Roles of Two Phytoene Synthases and Orange Protein in Carotenoid Metabolism of the β-Carotene-Accumulating Dunaliella salina

Phytoene synthase (PSY) as the key rate-limiting enzyme in carotenoid metabolism can be regulated by various regulators and factors. We found that DsPSY1 played a dominant role in carotenogenesis in the β-carotene-accumulating Dunaliella salina, and two amino acid residues critical in the substrate binding were associated with the functional variance between DsPSY1 and DsPSY2. ABSTRACT Phytoene synthase (PSY) is a key enzyme in carotenoid metabolism and often regulated by orange protein. However, few studies have focused on the functional differentiation of the two PSYs and their regulation by protein interaction in the β-carotene-accumulating Dunaliella salina CCAP 19/18. In this study, we confirmed that DsPSY1 from D. salina possessed high PSY catalytic activity, whereas DsPSY2 almost had no activity. Two amino acid residues at positions 144 and 285 responsible for substrate binding were associated with the functional variance between DsPSY1 and DsPSY2. Moreover, orange protein from D. salina (DsOR) could interact with DsPSY1/2. DbPSY from Dunaliella sp. FACHB-847 also had high PSY activity, but DbOR could not interact with DbPSY, which might be one reason why it could not highly accumulate β-carotene. Overexpression of DsOR, especially the mutant DsORHis, could significantly improve the single-cell carotenoid content and change cell morphology (with larger cell size, bigger plastoglobuli, and fragmented starch granules) of D. salina. Overall, DsPSY1 played a dominant role in carotenoid biosynthesis in D. salina, and DsOR promoted carotenoid accumulation, especially β-carotene via interacting with DsPSY1/2 and regulating the plastid development. Our study provides a new clue for the regulatory mechanism of carotenoid metabolism in Dunaliella. IMPORTANCE Phytoene synthase (PSY) as the key rate-limiting enzyme in carotenoid metabolism can be regulated by various regulators and factors. We found that DsPSY1 played a dominant role in carotenogenesis in the β-carotene-accumulating Dunaliella salina, and two amino acid residues critical in the substrate binding were associated with the functional variance between DsPSY1 and DsPSY2. Orange protein from D. salina (DsOR) can promote carotenoid accumulation via interacting with DsPSY1/2 and regulating the plastid development, which provides new insights into the molecular mechanism of massive accumulation of β-carotene in D. salina.

determining the functional variance between the two PSYs (DsPSY1/2) as well as the role of DsOR in carotenoid metabolism in the b-carotene-overproducing D. salina.

RESULTS AND DISCUSSION
Cloning and functional identification of two phytoene synthase genes from D. salina. According to the draft nuclear genome sequence of Dunaliella salina strain CCAP 19/18 at the phytozome website (31), two phytoene synthase genes, with gene identifiers with Dusal.0943s00001 and Dusal.0031s00032, were identified, and designated DsPSY1 and DsPSY2, respectively. Their full-length coding sequences were 1,278 bp with 425 amino acids (aa). Predicted by Plant-mPLoc, DsPSY1 and DsPSY2 were localized in the chloroplast. The length of the predicted chloroplastic transit peptides of DsPSY1/2 were 40 and 41 amino acid residues at the N terminals, respectively. In addition, we also cloned the PSY gene from Dunaliella sp. FACHB-847 (DbPSY) according to our previous transcriptome data, with the full-length coding sequence (CDS) of 1,305 bp encoding 434 aa (32). The length of the predicted chloroplastic transit peptide of DbPSY was 52 amino acid residues at the N terminals. The protein sequence of DsPSY1 displayed high identities with DbPSY (82.38%) and DsPSY2 (82.33%), with the main sequence differences in the transit peptide regions. Dunaliella PSYs (DsPSY1/2 and DbPSY) showed high homologies with PSYs from other green algae and higher plants, and showed ;50% identities with cyanobacterial crtB and ;30% identities with bacterial crtB (Fig. 1A). Many plants have 2 to 3 PSY homologous genes with tissue expression specificity; for example, tomato has 3 PSY genes, among which SlPSY1 is specifically expressed in fruits and flowers, and SlPSY2 is mainly expressed in leaves (33). Dicot PSY3 including SlPSY3 is strongly expressed upon root colonization by symbiotic arbuscular mycorrhizal fungi for apocarotenoid formation (34), distinct from PSY3 genes in the Poaceae, which can be induced by abiotic stress such as salt and drought (35). However, there is only one PSY gene in Arabidopsis thaliana. In contrast, most algae, including Dunaliella sp. FACHB-847, possess only one PSY gene (36); however, two PSY homologous genes were found in some algae, such as D. salina CCAP 19/18, Ostreococcus, and Micromonas based on the phytozome website (31,37). Algal PSY genes have been found to be induced by multiple stress factors such as high light intensity, blue light, salt stress, and nutrient deficiency (36,38,39).
