H. pluvialis OUC H2 used in this experiment was obtained from the Algae Genetics and Breeding Laboratory of Ocean University of China. Green growth stage: Bold Basal Medium was used. Light intensity was set as 30 μmol·m⁻²·s⁻¹ (light/dark, 12 h/12 h), the temperature was 23 ± 1°C. Red induction stage: BBM was replaced by oligotrophic medium containing NaAc and Fe²⁺, the light intensity was 195 μmol·m⁻²·s⁻¹ (light/dark, 12 h/12 h) and temperature was 23 ± 1°C. C. reinhardtii CC849 was obtained from Chlamydomonas Resource Center, University of Minnesota, United States. Alga was cultured by TAP (Tris Acetate Phosphate) medium. The light intensity was 50 μmol·m⁻²·s⁻¹ (light/dark, 12 h/12 h), the temperature was 23 ± 1°C, 100 rpm.

Total DNA of H. pluvialis OUC H2 was extracted by using Takara MiniBest Plant Genomic DNA Extraction Kit after grinding with liquid nitrogen. C. reinhardtii was shaken with glass beads at 3000 rpm for 5 min to break cells, and then DNA was extracted according to the instructions of the same kit. E.Z.N.A plant RNA Kit (Omega) was used to extract RNA, and then reverse transcription was performed by using 5 × PrimeScript RT Master Mix (Perfect Real Time, Takara) to construct cDNA library. Gel purification was carried out with Gel Extraction Kit (200) (Omega). Plasmid Miniprep Purification Kit (GMbiolab) was used to extract plasmids from Escherichia coli.

The primers for cloning BKT, BCH, MECR are displayed in Table S1. The cDNA and DNA of H. pluvialis OUC H2 was used as template. TA cloning was performed to construct cloning vector (Tsingke Biotechnology, China) for sequencing (Sangon Biotech, China). The plasmid pUC19 was used as a transition vector to add promoter and terminator, and then the expression frame was linked into the C. reinhardtii expression vector pHyg3. All three genes use β-2 tubulin promoter (P_β2-Tub) and rbcS2 terminator (T_rbcS2) from pHyg3. The sequences were constructed into the vector by enzyme digestion and ligation. The restriction enzymes used are listed in the Table S1. The expression vectors carrying different genes are, respectively, named as pHyg3-BKT-BCH and pHyg3-BKT-BCH-MECR.

The alga was centrifuged (4,000 g, 10 min) and suspended by fresh TAP medium to prepare cell suspension with cell density of 10⁸ cells/ml. 400 μl algal cells were mixed with 5–10 μg plasmid DNA and 400 mg sterile glass bead, and then vortexed for 25–45 s. Then fresh TAP medium was inoculated and the algal cells were cultured under dark environment of 23°C and 50 rpm agitation for 18 h. After centrifuged the recovered alga (4,000 g, 10 min) and added 400 μl fresh TAP medium, 100 μl of culture was spread on the resistant plate (with 10 μg/ml hygromycin B) (21). Positive algal colonies were observed at about 15 days later. Because the glass bead-based transformation of a foreign vector carrying the foreign gene(s) is generally randomly inserted in the genome, at least 20 clones were selected for further research. The algae strains with good growth state, stable gene integration and high abundance of expressed products were used for functional studies.

Two groups of different endonucleases were used to digest genomic DNA (37°C, 3 h). The information of genes and enzymes is listed in Table S2. The digestion products were separated by agarose gel electrophoresis, then denatured solution (1.5 mol/l NaCl，0.5 mol/l NaOH) was added to break the double strand. And then, the denatured nucleic acid was transferred to nylon membrane (0.22 μm, Pall, United States) for fixation after washing with neutralization solution (0.5 mol/l Tris–HCl，1.5 mol/l NaCl). Then the digoxigenin labeled probe was prepared by Random Priming method and hybridized with genomic DNA. The primers used to prepare the probe are listed in Table S3.

Transformants and wild type were cultured at normal or high light intensity for 9 days, and the RNA was extracted every 3 days. The primers used for amplification of exogenous BCH, BKT, MECR were designed by the gene sequences from H. pluvialis (Table S4). PCR products were then quantified continuously with the BIOER LineGene 9,640 using TB Green Premix Ex Taq II (Takara) according to manufacturer’s instructions. After heating at 95°C for 30 s, cycling parameters were: 40 cycles of 95°C for 5 s, 56°C for 20 s. Finally, the specificity of the qRT-PCR products was confirmed by performing a melting temperature analysis at temperatures ranging from 60°C to 95°C at 4°C/s. Transcription levels of the target genes were calculated by the 2^−ΔΔCT method. To standardize the results, the relative abundance of RCK gene (G protein beta subunit) was also determined and used as the internal standard.

