NTRC regulates CP12 to activate Calvin–Benson cycle during cold acclimation

Significance

While roles of NTRC (NADPH-dependent thioredoxin reductase C) and CP12 (Calvin–Benson Cycle Protein 12) in cold acclimation have been proposed previously, the underlying molecular mechanisms have not been fully understood. In this report, we demonstrate a previously unknown interaction between these two proteins leading to redox regulation of the Calvin–Benson cycle under cold stress. We identified NTRC as a previously undescribed redox regulator of CP12, triggering the dissociation/rearrangement of the autoinhibitory PRK/CP12/GAPDH (phosphoribulokinase/CP12/glyceraldehyde-3-phosphate dehydrogenase) supracomplex during cold acclimation. Despite minor differences in the molecular working mode in Chlamydomonas reinhardtii and Arabidopsis thaliana, the results should apply broadly to two major groups of photoautotrophic eukaryotes, i.e., green algae and land plants.

Abstract

NADPH-dependent thioredoxin reductase C (NTRC) is a chloroplast redox regulator in algae and plants. Here, we used site-specific mutation analyses of the thioredoxin domain active site of NTRC in the green alga Chlamydomonas reinhardtii to show that NTRC mediates cold tolerance in a redox-dependent manner. By means of coimmunoprecipitation and mass spectrometry, a redox- and cold-dependent binding of the Calvin–Benson Cycle Protein 12 (CP12) to NTRC was identified. NTRC was subsequently demonstrated to directly reduce CP12 of C. reinhardtii as well as that of the vascular plant Arabidopsis thaliana in vitro. As a scaffold protein, CP12 joins the Calvin–Benson cycle enzymes phosphoribulokinase (PRK) and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) to form an autoinhibitory supracomplex. Using size-exclusion chromatography, NTRC from both organisms was shown to control the integrity of this complex in vitro and thereby PRK and GAPDH activities in the cold. Thus, NTRC apparently reduces CP12, hence triggering the dissociation of the PRK/CP12/GAPDH complex in the cold. Like the ntrc::aphVIII mutant, CRISPR-based cp12::emx1 mutants also exhibited a redox-dependent cold phenotype. In addition, CP12 deletion resulted in robust decreases in both PRK and GAPDH protein levels implying a protein protection effect of CP12. Both CP12 functions are critical for preparing a repertoire of enzymes for rapid activation in response to environmental changes. This provides a crucial mechanism for cold acclimation.

Plants and other phototrophs in temperate climate zones experience both annual and diurnal temperature fluctuations. Low temperature is one of the most prominent abiotic stresses impeding growth and survival of these organisms, and thus a major concern in agriculture (1–3). Further, cold stress impacts many aspects of photosynthesis in both land plants and green algae (4–6). During photosynthesis, light energy absorbed by photosystems (PS) is converted into chemical energy, which is consumed by the Calvin–Benson cycle in which enzyme kinetics are highly temperature sensitive (7, 8). The balance between the energy source and the energy sink is termed as photostasis (9). Disparity occurs when the light energy absorption exceeds the metabolic utilization, resulting in excess excitation and overreduction of PS, mainly PSII (10). Overreduction causes not only reversible photoinhibition of PSII but also reactive oxygen species (ROS) generation via the Mehler reaction when photosynthetic electrons are accepted by oxygen molecules instead of NADP⁺ (11, 12). It has been reported that chloroplast ROS content increases in the cold, causing oxidative damage to the cell (13, 14).

Plants and algae cope with cold stress by adopting numerous mechanisms involved in cold acclimation (15–17). Re-establishment of photostasis is one of them. Upregulation of the Calvin–Benson cycle activities has been observed in cold-acclimated plants and is thought to increase the electron sink, avoiding electron leakage from the photosynthetic electron transport chain (ETC) (18–20). Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) in the Calvin–Benson cycle is especially important, because it is the enzyme that consumes NADPH and regenerates NADP⁺, which accepts electrons from the ETC. There are two subunits of GAPDH in plant chloroplasts, namely GapA and GapB, which form an A₂B₂-GAPDH heterotetrametric enzyme or an A₄-GAPDH homotetramer isoform (SI Appendix, Fig. S1) (21, 22). Interestingly, the most predominant difference between the two subunits is the C-terminal extension (CTE) of the GapB subunit, a 30 amino acid extension that is absent in the GapA subunit. The CTE contains two cysteine residues allowing the formation of a regulatory disulfide (23, 24). However, the GapB subunit is phylogenetically absent in cyanobacteria, green and red algae (SI Appendix, Fig. S1) (25–28); the only isoform in these organisms is the A₄-GAPDH, which was thought to be “nonregulatory” until the discovery of the Calvin Cycle Protein 12 (CP12) (29).

CP12 is a small, nuclear-encoded chloroplast protein. It is universally distributed in oxygenic photosynthetic organisms, including cyanobacteria, algae, mosses, and angiosperms (30, 31). Its amino acid sequence is highly similar to that of the GapB’s CTE but with an additional cysteine pair at the N terminus (29). The cysteine pairs at the C and N termini of CP12 are pivotal in binding GAPDH and phosphoribulokinase (PRK) enzymes, respectively, forming a PRK/CP12/GAPDH supracomplex of around 460 to 640 kDa (30, 32, 33). It has been demonstrated in vitro that the complex integrity depends on CP12 redox status, and dissociation of the complex releases active enzymes for catalytic activities (30, 32–35). In vitro studies also suggested that the complex is regulated by plastidial thioredoxins (Trx), namely Trx-f and Trx-m1/2 in response to light intensity (33, 35, 36).

Besides its extensively studied role in light acclimation, CP12 has been reported to be involved in cold tolerance of Stylosanthes guianensis, a forage legume (37). However, the molecular mechanisms of CP12 function in cold acclimation remain unknown. Interestingly, NADPH-dependent thioredoxin reductase C (NTRC), which is known as an alternative to the classical ferredoxin–thioredoxin (Fdx-Trx) system in chloroplast redox regulation, has also been reported to be involved in cold stress tolerance, although this was attributed to its chaperone functions (38). NTRC contains two functional domains on a single polypeptide—a NADPH-thioredoxin reductase (NTR) domain and a Trx domain (39). With the NTR domain, NTRC derives reducing potential from NADPH and hence, can function in both light and dark conditions. Furthermore, with its Trx domain, it behaves like other versatile Trx proteins of the Fdx–Trx system, having its own set of downstream redox targets (40–45). 2-Cys peroxiredoxins (2-Cys Prxs) are efficiently reduced by NTRC, which in turn maintains the reducing capacity of chloroplast Trxs and leads to a concerted redox function of the NTRC and Fdx–Trx systems (39, 46–48). While NTRC is known as an important redox hub to regulate acclimation of plants to different light conditions (42, 49), little is known on its role in cold acclimation.