Furthermore, we used color complementation in the carotenoid-producing E. coli platform to investigate the functional activity of DsPSY1/2 compared with the activity of DbPSY. The orange-colored Ec07 (harboring pAC-BETA) could result in the accumulation of b-carotene (40) (Fig. 2), whereas no carotenoids were observed in the 07 dB strain (carrying pAC-EIYdB) with the disruption of the bacterial crtB gene encoding phytoene synthase. When DsPSY1/2 and DbPSY without chloroplast transit peptides were overexpressed into 07 dB, respectively, the resulting strain 07 dB1Yb could accumulate comparable b-carotene with Ec07, and 07 dB1Y1 could lead to the considerable accumulation of b-carotene, but produced lower b-carotene than 07 dB1Yb. However, almost no b-carotene could be detected in 07 dB1Y2 (Fig. 2). This indicated that DsPSY1 and DbPSY had functional activity of PSY, but DsPSY2 had almost no PSY activity. Apparently, DsPSY1 was less effective in promoting carotenoid biosynthesis than DbPSY. Dunaliella sp. FACHB-847 was rich in lutein, which could be due to this efficient enzyme activity of DbPSY, the bifunctional lycopene band « -cyclases, and multiple pathways for lutein biosynthesis by carotene hydroxylase as well as the high expression of carotenogenic genes (25,30,36).
Two key amino acid residues associated with the functional variance between DsPSY1 and DsPSY2. Searched by NCBI-conserved domains, DsPSY1/2 and DbPSY possessed multiple distinct conserved domains, such as the substrate binding pocket, substrate-Mg 21 binding site, active site lid residues, catalytic residues, and two aspartate-rich

Roles of Two PSYs and/or Protein in Dunaliella Salina
Microbiology Spectrum regions (Fig. 1B). The protein sequences of DsPSY1/2 were highly homologous, but their catalytic efficiencies were distinctly different (Fig. 2). In tomato, fruit-specific SlPSY1 was reported to exhibit significantly weaker carotenogenic activity than green tissue-specific SlPSY2, and it was proposed that a neighboring aromatic-aromatic amino acid combination (Phe-135 and Asn-136 in SlPSY1, Phe-159 and Tyr-160 in SlPSY2) governed activity divergence between these two PSYs from tomato (33). Therefore, we speculated some key amino acid sites leading to the differentiation of functional activity of DsPSY1/2. In order to elucidate the reason for differential catalytic activity of DsPSY1/2, we first predicted a total of 13 amino acid positions (at 121, 136, 144, 147, 214, 234, 249, 251, 281, 282, 284, 285, and 365) that led to harmful mutations of DsPSY1 with high possibilities, through the PolyPhen-2 website (http://genetics.bwh.harvard.edu/pph2/) (Fig. 1B). We also used DbPSY for mutation experiments by testing the change in catalytic activity by amino acid substitution. Combined with the result of sequence alignment and conserved domain, two possible key amino acid sites (F144 and G285 in DsPSY1; H144 and W285 in DsPSY2; F155 and G296 in DbPSY) correlated with substrate binding pocket sites were used for further mutation experiments (Fig. 1B). Accordingly, by site-directed mutagenesis, we generated mutants DsPSY1 F144H , DsPSY1 G285W , DsPSY1 F144H G285W , DsPSY2 H144F , DsPSY2 W285G , DsPSY2 H144F W285G , DbPSY F155H , DbPSY G296W , and DbPSY F155H G296W for mutational analysis by testing catalytic activity. The mutated PSYs coupled with pAC-EIYdB were expressed in E. coli BL21(DE3). As shown in Fig. 2, 07 dB1Y1H, 07 dB1Y1W, and 07 dB1Y1HW carrying the mutant forms of F144H or/and G285W of DsPSY1 could greatly reduce the cell color with almost no detectable b-carotene content, as did 07 dB1YbW and 07 dB1YbHW. F155H in DbPSY also largely decreased the b-carotene content. Although the amino acid substitutions H144F and W285G as well as their double mutation in DsPSY2 did not result in b-carotene accumulation, it was sufficient to suggest that the two amino acid residues at positions 144 and 285 critically determined the catalytic activity difference between DsPSY1 and DsPSY2.