Cell density: The absorbance of the algal solution was measured to calculate the cell density according to the standard curve obtained in the previous: y = OD₇₅₀ × 1.2335 × 10⁷ (cells/ml). Dry cell weight: The algal liquid was centrifuged for 10 min at 4000 rpm. The algal cell precipitation was washed with ddH₂O twice. The clean algal mud was dried in a freeze-dryer (Shanghai Bilang Instrument Manufacturing) for 6–7 h until the algal mud was completely dry and powdered. Then, the weight of algal powder was measured.

Total carotenoids from C. reinhardtii were extracted by using 80% of acetone as previously described (22). Acetone was added to the algal powder. After ultrasonic crushing, 0.22 μm filter membrane was used for filtration. Then the samples were analyzed by HPLC using Thermo UItiMate 3,000 UHPLC chromatograph (Thermo, United States). The analytical conditions: YMC Carotenoid S-5 μm column (4.6 × 250 mm, 5 μm), flow rate = 1 ml/min, detection, 450 nm. Mobile phase: A: MeOH: MTBE: H₂O (81:15:4); B: MeOH: MTBE (6.5:93.5). The sample size was 5 μl. The determination of carotenoids was achieved by the comparison of characteristic absorption spectrum and reduced retention time with standards. And they were quantitatively determined by external standard method. 100 ml methanol was used to dissolve lutein (20 mg), zeaxanthin (5 mg), β- Cryptoxanthin (40 mg), α- Carotene (5 mg), β- Carotene (20 mg), neoxanthin (10 mg), violaxanthin (10 mg). The standard solution was mixed and use 0.1% BHT-acetonitrile: methanol: dichloromethane (75,20:5) to prepare 1/100, 5/100, 10/100, 20/100, 40/100, 60/100, 80/100 μg/ml of working fluid. Astaxanthin standard was prepared with concentration gradient of 0.1, 0.2, 0.5, 1, 2, 5 μg/ml. Carotenoid content of sample (μg/g) = (c × V × F) / m, where c is the concentration of carotenoids in the sample calculated according to the standard curve, V is the volume of the extract, m is the sample weight of about 0.2 g, and F is the dilution factor.

To determine the astaxanthin content in cells, 5 mg of algal powder was resuspended with 5 ml KOH/methanol solution, and then heated at 60°C for 5 min to destroy chlorophyll and convert astaxanthin ester into free astaxanthin. After heating, the algal cells were centrifugated at 4000 rpm for 10 min. After discarding the supernatant, 2 ml DMSO was add to the precipitation for extracting astaxanthin, and the extraction was repeated until the precipitation turns white. The supernatant was combined and diluted to a certain volume with acetone. The absorbance was measured at 474 nm. Finally, the astaxanthin content was calculated according to the formula P = AV/ (10b ε m). In the formula, A is the absorbance at 474 nm of the sample to be measured. V is the constant volume (ml). b is the length of the optical path of the colorimetric cell, 1 cm. ε is the light absorption coefficient of astaxanthin, 0.191 cm²/μg. m is the mass of algal powder (mg).

3 ml chloroform methanol (v/v, 2/1) mixture was added into 20 mg of algae powder, and fully mixed. Ultrasonic crushing for 15 min, centrifugation of 10,000 g for 10 min, the supernatant was transfer to another centrifuge tube, and repeated twice. 3 ml of tri distilled water was added into the supernatant, and centrifuged 10,000 g for 10 min for stratification. The lower layer of liquid was taken into a glass bottle that has been weighed (m1), and placed in the fume hood until the chloroform volatilizes completely to constant weight. The total mass of the glass bottle and the algae lipid (m2) was weighted, and the total fat content ω was calculated. The calculation formula is as follows: ω = (m2-m1)/20 × 100%. Fatty Acid Methylation: The lipid was saponified with KOH-methanol solution to obtain fatty acid. Under the catalysis of BF₃, fatty acids and methanol formed methyl esters, which are extracted with n-hexane for gas chromatography–mass spectrometry analysis (MassHunter GC/MS, Agilent). The full scan method of GC–MS was used for detection, the NIST standard mass spectrometry library ratio method was used for qualitative analysis, and the peak area percentage method was used for relative percentage content quantification. Gas chromatographic conditions: chromatographic column, DB-INNOWax quartz capillary column (30 m × 0.25 mm × 0.25 μm). The initial temperature of the chromatographic column is 50°C, which shall be kept for 1 min, and the temperature shall be raised to 250°C at 10°C/min, which shall be kept for 5 min. The sample inlet temperature is 250°C. The carrier gas is high-purity helium, in constant flow mode, with a flow rate of 1.0 ml/min. Injection volume is 1 μl. Mass spectrometry analysis conditions: ion source EI. Ion source energy 70 eV. The ion source temperature is 250°C. The scanning mode is full scanning. The mass scanning range is 50 ~ 500 amu. The solvent is delayed for 3 min.