In the present study, we found that redox activity of NTRC plays an important role to mediate cold acclimation responses in Chlamydomonas reinhardtii and Arabidopsis thaliana. In the search for downstream redox targets of NTRC responsible for cold tolerance, we identified CP12 as a binding partner of NTRC. We found that their interaction is both redox- and cold-dependent, triggering the dissociation/rearrangement of the PRK/CP12/GAPDH supracomplex. These findings led us to further scrutinize the molecular mechanism and impacts of NTRC and CP12 in carbon metabolism during cold acclimation.

Results

ntrc Mutants Display a Redox-Dependent Cold Phenotype.

It has previously been reported that NTRC is involved in cold stress and freezing tolerance in A. thaliana (38). Consistent with this report, we found that compared to wild type or a 35S::NTRC overexpression line, growth of an A. thaliana ntrc knockout mutant was much more strongly decreased at 4 °C than at 22 °C (Fig. 1D and SI Appendix, Fig. S2A). Moreover, both the wild type and the 35S::NTRC line showed visible induction of anthocyanin accumulation in the cold, but ntrc did not (Fig. 1D). This is consistent with previous studies indicating that the synthesis of this antioxidant pigment is under redox control (50). In addition to growth and metabolic parameters, NTRC deficiency also led to a much stronger decline of photosynthetic performance (Fv/Fm) in the cold, compared to warm conditions (Fig. 1E). When chlorophyll fluorescence kinetics were performed during dark-light transitions, loss of NTRC led to a much stronger decrease in photosynthetic efficiency of PSII and increase in nonphotochemical quenching at 4 °C as compared to 22 °C, which is consistent with an increased limitation in the Calvin–Benson cycle (SI Appendix, Fig. S2 B and C). This indicates a previously unknown role of NTRC in mediating cold acclimation of photosynthesis and metabolism, in addition to its known function in light acclimation in A. thaliana (42, 49).

Fig. 1.

Characterization of C. reinhardtii and A. thaliana with NTRC loss-of-function mutations in the cold. ntrc mutant strains of C. reinhardtii and A. thaliana were characterized at different growth temperatures. ntrc::aphVIII or ntrc denotes NTRC knockoutNTRC:HA-TG indicates ntrc::aphVIII complementation and 35S::NTRC indicates a representative NTRC overexpression line, while NTRC:HA-C455S and NTRC:HA-C455S-C458S were generated from the ntrc::aphVIII strain complemented with NTRC carrying point mutations leading to single or double cysteine to serine exchanges at the thioredoxin domain CGPC motif active site, respectively. The NTRC:HA-TG strain and the cysteine-mutants carry a HA tag at the NTRC’s C terminus. (A and B) Western blot quantification of NTRC protein levels, with RbcL serving as control. In B, loading of 35S::NTRC was half compared to wild-type and ntrc, when NTRC was analyzed, to allow a better comparison of NTRC protein amounts between genotypes. With respect to RbcL, loading was equal across genotypes. Quantification of three replicate blots show that NTRC protein level was 4.5-fold higher in the 35S:NTRC line, compared to wild type. (C) Growth studies of C. reinhardtii strains with different NTRC genotypes in TAP liquid medium at 15 °C (see SI Appendix, Fig. S4 D and E for growth at 23 °C). Culture densities were accessed every 2 d as OD₇₅₀. Data were compiled from three independent cultures; growth curves were then plotted using GraphPad Prism v8 software. Error bars denote SD. (D) Growth of A. thaliana ntrc mutant at 4 °C compared to 22 °C. Photos show representative plants (see also SI Appendix, Fig. S2A for fresh-weight analysis). Note that unlike the wild type, the ntrc mutant shows no visible anthocyanin accumulation in the cold. (E) Photosynthetic performance (Fv/Fm) of the plant ntrc mutant. Data represent mean ± SE (n = 7). (C and E) Doubling times in hours are shown underneath the graphs, with the corresponding SD values shown after “±” signs and different letters are indicating significant differences between the genotypes according to Two-Way ANOVA, Tukey test, P-value ≤ 0.05 (see SI Appendix, Fig. S3 for chlorophyll fluorescence kinetics performed during dark-light transitions).

To test whether NTRC cold-related function is evolutionarily conserved, a C. reinhardtii ntrc knockout mutant was analyzed in parallel. Similar to A. thaliana, the algal ntrc mutant displayed the well-studied light phenotype (42, 43) at 23 °C that could be reverted to wild-type growth by the addition of acetate to the medium (SI Appendix, Fig. S4 C–E). However, when mixotrophically growing lines were exposed to 15 °C, they grew more slowly and exhibited a longer lag phase than the wild type or ntrc::aphVIII complemented (denoted as NTRC:HA-TG) (Fig. 1 A and C and SI Appendix, Fig. S4D). Thus, cold-related phenotypic effects can be followed separately from light effects in the presence of acetate. All subsequent experiments involving C. reinhardtii were therefore performed under mixotrophic conditions. The growth study was then expanded to two other Chlamydomonas strains carrying either single or double cysteine-to-serine mutation(s) at the C455 and C458 residues of the NTRC protein, i.e., NTRC:HA-C455S and NTRC:HA-C455S-C458S. These two cysteine residues locate at the CGPC motif of NTRC’s thioredoxin domain and are responsible for redox reactions between NTRC and its targets (51). Both NTRC:HA-C455S and NTRC:HA-C455S-C458S exhibited a cold phenotype similar to the ntrc::aphVIII mutant, suggesting a redox-dependent mechanism is involved in NTRC-mediated cold acclimation (Fig. 1 A and C and SI Appendix, Fig. S4 B–E).

NTRC Interacts with CP12 in the Cold.

To identify NTRC’s redox target(s) that are responsible for cold acclimation, NTRC proteins were affinity-purified from NTRC:HA-TG cells by using the engineered HA-epitope tag at the protein’s C terminus (Fig. 2A). The single chloroplastic 2-Cys Prx of C. reinhardtii, named PRX1, was used as a positive control (52). Mass spectrometry analysis of the coimmunoprecipitation (co-IP) output revealed several candidates for potential cold-induced NTRC-binding partners. Among those, CP12, a chloroplast protein with four redox-sensitive cysteine residues (29, 30), exhibited a strict redox-dependent binding to NTRC, i.e., showed decreased or nondetectable binding enrichments when either the C455 alone or both the C455 and the C458 residues of NTRC were mutated (SI Appendix, Table S1 and Fig. 2B). This result suggested that CP12 represents a previously unrecognized target of NTRC’s redox activity and raised the possibility that CP12 might be involved in NTRC-mediated cold acclimation.

