Macular pigment-enriched oil production from genome-edited microalgae

Since dZAs generated serial mutations, we determined the effect of the AGP mutation on the growth of dZAs using growth analysis and microscopic fluorescence analysis. The growth patterns of dZA1 and dZA2 were almost identical to those of the parental zep line and wild type (Fig. 1C). This indicates that the edited lesions of the AGP gene in dZAs does not have any detrimental effect on the growth of dZAs and increases the oil content. To validate the higher TAG accumulation in the dZAs, oil droplets in the cells were visualized through microscopic fluorescence analysis and compared with the parental strains (Fig. 1D). Further, for the quantification of triacylglycerol (TAG), we measured the Nile red fluorescence of each cell line growing under nitrogen starvation conditions and normalized this to the same cell numbers. The two dZA strains displayed more than two-fold higher Nile red fluorescence than the parental strains (Fig. 1E). These results showed that the dZAs exhibited increased TAG synthesis through metabolite changes when AGP was knocked out.

To quantify starch synthesis, dried biomass was hydrolysed using an amyloglucosidase/α-amylase combination method, and the starch was quantified by the amount of dissociated glucose as explained in the Materials and Methods section. The starch content based on a DCW was approximately 35% in the wild-type and zep line, and approximately 2% in the dZAs (Fig. 2B), which was consistent with the results of iodine staining (Fig. 1A). Total lipid analysis showed an increase in lipid content in the dZAs with 450.0 mg g⁻¹ DCW in dZA1 and 505.6 mg g⁻¹ DCW in dZA2 (Fig. 2C). Whereas, the total lipid content of the wild type and zep was 221.3 mg g⁻¹ DCW and 254.3 mg g⁻¹ DCW, respectively (Fig. 2C). Such enhancements of lipid content in dZAs indicated that carbon flux to starch production had been changed to fatty acid biosynthesis owing to the defective starch biosynthesis shown in AGP knockout mutants (dZAs) under N-starvation conditions.

The culture method using this medium was called one-step oil induction process (one-step process), in which cell growth and oil induction occurred in the same medium. Lipid and pigment production from dZA1 were compared between one-step process and the conventional two-step process. Cells growing in the one-step culture condition reached the stationary phase slowly, and the final cell density at 4 days in one-step culture was almost 2/3 than in the normal media (Fig. 1C; Additional file 1: Fig. S1A). In contrast, cells in a two-step process reached the stationary phase fast at normal medium but maintained the cell density during the N-deprivation for 3 more days (Additional file 1: Fig. S1B). Interestingly, at the end of cultivation, the lipid content of dZA1 and dZA2 in the one-step process was enhanced by 1.6-fold and 1.9-fold, respectively, compared to that in the parental strain (zep), based on dry weight (Additional file 1: Table S1). However, when we compared the lipid productivity of dZA strains between one-step and two-step processes, the average value of lipid productivity (144.51 mg/g/day) for dZAs in a one-step process was 1.8-fold that of the two-step process (79.64 mg/g/day) (Fig. 4C).

Pigment production by dZA mutants under oil induction conditions was compared with that of the parental line (zep). The average value of the macular pigment content of dZAs in the one-step process was 38% higher than that of the parental line (zep), but this value in the two-step process was 23% higher than that of zep (Additional file 1: Table S1). Further, the macular pigment productivity of dZAs between the two processes was compared, and the average value of pigment productivity in the one-step process was slightly higher or at least the same as that in the two-step process (Additional file 1: Table S1; Fig. 4D).

A one-step process consisting of reduced nitrogen and phosphate provided by the 1/4NP2A medium resulted in slower cell growth compared to that in normal medium (Fig. 1C and Additional file 1: Fig. S1A). However, for the production of oil and pigment together, this process can shorten the culture time by approximately 33% compared to that with the N-starvation process, and medium replacement is not required; thus, without losing pigments, this one-step process enables the production of pigment-enriched microalgae oil for possible food applications.

The linear term for time and ratio for lipids, the linear and quadratic terms for the ratio, the interaction ratio with time for lutein, and the linear term for the ratio for zeaxanthin were significant. The p value for the lack-of-fit for the three dependent variables was much higher than 0.05, which indicated that the regression model correctly described the experimental data. The extraction of pigments was not significantly affected by the extraction temperature and time, similar to the findings of a previous study [39]. Unlike in previous studies [40, 41], the temperature did not affect the extraction of lipids. This phenomenon could be explained by the presence of isopropanol in this study, which helps to diffuse lipids from the cells. There was an increasing tendency for the dependent variables with increasing extraction time and solvent-to-biomass ratio. From the response surface analysis, we determined that the optimal conditions were 35.7 °C for extraction temperature, 150 min for extraction time, and the ratio of solvent to solid of 100 mL g⁻¹.