Similar to PSY that catalyzes head-to-head condensation of GGPP to generate phytoene, dehydrosqualene synthase (CrtM) and squalene synthase (SQS) catalyze head-tohead condensation of two molecules of farnesyl pyrophosphate to form dehydrosqualene and squalene, respectively (33). PSY proteins shared conserved core structures such as the hydrophobic flap and Asp-rich motif 1/2 with CrtM and SQS (41) (Fig. 1C). Here, we analyzed the protein structures of Dunaliella PSY proteins by homology modeling using SWISS-MODEL based on the crystal structure of CrtM from Enterococcus hirae ATCC 9790 (PDB: 5IYS) ( Fig. 1D and 3). DsPSY1/2 and DbPSY with high sequence similarities had a similar protein structure with RMSD (root mean square deviation) lower than 3. The higher sequence similarity, the lower the RMSD and the better the protein superposition (Fig. 1D).
As shown in Fig. 3, the backbone and sidechain of amino acid residues responsible for substrate binding pockets (looking like "pockets") were shown in the protein structures, which were highly similar with other PSY enzymes, CrtM, and squalene synthase (SQS) (41,42). F144 and G285 in DsPSY1 were highly conserved with many PSYs from other green algae and higher plants, of which F144 played a critical role in substrate binding, active site, and catalysis. G285 was also particularly important for substrate binding (Fig. 1B). F144 in DsPSY1 was conserved with one of the neighboring aromatic-aromatic amino acids that determined the activity divergence between the two PSYs from tomato (33). Interestingly, the spatial position of residues at 144 and 285 of DsPSY1/2 in 3D protein structure was almost the opposite, and located in the opening of the substrate binding pocket. For DsPSY1 (Fig. 3A), when F144 containing a 6-carbon benzene ring was mutated into H144 with an imidazole side chain, the longer spatial distance was observed between residues at 144 and 285 (Fig. 3B), which may lead to the loose binding between PSY enzyme and substrate, thereby reducing the catalytic efficiency. When G285 with simple molecular structure in DsPSY1 was mutated into W285 with an indole ring, much shorter spatial distance was observed between residues at 144 and 285 (Fig. 3C), which may hinder the substrate binding and result in the great reduction in catalytic activity. Similarly, the narrowing of the opening of the substrate binding pocket (Fig. 3D) may account for the near inactivation of DsPSY2. Besides, multiple amino acid residues at positions 121, 136, 147, 214, 234, 249, 251, 281, 282, 284, and 365 that were potentially harmful for the PSY catalytic activity were present in DsPSY2, which led to a near loss of enzyme activity. It was reported that the amino acid replacement of P192L led to the misfolding of SlPSY1 protein and affected its activity, thus reducing the accumulation of lycopene and other carotenoids in tomato fruits (43). An amino acid mutation of A191D in the PSY gene from cassava could promote the accumulation of provitamin A carotenoids (6). Overall, the key amino acid residues have a great impact on PSY enzyme activity, and the rational design of the PSY gene can be used for production of carotenoid-rich crops (7,8).