Fig. 2.

Studies of NTRC-CP12 interaction in C. reinhardtii. (A and B) co-IP analysis of HA-tagged NTRC. Anti-HA antibodies was used in the pull-down experiments; co-IP eluates from C. reinhardtii NTRC:HA-TG strain in which NTCR proteins were HA-tagged, were compared to those from the wild type used as a negative control. NTRC:HA-C455S-C458S strain was employed additional control to access the redox dependency of protein interactions. 2-Cys peroxiredoxin (PRX1) is a known redox target of NTRC, and hence was used as a positive control A, Silver-stained SDS-PAGE gel separating the co-IP eluates derived from wild-type and NTRC:HA-TG strains. (B) Shotgun mass spectrometry analysis of eluates derived from HA-tagged NTRC co-IP. All data were derived from four biological replicates. Top, Volcano plot demonstrating binding of CP12 to NTRC in NTRC:HA-TG strain which had been grown at 15 °C for 48 h. Each point indicates a different protein, ranked according to P-value (y axis, -log₁₀ of P-values) and relative abundance ratio (x axis, log₂ Fold Change NTRC:HA-TG/ WT). Protein candidates of interest are labelled in red. Bottom, Table showing binding enrichments of potential protein candidates to NTRC at different growth temperatures and different CGPC motif status. “–” sign denotes nondetectable binding. All data presented are with P-value <0.05. −log10 (0.05) = 1.3. (C) In vitro redox activity assays assessing reducing reactivity of recombinant NTRC on CP12. Recombinant NTRC and CP12 proteins of C. reinhardtii and A. thaliana were used in this assay. NTRC of Oryza sativa (rice) was used to substitute AtNTRC. In general, 100 µg CP12 proteins were either treated with 30 mM of NADPH, 150 µg NTRC proteins or combination of both. Plain buffer solution was used as the untreated control, while treatment with 5 mM of either reduced or oxidized DTT served as positive and negative controls respectively. The reaction mixes were then incubated for 1.5 h at 37 °C before proceeding to nonreducing SDS-PAGE followed by Coomassie blue staining. Pictures are from single experiment, which had been repeated at least twice with similar results.

Next, in vitro redox activity assays were performed to further substantiate that CP12 is indeed a redox target of NTRC. As previously reported (27), DTT-reduced recombinant C. reinhardtii (Cr) CP12 migrated more slowly than the oxidized form during nonreducing Sodium dodecyl-sulfate polyacrylamide gel electrophoresis (SDS-PAGE), indicating intramolecular disulfide bridges that affect CP12 conformation (Fig. 2 C, Upper gel). The resulting shift was detected again when CrCP12 was treated with a combination of CrNTRC and NADPH, but not with these components alone, demonstrating that NTRC reduces CrCP12 with electrons derived from NADPH (Fig. 2 C, Upper gel). Furthermore, a cross-species reactivity was observed for CP12 reduction via NTRC. The algal NTRC can also reduce A. thaliana (At) CP12 isoform-2 (Fig. 2 C, Lower gel). These data reveal that the reduction of CP12 by NTRC is phylogenetically conserved across oxygenic photosynthetic organisms.

NTRC Dissociates the PRK/CP12/GAPDH Complex In Vitro.

CP12 is known to function as a linker protein for the Calvin–Benson cycle enzymes GAPDH and PRK across all phototrophs, thereby forming an autoinhibitory PRK/CP12/GAPDH supracomplex (29, 30). This raises the question if NTRC via CP12, also has an impact on this complex assembly. Therefore, the native PRK/CP12/GAPDH complex status in protein extracts from the wild type and ntrc::aphVIII mutant of C. reinhardtii was analyzed by using size exclusion chromatography (SEC; Fig. 3). The complex status was followed by PRK, GAPDH subunit-A, and CP12 protein detection using their respective antibodies. In both strains, all the three proteins were detected predominantly in a size range of 440 kDa, with overlapping peaks of elution in fractions 10 and 11 (SI Appendix, Fig. S5). These high molecular weight (HMW) complexes likely represent the PRK/CP12/GAPDH complex of C. reinhardtii, which has a reported size of 460 kDa (34, 53).

Fig. 3.

Effects of NTRC and chilling on the PRK/CP12/GAPDH complex status. SEC assay analysing the in vivo status of PRK/CP12/GAPDH complex in C. reinhardtii and A. thaliana. Expected sizes of PRK/CP12/GAPDH complexes, A₄-GAPDH tetramers and PRK dimers based on previous studies, were labelled above the molecular weight scale (34, 53, 54). The complex SEC-profiles presented are from single experiment out of biological triplicates with similar results. (A) SEC study of the in vivo complex status of PRK/CP12/GAPDH in C. reinhardtii. The study compared the complex status in wild-type, ntrc::aphVIII, and NTRC:HA-TG strains of C. reinhardtii cultured at 15 °C or 23 °C for 48 h. (B) SEC assay profiling the PRK/CP12/GAPDH complex status in different A. thaliana lines at different growth temperatures. ntrc::aphVIII indicates NTRC knockout mutant, whereas 35S::NTRC denotes NTRC overexpression. Four weeks old plants were grown under long day conditions of 16 h at 120 µE/m²/s light, with a temperature of 4 °C or 22 °C for 2 d before leaf soluble proteins were extracted for SEC analysis followed by western blot detection of PRK and GapA proteins. Arrows labelled A and B denote GapA (42.5 kDa) and nonspecific GapB (47.6 kDa) detection bands respectively.

Previously, NADPH binding to the GAPDH enzyme has been demonstrated to favor the dissociation of the PRK/CP12/GAPDH complex (30, 31, 55, 56). In line with that, treatment of the soluble protein extract from the wild-type C. reinhardtii with NADPH induced pronounced shifts in PRK, GapA, and CP12 protein distributions toward the LMW region in SEC analysis (SI Appendix, Fig. S5, Left). This elution profile most likely reflects PRK dimers, A₄-GAPDH tetramers and CP12 monomers, indicating a complete dissociation of the complex. However, in ntrc extracts lacking the NTRC enzyme, treatment with NADPH only dissociated the complex partially. A full complex dissociation was only achieved by either prolonging the incubation time or when recombinant CrNTRC was added in parallel (SI Appendix, Fig. S5, Right). These in vitro observations suggest that CrNTRC, either native or recombinant, promotes the dissociation of the PRK/CP12/GAPDH complex in the presence of NADPH, most likely via CP12 reduction.