SPSS and Excel software were used for data analysis. The data were obtained as the mean value ± SD. Significant differences were statistically analyzed by ANOVA. Statistical significance was defined as value of p <0.05. All experiments were performed at least three biological replicates to ensure reproducibility.

The structure of expression vector is shown in Figure 1A. The vector pHyg3 was containing the C. reinhardtii β-2 tubulin promoter (P_β2-Tub), rbcS2 gene intron 1, streptomyces hygroscopicus aminoglycoside phosphotransferase gene (aph7) and rbcS2 3′ untranslated region (T_rbcS2). This cassette conferred a resistance against hygromycin B. The P_β2-Tub and T_rbcS2 were cloned to construct a cassette, then BKT and BCH were introduced into the cassette, respectively. Then BKT-cassette and BCH-cassette were inserted into vector pHyg3 using the enzyme sites as shown in Table S1 to get the vector pHyg3-BKT-BCH. All exogenous genes are regulated by C. reinhardtii β-2 tubulin promoter and rbcS2 terminator. The cells grown on the selection medium containing hygromycin (10 μg/ml) were termed as transformants. In order to confirm the stable inheritance of the transformed genes, the transformants were continuously cultured for 10 generations under selective conditions (Figure 1B).

The finally obtained transformant was named as T1 and the untransformed strain CC849 was regarded as WT. Genomic DNA of T1 and WT was extracted for southern blotting analysis (Figure 1C). After DNA was digested with SacI/HindIII and SalI/BamHI enzymes, it was hybridized with the probe of BKT gene. A clear single band appeared in T1. After DNA was digested by MluI/BglII and XhoI/NdeI enzymes respectively, clear bands also appeared in the hybridization with the probe of BCH gene. The untransformed algal strains were digested by same enzymes and hybridized with same probes, but no bands appeared, indicating that the transformed BKT and BCH were integrated into the genomic DNA of T1.

Real time fluorescent quantitative PCR was used to detect the transcription of foreign genes in T1 under 50 μmol·m⁻²·s⁻¹ and 195 μmol·m⁻²·s⁻¹ light intensity. The transcriptional level on the 3rd, 6th and 9th day were analyzed using the transcription level on the first day as the control. In general, BKT and BCH were successfully expressed in cells. Under normal light, the expression of BKT was higher than that of BCH, and the expression of both genes decreased with time. Under high light condition, the transcription level of each gene has been greatly improved. BKT gene was always at a high transcription level under high light intensity, and the BCH and MECR gene expression was significantly higher on the third day of high light treatment (Figure 2).

Carotenoids in T1 and WT were measured by HPLC (Figure 3). Compared with WT, neoxanthin, zeaxanthin and astaxanthin in T1 are significantly increased under normal light and high light. Violaxanthin, α-carotene decreased significantly. β-cryptoxanthin increased significantly under normal light and decreased significantly under high light. Lutein did not change significantly under normal light, but decreased significantly under high light. β-carotene increased significantly under normal light, but did not change significantly under high light. In the carotenoid metabolic pathway, zeaxanthin is the common precursor of astaxanthin and violaxanthin. Therefore, we speculate that the expression of BKT and BCH leads to the redistribution of carbon flow, and more astaxanthin was synthesized, resulting in an increase of the synthesis of its precursors zeaxanthin. Similarly, the reduction of α-carotene may also be due to the redistribution of carbon flow. In general, high light led to the increase of carotenoid content, which indicates that there is a mechanism to avoid light damage by increasing carotenoid content in C. reinhardtii, which seems to be common in green algae. It is worth noting that astaxanthin content in WT can hardly be detected, which is consistent with previous studies. However, the significant increase of astaxanthin in T1 indicates that T1 can be used as a platform strain for later experiments.