Cloning and characterization of orange protein from D. salina. According to the genome data of D. salina, only partial CDS of the orange gene from Dunaliella salina (designated DsOR) could be identified with the gene identifier with Dusal.0142s00018. Based on our transcriptome data of D. salina, the full-length CDS of DsOR with 1,071 bp encoding 356 aa could be cloned. Transcriptome sequencing of Duneliella sp. FACHB-847 found that the full-length CDS of DbOR was 1,044 with 347 aa (32), which displayed 75.76% identities with DsOR. The protein sequence of DsOR showed 50 to 76% identities with ORs from other green algae and higher plants (Fig. 4A). DsOR belonged to the DnaJ-like zinc finger domaincontaining protein family and DnaJ_zf superfamily, and shared the highly conserved quadruple repeat of the CxxCxxxG signatures at the C terminal (Fig. 4B). Analyzed by Phyre2, DsOR and DbOR were predicted as a membrane protein with two transmembrane helices (Fig. 4B). DsOR was predicted to be localized in the chloroplast and mitochondrion by Plant-mPLoc prediction, and localized in the nucleus by LOCALIZER prediction. In plants, the multifunctional OR proteins have been reported to be localized in different subcellular localizations, including nucleus and plastids (44,45). Predicted by LOCALIZER, a putative chloroplast transit peptide and a nuclear localization signal (RRNKIFLMMEEVRRLRI) were present in DsOR and DbOR (Fig. 4B). An arginine to histidine "golden SNP" has been shown to contribute to large amounts of carotenoids (18). Accordingly, a highly conserved arginine at

Roles of Two PSYs and/or Protein in Dunaliella Salina
Microbiology Spectrum positions 129 and 115 was found in DsOR and DbOR (Fig. 4C), respectively, and we constructed the mutants DsOR His (R129H) and DbOR His (R115H) for further overexpression experiments. DsOR interacted directly with DsPSY1/2. To confirm the physical interaction between Dunaliella OR and PSY, we performed protein interaction analysis by the matingbased split ubiquitin system (mbSUS) of yeast two-hybrid assay (46), as OR was a transmembrane protein. The full-length PSY coding sequence was fused to NubG, a mutant of the N-terminal moiety of ubiquitin (Nub) with severely decreased affinity for the C-terminal moiety of ubiquitin (Cub), and OR was fused to Cub. As shown in Fig. 4D, all yeast transformants grew on synthetically defined (SD) medium lacking leucine, tryptophan, and methionine (SD/-Leu/-Trp/-Met), indicating that exogenous PSY are not toxic on yeast. Furthermore, yeast cells coexpressing NubG-DsPSY1/2 and DsOR-Cub could grow on SD medium lacking leucine, tryptophan, histidine, and adenine (SD/-Leu/-Trp/-His/-Ade) containing 75 and 400 mM methionine (Met), which was consistent with the positive control, but no growth was observed on the same media when there was coexpression with NubG-DsPSY1/21Cub and NubG1DsOR-Cub, suggesting that DsOR could interact with DsPSY1/2. No growth was also observed on the same media when there was coexpression with NubG-DbPSY1DbOR-Cub, suggesting no interaction between DbOR and DbPSY, which may be due to the fact of the non-b-carotene-accumulating Dunaliella sp. FACHB-847 (25). Clearly, the mbSUS yeast two-hybrid assay demonstrated that DsOR interacted with DsPSY1 and DsPSY2 without significant difference, implying the key role of DsOR protein in regulating carotenoid metabolism in D. salina.