To test whether the PRK/CP12/GAPDH complex assembly indeed inhibits PRK and GAPDH as suggested by previous in vitro findings (34, 57, 58), their enzymatic activities were assayed in SEC fractions 10 and 13 from wild-type algal extract containing CP12-complexed and free enzymes, respectively (SI Appendix, Fig. S5). Untreated extract contained mostly the PRK/CP12/GAPDH complex and showed a relatively low GAPDH activity (SI Appendix, Fig. S5) in both fractions. In contrast, upon addition of NADPH the PRK/CP12/GAPDH complex was dissociated, and GAPDH activity clearly increased in fraction 13 containing free enzymes (SI Appendix, Fig. S5). Moreover, PRK/CP12/GAPDH complex dissociation was also triggered by artificial reducing agents, i.e., reduced DTT (SI Appendix, Fig. S7A). Similar to NADPH treatment, this was accompanied by a drastic increase in the catalytic activity of GAPDH (SI Appendix, Fig. S6B). These analyses clearly show the autoinhibitory effect of PRK/CP12/GAPDH complex, which is relieved upon complex dissociation.

NTRC Deletion Suppresses PRK/CP12/GAPDH Complex Dissociation in Arabidopsis but Not in Chlamydomonas.

As a next step, native PRK/CP12/GAPDH complex formation was followed in the wild type, the mutant ntrc::aphVIII, and the complemented NTRC:HA-TG strains of C. reinhardtii at 23 °C and 15 °C (Fig. 3A). All analyzed algal strains exhibited SEC-elution peaks of both PRK and GAPDH at ca. 440 kDa (fractions 10 and 11), indicating neither temperature nor NTRC effects on complex integrity in vivo (Fig. 3A).

Intriguingly and in contrast to C. reinhardtii, the PRK/CP12/GAPDH complex status in A. thaliana was strongly affected by either the lack of NTRC or the temperature (Fig. 3B). At room temperature, PRK was detected in the LMW fractions 13 to 15 in both wild-type and 35S::NTRC overexpression lines, consistent with the reported size of a PRK dimer (Fig. 3 B, Upper) (54). Conversely, the ntrc knockout plant exhibited an additional elution peak at 440 kDa (fractions 10 and 11; Fig. 3 B, Upper). These HMW complexes likely represent the PRK/CP12/GAPDH complex (54), which was also detected when a GapA antibody was applied (Fig. 3 B, Lower), indicating that complex dissociation was impeded in the absence of NTRC.

The impact of NTRC on complex integrity was even stronger in cold-treated A. thaliana plants. At 4 °C, NTRC deletion caused almost complete inhibition of complex dissociation as indicated by the virtual absence of free PRK and GAPDH enzymes (Fig. 3B). On the other hand, comparably higher levels of PRK dimer and GAPDH tetramers were detected in the wild type (Fig. 3B). This effect was even more pronounced in 35S::NTRC plants, illustrating a dose-dependent effect of NTRC proteins on dissociating the complex in vivo (Fig. 3B). Together, these results demonstrate that, in A. thaliana, NTRC triggers the dissociation of the PRK/CP12/GAPDH complex in vivo, particularly in the cold.

Loss of NTRC Function Impairs Cold-Induced Elevation of GAPDH Enzyme Activity.

To find out if NTRC also affects enzyme activities of the PRK/CP12/GAPDH complex in vivo, GAPDH activity in ntrc::aphVIII knockout and cysteine-substituted mutant strains of C. reinhardtii was examined. A general increase in enzyme activity occurred when the algae were shifted from 23 °C to 15 °C for cultivation (SI Appendix, Fig. S8). However, this induction was not as strong in the ntrc::aphVIII and NTRC:HA-C455S-C458S mutant strains; there was only a fourfold increase in the mutants as compared to a sevenfold increase in the wild type and NTRC:HA-TG (SI Appendix, Fig. S8). These results indicate that the loss of NTRC’s redox function in C. reinhardtii did affect the catalytic activity of GAPDH, despite the lack of detectable changes in the PRK/CP12/GAPDH complex status (Fig. 3A). In line with this, the same cold-induced increase in GAPDH enzyme activity occurred in wild type and 35S::NTRC plant lines in vivo (SI Appendix, Fig. S9). This effect was abolished in the Arabidopsis ntrc mutant, further supporting the idea of NTRC acting as a regulator of PRK/CP12/GAPDH complex formation and resulting Calvin–Benson cycle activity in vivo (SI Appendix, Fig. S9).

cp12::emx1 Mutants Are Redox- and Cold-Sensitive.

The identification of CP12 as a downstream redox target of NTRC (Fig. 2C) at low temperature (Fig. 2B) suggested that the loss of CP12 function might also affect growth under cold conditions. Since A. thaliana contains three different CP12 isoforms, we focused on the C. reinhardtii system, where CP12 is encoded by a single gene (59) to address this question. To this end, we inactivated CP12 by generating a C. reinhardtii cp12::emx1 insertion mutant using the CRISPR-Cas9 technique (SI Appendix, Fig. S10A). Redox dependency was then tested by introducing a wild-type CP12 protein or a mutated version with the two pairs of cysteine residues being substituted with serine into this cp12::emx1 mutant background (CP12-TG and CP12:C→S, respectively) (SI Appendix, Fig. S10B).

PRK protein accumulations decreased remarkably in cp12::emx1 and the lack of disulfide bridge possession in CP12:C→S lines clearly affected enzyme accumulation (Fig. 4A). Similar observations have been reported for C. reinhardtii and A. thaliana, which revealed decreased PRK protein levels in CP12 mutant lines (60, 61). It was hypothesized that this is due to the loss of a protein-stabilizing effect exerted by CP12 on PRK (61–63). As expected, the cp12::emx1 mutant also failed in forming the PRK/CP12/GAPDH complex as evidenced by SEC assays (SI Appendix, Fig. S10E). At 23 °C, no significant differences in the growth rate of CP12 mutants and wild type were detected, in line with recent findings in C. reinhardtii (SI Appendix, Fig. S10 C and D) (61). In contrast, at 15 °C, both cp12::emx1 and the CP12:C→S mutant exhibited retarded growth as compared to both the wild type and the complementation line CP12-TG (Fig. 4B and SI Appendix, Fig. S10C). These differential growth rates under cold conditions provide direct evidence that CP12 is involved in cold acclimation in a redox-dependent manner.