To confirm the actual product, edible microalgal oil from dZA1 was produced using the optimized methods described above. dZA1 was cultivated under one-step cultivation conditions using a 1/4NP2A medium. The biomass yield of dZA1 in 1/4NP2A medium was approximately 70% that in the normal medium (Fig. 1C; Additional file 1: Fig. S1A). According to previous reports, the biomass productivity or cell growth of the starchless strain, such as dZA1, could be affected by the reduction of starch synthesis by AGP knockout under nitrogen-starvation conditions [14, 34, 42]. Under the same conditions, total lipid and pigment production were confirmed, and the productivity was compared to that of the zep. Lipid productivity of dZA1 was 131.53 ± 16.62 mg/g/day, which was 69% higher than that of zep (77.87 ± 8.72 mg/g/day). Pigment productivity of dZA1 (lutein plus zeaxanthin: 0.90 mg/g/day) was also 29% higher than that of zep (0.7 mg/g/day) (Fig. 4C, D; Additional file 1: Table S1). This result indicated that higher amount of pigment-enriched oil can be produced from dZA1 than parental strain in the optimized process.

Chlamydomonas reinhardtii CC-4349 cw15 mt⁻ (provided by Chlamydomonas Resource Center, University of Minnesota, USA), zep line made by Baek et al. [24] and double knockout mutants were grown in tris acetate phosphate (TAP) medium. Cells were cultured on a shaker (120 rpm) with light intensity of 60 μmol photons m^–2 s^–1 at 25 °C. To produce pigment-enriched edible oil, cells grown in normal TAP medium were cultured in nitrogen-depleted medium to induce lipid production. To simplify the culture process, the culture medium was adjusted to 1/4NP2A medium. The 1/4NP2A contained 1.875 mM NH₄Cl, 0.155 mM K₂HPO₄, 0.103 mM KH₂PO₄, and 2 mL L^–1 glacial acetic acid; the other components were the same as that of the TAP medium.

The C. reinhardtii zep mutant was gene edited as described in a previous study with slight modifications [24]. ADP-glucose pyrophosphorylase (AGP) gene (Cre03.g188250.t1.2) was selected to editing target to develop a double knockout mutant producing macular pigment enriched microalgal oil. The sgRNA target sequence of AGP gene was 5ʹ-TAGCATGGCCCTGAAGATGCGGG-3ʹ. Ribonucleoprotein (RNP) complexes were generated by premixing 100 μg of Cas9 protein purchased (ToolGen, Seoul, South Korea) and 70 μg of in vitro transcribed sgRNA (GeneArt™ Precision gRNA Synthesis Kit, Invitrogen, CA, USA). Then, RNP complex was incubated 5 × 10⁵ cells for 5 min on ice and transferred into a 0.4-cm electroporation cuvette. Cells and the RNP complex mixture were pulsed using a Gene Pulser Xcell Electroporation System (Bio-Rad, CA, USA) with a set pulse at 600 V (50 µF, and 200 Ω resistance). After electroporation, the cells were plated on TAP agar plates.

To confirm the AGP knockout mutant, liquid-cultured cells were visualized by iodine staining. Candidate mutants from the agar plates were transferred to 96-well culture plates. Transferred cells were cultured in nitrogen-deprived TAP medium for 3 days and were then stained with 1 μg of iodine to visualize starch accumulation (dark green or dark grey). The mutants without color change (green or yellowish green) were selected as AGP knock-out mutants, and the genomic DNA was isolated and subjected to PCR with specific primers which are adjacent to sgRNA target sites in AGP (F: 5ʹ-TGGGCACGACTTGCATTGTGT-3ʹ, R: 5ʹ-AATGGGCCAGCGCGAGCATA-3ʹ). The PCR products were analyzed by gel electrophoresis and by Sanger’s sequencing method (Macrogen, Seoul, South Korea). The AGP gene sequence of the mutants was aligned with the innate gene sequence to confirm mutation.

The content of microalgal starch was assayed using a assay kit including amyloglucosidase/α-amylase method (Megazyme, Wicklow, Ireland). The dried biomass was fragmented with a pestle and resuspended in 80% ethanol. Thermostable α-amylase (3000 U mL⁻¹) in sodium acetate buffer (100 mM sodium acetate buffer, pH 5.0) was added to each sample. The samples were incubated in a boiling water bath for 12 min, followed by heating at 50 °C. Amyloglucosidase (3300 U mL⁻¹) in sodium acetate buffer was then added to each sample before heating. The samples were then kept at 50 °C for 30 min and then centrifuged at 2000×g for 10 min. Glucose in the supernatant was assessed using the glucose oxidase reaction based on the colorimetric method (510 nm). Total starch content was calculated according to the following formula provided by the manufacturer:\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${\text{Starch}}, \% = \Delta {\text{A}} \times \frac{{\text{F}}}{{\text{W}}} \times {\text{FV}} \times 0.9$$\end{document}Starch,%=ΔA×FW×FV×0.9ΔA: absorbance; F: glucose amount (100 µg)/absorbance of standard glucose; FV: final sample volume (mL); and W: sample weight (mg).