Previous study showed that two Arabidopsis OR proteins (AtOR and AtOR-like) as well as the Brassica oleracea OR protein (BoOR) interacted directly with PSY (12). Interestingly, differential interactions of OR proteins with the PSY enzymes were observed in saffron as CsOR-a displayed strong interactions with CsPSY1a, CsPSY1b, and CsPSY3, and weak interaction with CsPSY2, and in contrast, no interactions were observed between CsOR-b and the CsPSY enzymes (47). It was showed that the OR protein possessed an N-terminal region responsible for interaction with PSY and a C-terminal cysteine-rich zinc finger domain
Heterologous expression of DsOR in carotenoid-producing E. coli. Previous studies showed that overexpression of OR, especially OR His , genes in plants and green alga C. reinhardtii could significantly enhance the accumulation of carotenoids (16 to 18, 23). To investigate the effect of Dunaliella OR protein on PSY activity and the role of OR protein in regulating carotenoid metabolism, we coexpressed Dunaliella OR expressed OR His in E. coli carrying crtEIY and Dunaliella PSY genes. As shown in Fig. 2, for the expression of DsOR or DsOR His genes, the b-carotene contents of 07 dB1Y1-DsOR and 07 dB1Y2-DsOR had slight improvement compared with 07 dB1Y1 and 07 dB1Y2, and more b-carotene contents were detected in 07 dB1Y1-DsORH and 07 dB1Y2-DsORH, indicating that DsOR and DsOR His can promote carotenoid accumulation, which was quite in line with the previous report (53). Interestingly, no significant change was observed in 07 dB1Yb-DbOR and 07 dB1Yb-DbORH compared with 07 dB1Yb, which may be due to the fact of the nonb-carotene-accumulating Dunaliella sp. FACHB-847 (25,30). In addition, coexpression of OR or the OR His mutant in b-carotene-producing E. coli carrying crtEIBY genes also could not significantly affect the b-carotene levels compared with Ec07, suggesting that OR protein could not act on bacterial crtB, which was consistent with the previous report (53). Apparently, our study demonstrated that DsOR plays a crucial role in promoting carotenoid biosynthesis.
Overexpression of DsOR/DsOR His enhanced b-carotene accumulation in D. salina. By using efficient endogenous promoters in D. salina and the fusion of bleomycin-resistant gene Ble and the enhanced green fluorescent protein EGFP as reporter, the overexpression vectors, pCROE-DsOR, and pCROE-DsORHis, were constructed (Fig. 5A). The electrotransformation technology was used to transform these two overexpression vectors into D. salina cells, respectively. Three days after transformation, confocal laser scanning microscopy was used to observe the transgenic D. salina cells, and it was found that the cells after transforming the overexpressed vector showed green fluorescence (Fig. 5B), indicating that the overexpression vector was successfully transformed into D. salina cells. Meanwhile, bleomycin-resistant plates were used to screen the monoclonal algal transformants. Subsequently, two OE-DsOR strains (DsOR#1 and DsOR#9) overexpressing DsOR and one OE-DsORHis strain (DsORHis#9) overexpressing DsOR His were identified by direct PCR using algal cells as the template and confirmed by PCR using algal genomic DNA as the template (Fig. 5C).
Phenotypic identification and carotenoid detection ( Fig. 5D; Table S1 at https:// www.biosynnatlab.com/wp-content/uploads/2023/03/Supporting-Information2023OR .pdf) showed that the chlorophyll and carotenoid contents per cell of the transformants at day 25 were higher than that of the control, of which DsOR#1 and DsOR#9 showed 22.9 to 30.7% higher single-cell carotenoids than the control, and DsORHis-9 had 73.4% higher than the control. All-trans b-carotene and 9-cis b-carotene at the ratio of about 1:1 were the main carotenoids (accounting for .83%) in D. salina cells, which was consistent with a previous report (27). D. salina cells have turned red at day 25, and the carotenoid-to-chlorophyll ratio was 3.2 in the control D. salina cells and 3.5 to 4.4 in the transformants, which was in accordance with the previous report (54). Observed by light microscope and transmission electron microscope, all the transformants (DsOR#1, DsOR#9, and DsORHis#9) possessed larger and darker orange cells compared with the control, especially at the later stage of growth (Fig. 6). Interestingly, the transformants not only changed the cell size (enlarged cells), but also increased