Fig. 4.

Characterization of C. reinhardtii CP12 mutants under cold conditions. The cp12::emx1 mutant of C. reinhardtii was generated with aid of CRISPR-Cas9 and complemented with CP12 gDNA (CP12-TG) or CP12 carrying quadruple cysteine to serine exchanges (CP12:C→S). (A) Immunoblot analysis comparing CP12, PRK, GapA, NTRC protein accumulations in the mutants and wild type at 23 °C; RbcL serves as a loading control. Blots shown are representative from three biological replicates with similar results. (B) Growth studies of CP12 strains and wild type in TAP media at 30 µE/m²/s light and 15 °C. OD₇₅₀ of the cultures as read every 2 d. Data were derived from three separate cultures, and growth curves were drawn using GraphPad Prism software. Error bars represent SD. Values below the curves are doubling times (hours) and their respective SD values shown after “±” signs and different letters are indicating significant differences between the genotypes according to Two-Way ANOVA, Tukey test, P-value ≤ 0.05.

Finally, we generated a ntrc;cp12 double mutant via CRISPR- based inactivation of the CP12 gene in the ntrc::aphVIII mutant background of C. reinhardtii (SI Appendix, Figs. S10A and S11A). Analysis of its growth phenotype revealed an even higher sensitivity toward cold conditions as compared to either the ntrc::aphVIII or the cp12::emx1 single mutants (SI Appendix, Fig. S11B). This further underlines the critical role of the NTRC-CP12 regulatory system for adequate cold acclimation.

Discussion

The chloroplast contains two different thiol-redox systems, namely the light-dependent Fdx-Trx system and the NADPH-dependent NTRC system, which are known to be important in redox regulation of chloroplast functions and in acclimation of plants to different light conditions (64). Recently NTRC was also found to have a role in acclimation of plants to low temperature, but while it was suggested that this is due to its cryoprotective chaperone function, the underlying mechanisms still remained unclear (38). Here, we unambiguously demonstrate by substitution of the redox-regulatory cysteines that NTRC mediates cold tolerance in C. reinhardtii via its oxidoreductase activity. We clarified the molecular mechanisms and found that NTRC interacts with CP12 and reduces this regulatory protein by thiol-disulfide modulation, leading to the dissociation of the inhibitory PRK/CP12/GAPDH complex in A. thaliana though not in C. reinhardtii, to activate PRK and GAPDH, which will contribute to reestablish photostasis in the cold.

NTRC Interacts with CP12 to Regulate the Calvin–Benson Cycle via Ternary Complex Formation.

The data show that NTRC regulates the Calvin–Benson cycle in A. thaliana by a previously unknown mechanism. NTRC interacts with the Calvin–Benson cycle regulator CP12 by direct thiol-disulfide interaction, leading to an NADPH-dependent increase in the reduction state of the protein. This is followed by the dissociation of an inhibitory PRK/CP12/GAPDH supracomplex in A. thaliana in vivo (Fig. 3B) and C. reinhardtii in vitro (SI Appendix, Fig. S5), and increased GAPDH enzyme activity in the cold in both organisms (SI Appendix, Figs. S8 and S9). A role of CP12 in PRK/CP12/GAPDH supracomplex formation has already been found in previous studies on the regulation of the Calvin–Benson cycle in response to light (35, 54, 55), although light-induced complex dissociation has not been documented for Chlamydomonas and Arabidopsis in vivo (53, 60, 65). Based on in vitro experiments, it was proposed that CP12 is mainly reduced by Trx-f and m1/2, while the role of NTRC has not been investigated in this context (35, 36). NTRC has been found to interact with 2-Cys Prxs providing evidence for a major role of NTRC to regulate the redox poise of 2-Cys Prxs for antioxidative function (39). Interestingly, while in Arabidopsis genetic and biochemical evidence was provided that NTRC regulates enzymes of the Calvin–Benson cycle, such as FBPase, SBPase, GAPDH and PRK in response to light in vivo (42, 66–68), subsequent experiments showed no direct interaction between NTRC and FBPase or SBPase in vitro (68, 69), suggesting that NTRC acts on these enzymes by an indirect mechanism. Since the growth phenotype of NTRC loss of function mutants was reverted in the background of a 2-Cys Prx mutant, NTRC was suggested to modulate the reduction state of Calvin–Benson cycle enzymes in an indirect manner by acting on the redox balance of 2-Cys Prx (47, 48, 68). This contrasts with studies using bimolecular fluorescence complementation tests, which found interactions of NTRC and the Calvin–Benson cycle enzymes FBPase and PRK in vivo, while the underlying mechanisms remained unclear (67). In the present study, we provide in-vitro and in-vivo evidence that NTRC regulates PRK and GAPDH activities by modulating the formation of an inhibitory PRK/CP12/GAPDH supracomplex via direct interaction and reduction of CP12, providing a mechanism for the regulation of the Calvin–Benson cycle by NTRC.

While in Arabidopsis, ternary complex formation was affected in response to cold and NTRC deficiency in vivo, a discrepancy was observed in the native complex status of C. reinhardtii, which was unresponsive to temperature and light intensity changes as well as the absence of NTRC (Fig. 3A). Complex dissociation in C. reinhardtii was only achieved under the trigger of artificial reductants in a much higher dosage than what occurs in the physiological condition (SI Appendix, Figs. S5, S6A, and S7A). These discrepancies appear to be related to intrinsic differences between the two organisms from molecular level to ecological level. Unlike A. thaliana, instead of a highly heterogenous conformation and autonomous GAPDH regulation, C. reinhardtii CP12 has its C terminus forming a stable α-helix structure making its binding to the GAPDH fit better the “lock and key” model and thus reducing the cost of entropy, contributing to the unusually high binding affinity between CP12 and GAPDH in C. reinhardtii (70, 71). The GAPDH-CP12 binding affinity in C. reinhardtii has been reported to be around 400-fold higher than that in A. thaliana, making the complex dominating and hindering the detection of the free enzymes in algae (SI Appendix, Fig. S12) (32, 72). Phylogenetic variation in the amino acid sequence of CP12 could partly explain these differences in enzyme regulation among species (73). Furthermore, with the absence of regulatory GapB subunit, which is responsive to thioredoxins and other metabolites, complex formation with CP12 is thus far the only known regulatory mechanism for the algal A₄-GAPDH catalytic activity (23, 28, 74, 75). Hence, it is not surprising to observe a more robust change in GAPDH activity induced by NTRC knockout in C. reinhardtii than A. thaliana, despite the unaltered algal in vivo complex status (SI Appendix, Figs. S8 and S9). Last, the presence of a pyrenoid with densely packed RuBisCO enzymes in C. reinhardtii could provide a localized reaction site for the Calvin–Benson cycle to take place. Hence, instead of a full release of GAPDH and PRK enzymes from the supracomplex, a conformational change of the complex might be sufficient to expose enzyme reaction sites for the access of substrates when all the Calvin–Benson cycle enzymes are in close proximity. More studies will be necessary to investigate the differences in CP12 complex formation and thereby regulating of carbon metabolism in algae and plants.

