in land plants only. The evolution of these land plant-specific thiol/disulfide-modulating proteins may reflect plant adaption to new terrestrial environments, the sessile life style, and excess light, during transition from water to land. Cyanobacteria and microalgae have recently become great potential manufacturers of biofuels.^59-64 Therefore, transgenic expression of land plant-specific thiol/disulfide-modulating proteins might be a strategy to improve photosynthetic efficiency and cell growth in biofuel producers such as cyanobacteria and algae.

Funding Statement

This work was financially supported by the U.S. National Science Foundation [grant number MCB-1244008] and the Western Michigan University Faculty Research and Creative Activities [Award number W2016-023].

Acknowledgments

The authors thank Himadri B. Pakrasi, Yinjie J. Tang, and Arul M. Varman at the Washington University for sharing the pSL2035 vector; Beronda Montgomery at Michigan State University for sharing the wild-type Synechocystis sp. PCC6803 strain and comments on TEM images; Alicia Withrow at Michigan State University for TEM sample preparation and imaging; James P. O’Donnell, Manasa B. Satyanarayan, Amy T. Kobylarz, and Sianoush B. Fereidani for technical assistance; Christopher D. Jackson for growth chamber management; and Todd J. Barkman, Jian Yao, and Silvia Rossbach for comments on experimental design.

Disclosure of Potential Conflicts of Interest

No potential conflicts of interest were disclosed.