the size of plastoglobuli and changed the distribution of pyrenoid and starch granules (Fig. 6). Normally, orange Dunaliella cells contain a single cup-shaped chloroplast with uniform size of many plastoglobuli as well as a central pyrenoid surrounded by starch granules (55,56). Due to the lack of a cell wall in D. salina, some cells tended to crack after enlarging (Fig. 6). The growth state of the transformants with low cell density was not as good as that of the control. This may be because the transformants with larger cells needed higher energy and more carbon sources, resulting in slightly slower cell growth. Also, pyrenoid is a subcellular structure that colocalizes CO 2 -concentrating mechanism and the ribulose-1,5-bisphosphate carboxylase/oxygenase (RuBisCo) enzyme, thus maximizing CO 2 fixation in Dunaliella (56,57). Therefore, the change in the distribution of pyrenoid and starch granules may influence the efficiency of CO 2 concentration and photosynthesis. In conclusion, overexpression of the DsOR or DsOR His gene could significantly improve the single-cell carotenoid content of D. salina, with much more

Roles of Two PSYs and/or Protein in Dunaliella Salina
Microbiology Spectrum single-cell carotenoids in D. salina overexpressing DsOR His , which was well consistent with the OR Rubisco OR His transgenic lines of Chlamydomonas reinhardtii (23), tomato (17), and melon (18). Major conclusion. In the b-carotene-accumulating D. salina, the efficient DsPSY1 played a dominant role in carotenogenesis, and DsPSY2 almost had no catalytic activity. Two key amino acid residues at positions 144 and 285 in the conserved domains were correlated with the functional variance between DsPSY1 and DsPSY2. DsOR could promote carotenoid accumulation via interacting with DsPSY1/2 and regulating plastid development, which provides important clues for the molecular mechanism of carotenoid metabolism regulation in D. salina. Overexpression of DsOR and DsOR His genes not only increased the single-cell carotenoid accumulation, but also enlarged cells, increased the size of plastoglobuli, and changed the distribution of pyrenoid and starch granules. In the lutein-rich Dunaliella sp. FACHB-847, DbPSY had more efficient catalytic activity than DsPSY1, but DbOR could not interact with DbPSY, which provides an explanation for the non-b-carotene overproduction in Dunaliella sp. FACHB-847. The rational design of highly efficient PSY, the regulatory factors or proteins for finetuned regulation of PSY expression, and the molecular mechanism of massive accumulation of b-carotene in D. salina need to be further explored.

MATERIALS AND METHODS
Algal strains and cultivation conditions. Dunaliella salina CCAP 19/18 was obtained from the Culture Collection of Algae and Protozoa (CCAP), Scotland, United Kingdom. D. salina was cultivated in 2ASW (artificial seawater) medium containing 1.5 M NaCl (Table S2 at

Roles of Two PSYs and/or Protein in Dunaliella Salina
Microbiology Spectrum hemocytometer. The genomic DNA of Dunaliella cells was extracted by EZNA HP Plant DNA kit (Omega Bio-Tek, USA). Gene mining and bioinformatics analysis. Ten milliliters of algal cells collected at the late log phase were used for RNA isolation by RNAiso Plus (TaKaRa, Japan). Then, the complementary DNA (cDNA) was synthesized by RevertAid First Strand cDNA Synthesis kit (Thermo Scientific, USA). Genes from Dunaliella cells could be cloned by using the above cDNA template and PrimeSTAR HS DNA polymerase (TaKaRa, Japan), and primers (Table S3 at https://www.biosynnatlab.com/wp-content/uploads/2023/03/Supporting -Information2023OR.pdf) designed according to the genome or transcriptome data (31,32).
The CDS of Dunaliella OR or OR His was subcloned into the EcoRV-XhoI site of pCDFDuet-1 or vectors carrying the PSY gene (Table S4 at https://www.biosynnatlab.com/wp-content/uploads/2023/03/Supporting -Information2023OR.pdf). The plasmids carrying PSY and OR (or OR His ) genes were coupled with 07 dB plasmid for mutational analysis by color complementation, so as to determine the role of OR in affecting carotenoid biosynthesis. The 30-mL overnight culture of the engineered E. coli strain was transferred into 120-mL LB medium with the appropriate antibiotics at 37°C, 220 rpm until the OD 600 was 0.6. Then the culture was induced by 1.0 mM isopropyl b-D-1-thiogalactopyranoside (IPTG) for 48 h at 30°C, 200 rpm.