On the basis of vector pHyg3-BKT-BCH, MECR gene was ligated with P_β2-Tub and T_rbcS2 and then inserted into vector phyg3-BKT-BCH to get the vector pHyg3-BKT-BCH-MECR (Figure 4A). The final transformants are named as T2. Similarly, southern blotting was used to analyze the integration results of foreign genes. The genomic DNA of WT and T2 was digested by two groups of enzymes (XbaI/BamHI and NdeI/MluI). One band appeared in WT digested with XbaI/BamHI, and two bands appeared in T2 digested with the same enzyme, one of which was in the same position as WT. It was speculated that it might be a non-specific hybridization, and the other band was unique to T2, which should be a specific hybridization band of MECR gene. WT digested with NdeI/MluI showed no band, while T2 digested with NdeI/MluI showed a specific band. The above results demonstrate that MECR gene has been successfully integrated into T2 genome (Figure 4B). qPCR results also confirmed the successful expression of MECR gene (Figure 2). After obtaining appropriate transformants, the total lipid content in the transformants and WT was detected (Figure 5). The results showed that the total lipid in T1 which only transformed BKT and BCH was significantly higher than that in the WT, which indicated that the increase of astaxanthin promote the lipid accumulation. This result shows that the metabolic correlation between lipid and astaxanthin is mutual. In addition, on the basis of T1, the total lipid content of T2 strain transformed with MECR gene was significantly higher than that of T1. According to these results, MECR has been shown to promote total lipid.

In order to determine the effect of different genes on the fatty acids in the transformants, GC–MS was used to measure the relative quantity of different fatty acids (Figure 6). The results showed that the fatty acids in WT were mainly hexadecanoic acid, octadecanoic acid and eicosanoic acid, including monounsaturated fatty acids and polyunsaturated fatty acids. The main fatty acids are hexadecanoic acid (Palmitic acid), 9,12,15-octadecatienoic acid (Linolenic acid) and 5,8,11,14,17-eicosapentaenoic acid (EPA), with the contents of 22.46, 20.95 and 21.61%, respectively. In addition, there were other unsaturated fatty acids with high nutritional value, such as 5,8,11,14-eicosatetraenoic (arachidonic acid). Overall, the composition of fatty acids of the transformants did not change compared with WT. The foreign gene did not change the type of fatty acids in C. reinhardtii, but the relative content of different fatty acids in the transformants changed. We compared the differences of fatty acid categories between transformants and WT. The results showed that monounsaturated fatty acids in T1 and T2 were significantly lower than those in wild type, but polyunsaturated fatty acids were significantly higher than those in wild type. Moreover, the content of monounsaturated fatty acids in T2 was higher than T1 (Figures 7A,B). In the comparison of fatty acids with different chain lengths, there was no significant difference in C16 between WT and transformants. The content of C18 in transformants was lower than that of WT, but T2 was higher than that of T1. The content of C20 in T1 and T2 was significantly higher than that in WT (Figures 7C,D,E). The above results indicate that MECR gene affect directly or indirectly the fatty acid elongation and desaturation metabolic pathway in C. reinhardtii. In addition, the change of fatty acid type in T1 further proves the influence of astaxanthin metabolism on lipid metabolism mentioned above.

Based on the above research, we have successfully expressed H. pluvialis genes that catalyze astaxanthin synthesis (BKT, BCH) and fatty acid elongation (MECR) in C. reinhardtii. Next, the effect of lipid increases and fatty acid changes on astaxanthin production were examined. The mass percentage of astaxanthin in transformants and WT was measured, as shown in Figure 8. We found that the astaxanthin content in T1 was significantly higher than that in WT, which was consistent with the liquid chromatography results in 3.1, with an increase of nearly 50%. In addition, the astaxanthin content in T2 was significantly higher than that in T1, which was 227.5%higher than that of wild type. These results indicate that the promotion of lipid production has indeed increased astaxanthin production. It is feasible to promote astaxanthin accumulation by transforming foreign gene MECR.

In production, the growth of microalgae is a key indicator. We measured the growth curve and dry weight to characterize the growth of WT and transformant. The results showed that WT and transformants accumulated more than 50% biomass within 5 days after inoculation, then the growth rate slowed down and got into the platform period after 2 weeks of culture. However, we found that the biomass of the three transformants was significantly lower than that of WT, and there was no significant difference between different transformants (Figure 9). We measured the dry cell weight of algae after 15 days of culture, and the results were consistent with the growth curve. The dry weight of wild type was significantly higher than that of all transformants. And the dry weight of T2 is slightly higher than that of T1 (Figure 10).