The Role of NTRC and CP12 in the Redox Regulation of Cold Acclimation.

NTRC is known as an important redox hub to regulate acclimation of plants to different light conditions (42, 49). The results of the present study show that NTRC also affects cold-acclimation responses in A. thaliana and C. reinhardtii by its redox-regulatory function, rather than acting as a chaperone. This suggests NTRC acting as a major integrator of redox processes that allow algae and plants to acclimate to fluctuations in, both, light and temperature that occur very rapidly in nature. Our finding of CP12 as a direct redox target of NTRC unveils a previously unrecognized redox regulatory mechanism for the Calvin–Benson cycle that is important to allow algae and plants to acclimate to low temperature. Our results in C. reinhardtii mutants deficient in CP12 function further support the important role of this mechanism for cold acclimation in vivo. The activation of the Calvin–Benson cycle has been suggested as an adaptive mechanism under cold stress by which the enhanced regeneration of NADP⁺ will contribute to utilize excess electrons from the thylakoid ETC and hence minimize the generation of harmful H₂O₂ molecules to optimize photosynthesis (18–20). In confirmation of this, NTRC deficiency led to impaired photosynthetic efficiency in the cold (Fig. 1E and SI Appendix, Fig. S2 B and C), indicating that activation of GAPDH and PRK enzyme activity by NTRC via CP12 will contribute to re-establish photostasis in the cold (SI Appendix, Figs. S8 and S9). Interestingly, compared to wild-type, 4.5-fold overexpression of NTRC did not lead to a substantial increase in photosynthetic efficiency or growth in the cold. This indicates that when NTRC is increased above wild-type level, other factors may become limiting to establish improved cold acclimation of photosynthesis and growth. This may involve other plastidial redox modulators (i.e., Fdx-dependent Trxs) working in tandem with NTRC, biosynthetic capacities for osmoprotectants/cryoprotectants, antioxidative functions or lipid remodeling capacities for membrane stabilization (76). Moreover, in line with a recent finding, our targeted mutagenesis of CP12 also supports the idea of a moonlighting function of CP12, i.e., the protection of Calvin–Benson cycle enzymes, particularly PRK (57, 61). Both properties are important to maintain critical Calvin–Benson cycle enzymes in a state that is inactive but poised for rapid activation upon a decrease in temperature or an increase in light intensity.

In conclusion, the present study provides an understanding of the molecular mechanism of NTRC and CP12 in cold acclimation (SI Appendix, Fig. S13). Our data demonstrate that NTRC reduces CP12 by thiol-disulfide modulation and leads to the dissociation of the PRK/CP12/GAPDH complex in A. thaliana though not in C. reinhardtii, which in turn results in an increase in the activity of GAPDH enzyme activity of the Calvin–Benson cycle. This regulatory pathway is specifically important during cold acclimation of photosynthesis, providing a mechanistic understanding of why NTRC and CP12 are important for acclimation to cold stress.

Materials and Methods

Algal Culture and Growth Study.

C. reinhardtii strains were grown in Tris-acetate-phosphate (TAP) medium under constant light of 30 µE/m²/s at 23 °C, shaken at 120 rpm, unless otherwise specified. Growth temperature was changed to 15 °C for cold studies. TAPS medium—TAP supplemented with 1 % sorbitol (w/v)—was used instead for culturing cell wall–deficient (cw15) strains. ntrc::aphVIII knockout, complementation, and cysteine-substituted strains were generated from CC-4051 strain, which has intact cell wall, whereas cp12::emx1 insertion, complementation, and cysteine-substituted strains were of CC-406 (cw15) background.

Generation of Algal ntrc::aphVIII Knockout, Complementation, and Cysteine-to-Serine Mutant Strains.

The ntrc::aphVIII knockout mutant strain was generated by insertional mutagenesis of CC-4051 (SI Appendix, Fig. S4A) (77). This mutant strain was complemented with 1) NTRC gDNA carried by pBC1 plasmid vector to generate NTRC:HA-TG strain; 2) NTRC gDNA with cysteine-to-serine mutation(s) at C455 or both C455 and C458 residues to generate NTRC:HA-C455S and NTRC:HA-C455S-C458S mutant strains, respectively (SI Appendix, Fig. S4B). All constructs fuse an HA-tag to the NTRC C terminus. Detailed procedures of plasmid construct generation and C. reinhardtii transformation via electroporation are described in SI Appendix, Text.

Plant Material and Growth.

The A. thaliana T-DNA insertion line ntrc (SALK_012208) has been extensively characterized in previous studies (42, 49, 78). To generate transgenic 35S::NTRC A. thaliana plants, the full coding sequence of AtNTRC (AT2G41680) was PCR-amplified from the previously created 35S:NTRC:sCFP3A-pBar plasmids (66). To attach an XbaI restriction site to each end of the amplified coding sequence, the following pair of oligonucleotides was used: NTRC-fwd 5′-TATCTAGAATGGCTG CGTCTCCCAAGATAGGCATCG-3′/NTRC-rev 5′-TATCTAGATCATTTATTGGCCTCAATGAATTCTCGG-3′. The resulting amplicon was XbaI restricted and ligated into a CaMV 35S:XbaI:tocs expression cassette contained in the pPZP200BAR binary vector (GenBank accession number KX227611). Selected plasmids were sequenced for validation purposes and then introduced into Agrobacterium tumefaciens strain EHA 105. A. thaliana plants were transformed via the floral dip method as described previously (79). From four different lines selected, two independent transgenic lines overexpressing NTRC by threefold-to-fourfold of wild-type level were used for further analysis, showing a similar growth and photosynthesis phenotype in the cold. Representative data of one of these lines are shown in this manuscript.