To confirm the presence of lipid bodies in cells, neutral lipids were stained using BODIPY™ 505/515 (Thermo Fisher Scientific, MA, USA). Cells were mixed with 0.04 µg/mL of BODIPY using a vortex mixer, and the mixture was kept at room temperature for 5 min in the dark. BODIPY-stained lipid droplets were examined by fluorescence microscopy (Eclipse Ni, Nikon, Tokyo, Japan). The fluorescence detection wavelengths is 535 ± 23 nm for the FITC filter for BODIPY and 630 ± 30 nm with the Texas RED filter for chloroplast autofluorescence.

Total lipid contents were measured according to the method described by Bligh and Dyer [43]. The cells were collected after lipid induction and frozen. The cell pellets were resuspended in 4 mL of deionised water, and then 10 mL of methanol and 5 mL of chloroform were added sequentially in a 4:10:5 ratio. After overnight incubation in the dark at room temperature, 5 mL of deionised water and 5 mL of chloroform were added, resulting in a 9:10:10 ratio of water: methanol: chloroform. The extracts were centrifuged for 10 min at 2000×g to separate the two layers. The chloroform layer, which contained lipids, was transferred to a pre-weighed glass bottle and evaporated on a hot plate. The difference in the weight of the bottle before and after evaporation represented the lipid content.

To measure neutral lipids (TAG), the algal cells were stained with the Nile red method described by Chang et al. [44] with some modifications. The fluorescence of Nile red was measured using a Varioskan Flash microplate reader (Thermo Fisher Scientific, USA). The fluorescence intensity was normalized to that of the cells (100 × 10⁴ cells) in the sample. Excitation and emission wavelengths of 530 nm and 595 nm, respectively, were selected. Data are expressed as the mean and standard deviation of three replicates.

Pigments in cells were extracted with 90% (v/v) acetone. The extracts were analyzed using a Shimadzu Prominence HPLC model LC-20AD (Shimadzu, Kyoto, Japan) equipped with a Spherisorb 5.0 μm ODS1 4.6 × 250 mm cartridge column (Waters, Milford, USA). The pigment was detected by a photodiode array detector at 445 nm and 670 nm absorbance, as described previously [45].

Fatty acids were quantified according to the method described by Garcés and Mancha [46]. Lyophilized cells were treated with 2 mL of methylation mixture (MeOH: benzene: DMP (2,2-dimethoxy-propane): H₂SO₄, 39: 20: 5: 2) in a 4-mL vial with a Teflon cap. Then, 1 mL of heptane was added. The mixture was incubated at 95 °C for 2 h for the extraction. After cooling at room temperature, the separated organic phase (upper phase) was collected. Pentadecanoic acid (Sigma-Aldrich, MO, USA) was used as the internal standard. Fatty acid methyl esters (FAMEs) were analyzed using a gas chromatograph (7890A GC, Agilent technology, CA, USA) equipped with a flame ionisation detector and a DB-23 column (60 m × 0.25 mm × 0.25 µm); (Agilent Technology, CA, USA). Helium was used as carrier gas. For FAME analysis, the injection volume was 1 μL (split ratio, 1:10). The injector temperature was 250 °C, and the detector temperature was 280 °C. Initially, the oven temperature was held at 50 °C for 1 min, gradually increased to 250 °C and then held at 250 °C for 5 min. This study was supported by the National Instrumentation Center for Environmental Management (NICEM).

To extract the edible oil at low temperatures over a short time period, ultrasound-assisted extraction (UAE) was used. A 0.5 g sample of dried biomass and 50 mL of 3:2 (v/v) hexane/isopropanol mixture (HIP) was used to extract pigment-enriched oil, and the mixture was sonicated in an ultrasonic bath at a frequency of 40 kHz and power of 350 W for 150 min at 35.7 °C (Powersonics 505, Hwashin Technology, Seoul, Korea). The sonicated mixture was centrifuged at 3500×g for 10 min at 4 °C to separate the liquid phase. The supernatant was collected and mixed with 0.88% KCl solution of the same volume. To collect the organic phase, the supernatant with the KCl solution was transferred to a separating funnel, and the upper phase was collected. Finally, the organic solution was evaporated by rotary evaporation at 50 °C for 3 min. The final extract was mixed with acetone, filtered using a 0.2-µm syringe filter for HPLC analysis, and was mixed with chloroform for lipid analysis.

All experiments were performed at least in triplicate. Student’s t test was performed to compare the quantitative values of Nile red fluorescence, starch, and lipid contents between the wild type and the mutants. The p values that are p < 0.05 are considered to be statistically significant. Data obtained from pigment contents, total lipid productivity, and pigment productivity were analyzed by two-way ANOVA model as described by Assad [47]. Tukey’s multiple comparison was used to test the significance of differences among means. Data are expressed as mean ± standard deviation (SD).

Additional file 1: Table S1. Biomass, total lipid, pigment production from WT, zep, dZA1, dZA2 from the two-step and one-step process (n = 3). Figure S1. Growth curves of WT, zep, dZA1, dZA2 from A one-step process (n = 6) and B two-step process, TAP-N medium (grey) (n = 3).