Extraction and analysis of carotenoids. Carotenoids were extracted from E. coli cells or algal cells as previously reported (25), and carotenoids were detected by high-performance liquid chromatography (HPLC) according to our previous reports (30).
Mating-based split ubiquitin system (mbSUS) assay for protein interaction. The interactions between Dunaliella OR and PSY were determined by mbSUS yeast two-hybrid as described previously (46). The Dunaliella OR gene was cloned into the HindIII-PstI site of the pMetYCgate vector (for fusion proteins with Cub) to construct the bait vectors pMetYCgate-DsOR and pMetYCgate-DbOR, and then transformed into the yeast haploid strain THY.AP4. Similarly, the coding sequences of DsPSY1/2 and DbPSY without the stop codon were cloned into the EcoRI-SmaI site of pNXgate32-3HA (to express fusion proteins with Nub) to construct the prey vectors pNXgate32-DsPSY1/2-3HA and pNXgate32-DbPSY-3HA, respectively (for primers, see Table S3 at https://www.biosynnatlab.com/wp-content/uploads/2023/03/Supporting-Information2023OR .pdf). The bait vector and prey vector were then cotransformed into the yeast strain THY.AP4. Protein-protein interactions were tested by the growth of diploid yeasts on SD/-Leu/-Trp/-Met and SD/-Leu/-Trp/-His/-Ade with 0-, 75-, and 400-mM methionine (Met) at 30°C for 3 to 6 days.
Transmission electron microscopy (TEM). Ten milliliters of algal sample were centrifuged (3,500 g, 10 min), fixed overnight at 4°C using 2.5% glutaraldehyde, then postfixed in 1% osmium tetroxide for 2 h, and finally washed three times using phosphate-buffered saline (PBS). Subsequently, the samples were dehydrated with gradient ethanol solutions of 30%, 50%, 70%, 80%, 90%, 95%, and 100% for 15 min each time, then dehydrated with pure acetone for 20 min. After dehydration, the samples were embedded with epoxy resin. Ultrathin sections of 70 to 90 nm in thickness were cut using an ultramicrotome (EM UC7, Leica, Germany), then stained with lead citrate and uranyl acetate, and finally examined under a Hitachi transmission electron microscope (HT7700, Tokyo, Japan) at 75 kV.
Overexpression of DsOR/DsOR His in D. salina. For vector construction (primers seen in Table S3 at https://www.biosynnatlab.com/wp-content/uploads/2023/03/Supporting-Information2023OR.pdf), the fragments of DsLcyB (lycopene b-cyclase gene from D. salina) promoter (40), DsOR or DsOR His genes, and DsLcyB terminator were subcloned into the XhoI-XbaI site of the pZBET vector mentioned in our previous report, which carried the promoter and terminator of the Dunaliella j -carotene desaturase gene (DbZDS) and the fusion of Ble-EGFP as reporter (61), by the using pEASY-Basic Seamless Cloning and Assembly kit (Beijing TransGen Biotech). The overexpression vectors pCROE-DsOR and pCROE-DsORHis were then constructed. The electroporation procedure for introduction of foreign genes in D. salina cells and the selection of positive transformants were reported previously (40,62). The primers (BE-For and BE-Rev) for selection of the positive transformants are shown in Table S3 (https://www.biosynnatlab .com/wp-content/uploads/2023/03/Supporting-Information2023OR.pdf). The fluorescent signal of EGFP in the transformants was observed by using confocal laser scanning microscope (ZEISS LSM710, Germany) after 3 days of transformation.
Statistical analysis. All experiments were performed in triplicate. Data were expressed as means 6 SD (standard deviation). The significance of differences between groups was assessed by one-way analyses of variance or t test using IBM SPSS statistics 22. P , 0.05 indicated the presence of a statistically significant difference, and P , 0.01 was considered highly significant.