To analyze growth phenotype, fresh weight and chlorophyll fluorescence, plants were first grown on ½ MS medium under long day conditions of 16 h at 120 µE/m²/s with 22 °C for 1 wk, followed by 8 wk at 4 °C for the cold treatment or 10 d at 22 °C for the control treatment. After that, fresh weights of plant rosettes cut above the hypocotyl were measured using an analytical balance (ABT 220-4M, Kern), or the entire plants were subjected to chlorophyll fluorescence analysis.

For SEC analysis and GAPDH enzyme activity measurement, plants were grown on soil and long day conditions of 16 h at 120 µE/m²/s with 22 °C for 4 wk, followed by 2 d at 4 °C for the cold treatment or at 22 °C for the control. Those plants were harvested at the respective treatment temperature by cutting them from the hypocotyl and snap-freezing in liquid nitrogen. At least two plants were harvested per biological replicate. The material was subsequently pulverised in a liquid nitrogen cooled ball mill (MM 400, Retsch GmbH) and stored at −80 °C. The frozen leaf powder was then extracted for proteins for different biochemistry analyses.

Plant Chlorophyll Fluorescence Analysis.

The in vivo chlorophyll fluorescence analyses were performed on A. thaliana plants in the middle of the light period, using the Walz MAXI IMAGING-PAM. Prior to measurements, plants were dark acclimated for 30 min at their respective treatment temperature. The maximum quantum yield of PSII (Fv/Fm) was determined using a saturation pulse of 2,700 µE/m²/s for 0.8 s followed by a dark–light transition kinetic with an illumination of actinic light similar to growth light. The effective PSII and the nonphotochemical quantum yields were calculated according to Genty et al. (80) and Kramer et al. (81), respectively.

SDS-PAGE and Immunoblotting Analysis.

SDS-PAGE and immunoblot analysis were performed with standard procedures. First, algal total protein extraction was performed using ice-cold RIPA buffer (150 mM NaCl, 20 mM Tris-HCl pH 7.8, 1 mM EDTA, 1 % (w/v) Triton X-100, 0.5 % (w/v) sodium deoxycholate, 0.1 % (w/v) SDS, and 5 mM β-mercaptoethanol) supplemented with 8 M Urea and complete™ protease inhibitor cocktail (Roche #04693116001). The recipe was modified from the RIPA buffer recipe of the Cell Signalling Technology company. Lysates were homogenized (3 × 5 s, 10 % amplitude) on ice using Sonopuls HD2070 sonicator (Bandelin Electronics) followed by debris removal by centrifugation at 20,000 g, 4 °C for 10 min. Protein concentrations were then determined using Pierce® BCA Protein Assay Kit (Thermo Fisher Scientific, #23225). On the other hand, total plant proteins were extracted from the frozen, pulverized plant material using twofold Laemmli buffer (82) with 20 mM DTT instead of 2-mercaptoethanol. The lysate was shaken vigorously for 15 min at room temperature followed by a 5-min incubation at 90 °C. Subsequently, 30 µg of algal total protein extract or a lysate equivalent to 1 mg (unless otherwise specified in the corresponding figure legend) of the plant fresh weight was loaded per lane of the acrylamide gel. Noted that for the electrophoresis of SEC fractions and co-IP elutes, 40-µL sample was loaded per lane instead.

All primary antibodies except anti-ATPβ were derived from rabbit sera and used at 1:1,000 dilution. The anti-ATPβ antibody was produced by chicken and used at 1:4,000 dilution. Antibodies against GapA and PRK were provided by R. Scheibe (University of Osnabrueck) (83) while anti-RbcL antibody was a kind gift from G. F. Wildner (Ruhr University of Bochum). Antibodies against NTRC and CP12 were generated by our lab as described in SI Appendix, Text. Primary antibodies against HA (Sigma Aldrich, #H6908), ATPβ (Agrisera, #AS03-030-10), CP47 (Agrisera, #AS04-038) and D2 (Biogenes, #6441) were purchased from the respective companies. Horseradish peroxidase (HRP)-conjugated anti-rabbit (#A9169) and anti-chicken (#A9046) secondary antibodies were purchased from Sigma-Aldrich and used at 1:5,000 dilution.

Antibodies listed above were used in all western blot analyses except the one in Fig. 1B; the antibody against Arabidopsis NTRC was kindly provided by F. J. Cejudo (University of Sevilla) (39) and used at a 1:1,500 dilution. Moreover, antibody against the plant RbcL (#AS03-037) and HRP-conjugated anti-rabbit secondary antibody (#AS09-602) were purchased from Agrisera and used at a 1:5,000 dilution and 1:10,000 dilution, respectively.

Algal Soluble Protein Extractions.

Algal soluble protein extracts were prepared for co-IP, SEC, and enzyme activity measurements. Algal cells were first pelleted and resuspended in ice-cold IP lysis buffer (150 mM NaCl, 5 % glycerol (v/v), 1 mM EDTA, 20 mM Tris-HCl pH 7.8) modified from recipe of Pierce® IP lysis buffer of Thermo Fisher Scientific and supplemented with cOmplete™ protease inhibitor cocktail (Roche, #04693116001). Cell wall breakage was carried out with the help of 0.5-mm glass beads and BeadBug™ microtube homogenizer (5 × 40 s, 4,000 rpm; Biozym Scientific), followed by centrifugation at 20,000 g, 4 °C for 20 min to remove unbroken cells and debris. Protein concentrations were then determined using ROTI® Quant Bradford assay (Carl Roth, #K015).

co-IP.

The pull-down experiment was performed using Pierce® Anti-HA magnetic beads (Thermo Fisher Scientific, #88837) as per the manufacturer’s manual. NTRC was found to be a soluble protein (SI Appendix, Fig. S2), and hence soluble protein extracts from C. reinhardtii were used here. For each sample, 2-mg algal soluble proteins (in 500 µL) extracted as described above was incubated with 50 µL magnetic beads at 4 °C overnight. The beads were then collected with SureBeads™ magnetic rack (Bio-Rad) and washed 3 times with 500 µL IP lysis buffer. Then, acidic elution was performed at room temperature for 10 min using 50 µL 0.1 M glycine, pH 2.0, followed by neutralization with 50 µL 0.1 mM ammonium bicarbonate. Eluates were then either subjected to SDS-PAGE followed by silver staining or sent for mass spectrometry analysis (refer to SI Appendix, Text for the protocols).

SEC.

The gel filtration experiments were carried out using a ÄKTA purifier 10 HPLC system (GE Healthcare). Algal soluble protein extract was first prepared as described in the previous section, while plant soluble proteins were extracted from frozen leaf powder by vigorous shaking in ice-cold IP lysis buffer, supplemented with cOmplete™ mini protease inhibitor cocktail (Roche, #11836170001). After that, centrifugation at 20,000 g, 4 °C for 2 min was carried out to remove insoluble materials. Collected supernatants were then measured for protein concentrations using ROTI® Quant Bradford assay (Carl Roth, #K015). Subsequently, for each C. reinhardtii or A. thaliana sample, 5 mg of the prepared soluble protein extract (in 500 µL) was loaded onto Superdex®200 Increase 10/300GL column (GE Healthcare). Elution by IP lysis buffer was performed at flow rate of 0.5 mL/min, 4 °C, with fractionation volume of 1 mL. After that, 30 µL of each fraction was mixed with 10 µL ROTI®Load1 (4×) reducing loading dye (Carl Roth, #K929) before loading for SDS-PAGE and immunoblotting analysis.

For in vitro NADPH-induced PRK/CP12/GAPDH dissociation studies, 5-mg soluble algal protein extract was pretreated with 12 mM NADPH (Carl Roth, #AE14), 2-mg recombinant CrNTRC protein or both for 2 h at 4 °C, before loading onto the column. On the other hand, DTT induced PRK/CP12/GAPDH dissociation was performed by preincubation of the soluble algal protein extract with 20 mM reduced DTT (AppliChem, #A2948) at 30 °C for 1 h.

GAPDH Enzyme Activity Assay.

Enzyme activity of GAPDH consumes NADPH which can be measured with its maximal absorbance at 340 nm. Algal soluble protein extract was first prepared as described above. On the other hand, frozen leaf powder was shaken vigorously in ice-cold buffer containing 50 mM HEPES pH 7.4, 10 mM MgCl₂, 10 % glycerol (v/v), 1 mM EDTA, 1 mM EGTA, supplemented with complete™ mini protease inhibitor cocktail (Roche, #11836170001). This was followed by centrifugation at 20,000 g and 4 °C for 2 min, and the resulting supernatant was collected. A reaction mixture containing 100 mM HEPES pH 8.0, 10 mM MgCl₂, 1 mM EDTA, 5 mM ATP (Sigma Aldrich, #A7699), 0.2 mM NADPH (Carl Roth, #AE14) and 10 U/mL phosphoglucokinase (PGK; Sigma Aldrich, #P7634) enzyme was then prepared. Subsequently, 10 µg algal soluble protein extract or 15 µL of the plant supernatant was added to each reaction mix to make a final volume of 200 µL. The reaction was initiated by the addition of 0.3 mM glycerate-3-phosphate (3PG; Sigma Aldrich, #P8877) into each reaction mixture. The reaction took place at 30 °C, and the absorbance at 340 nm was measured using FilterMax® F5 multimode microplate reader (Molecular Devices).

In Vitro Redox Activity Assay.

CrNTRC, OsNTRC, CrCP12, and AtCP12 recombinant proteins used in this assay were prepared as described in SI Appendix, Text, while reaction buffer used was IP lysis buffer. Then, 100 µg CP12 proteins were mixed with 150 µg NTRC proteins, 30 mM NADPH (Carl Roth, #AE14) or both. For untreated, positive, and negative controls, 100 µg CP12 or PRX1 proteins were treated with blank buffer, 5 mM reduced DTT (AppliChem, #A2948) or oxidized DTT (Sigma Aldrich, #D3511), respectively. All reactions were equalized to a final volume of 20 µL in PCR tubes and incubated at 37 °C for 1.5 h. Subsequently, each reaction was stopped by adding 7 µL ROTI®Load2 (4×) nonreducing loading dye (Carl Roth, #K930). The proteins were then separated by SDS-PAGE and stained with Coomassie Brilliant Blue G250 (Carl Roth #9598).

Generation of cp12::emx1 Mutants via CRISPR/Cas9, Complementation and Cysteine-to-Serine Mutant.

CP12 insertional mutagenesis was carried out in C. reinhardtii CC-406 (cw15) strain using the CRISPR/Cas9 technique following the protocol established by Greiner (84). First, the CP12 crRNA sequence CP12 crRNA: 5′-AGCAGCGCUGAGCUCCUCAACGG-3′ was designed to target CP12 exon-2 (SI Appendix, Fig. S10A). The crRNA was then annealed with tracrRNA (Integrated DNA Technologies, #1072533) to generate noncoding guide RNAs (gRNA), followed by conjugation with SpCas9 endonuclease (Integrated DNA Technologies, #1081058) to form SpCas9/gRNA complexes. The preassembled SpCas9/gRNA complex was then cotransferred with mutation oligonucleotide containing the human gene EMX1 and pBC1 empty plasmid vector which carry frameshift tolerant stop codons and a paromomycin resistance cassette, respectively, into the microalga via electroporation using NEPA21 electroporator (Nepa Gene). Mutagenesis with CP12 crRNA was expanded to the ntrc::aphVIII mutant strain, which later gave rise to the ntrc;cp12 double mutant strains. This cp12::emx1 mutant was complemented with 1) CP12 gDNA to generate a CP12-TG strain; 2) CP12 gDNA with the two pairs of redox-sensitive cysteine residues substituted to serine resulting in a CP12:C→S strains. Both complementation constructs were expressed under the control of a PsaD promoter in a pCrZ4 plasmid vector harboring a zeocin resistance cassette (SI Appendix, Fig. S10B). Nuclear transformation was performed via the glass bead method (85). Transformed microalgae were then selected under paromomycin or Zeocin, respectively, followed by screening via colony-PCR and sequencing. An extended protocol can be found in SI Appendix, Text.

Supplementary Material

Appendix 01 (PDF)

Data, Materials, and Software Availability

DNA and amino acid sequences for microalgal and plant genes of interest were retrieved from Phytozome v12.1 of Joint Genome Institute (JGI), under the following loci identifiers: Cre01.g054150 (CrNTRC), Cre08.g380250 (CrCP12), Cre06.g257601 (CrPRX1), AT2G41680 526 (AtNTRC), AT3G62410 (AtCP12-2) and Pp3c20_5530 (PpNTRC) (86). http://genome.microbedb.jp/cyanobase/. For SyNTRC, sequence was retrieved from CyanoBase under gene identifier: Slr0600 (87). All other data are included in the article and/or SI Appendix.