Effects of different abiotic stresses on carotenoid and fatty acid metabolism in the green microalga Dunaliella salina Y6

Under different abiotic-stress conditions, the unicellular green microalga Dunaliella salina accumulates large amounts of carotenoids which are accompanied by fatty acid biosynthesis. Carotenoids and fatty acids both possess long carbon backbones; however, the relationship between carotenoid and fatty acid metabolism is controversial and remains poorly understood in microalgae. In this study, we investigated the growth curves and the β-carotene, lutein, lipid, and fatty acid contents of D. salina Y6 grown under different abiotic-stress conditions, including high light, nitrogen depletion, and high salinity. Both high-salinity and nitrogen-depleted conditions significantly inhibited cell growth. Nitrogen depletion significantly induced β-carotene accumulation, whereas lutein production was promoted by high light. The accumulation of lipids did not directly positive correlate with β-carotene and lutein accumulation under the three tested abiotic-stress conditions, and levels of only a few fatty acids were increased under specific conditions. Our data indicate that cellular β-carotene accumulation in D. salina Y6 positive correlates with accumulation of specific fatty acids (C16:0, C18:3n3, C14:0, and C15:0) rather than with total fatty acid content under different abiotic stress conditions.


Introduction
Carotenoids are naturally occurring yellow-to orangered-colored compounds with a polyisoprene backbone, mainly comprising β-carotene and lutein, which are found in bacteria, cyanobacteria, microalgae, fungi, and plants (Concepcion et al. 2018;Kanzy et al. 2015;Saini and Keum 2017). Carotenoids are used in the food supplements, cosmetics, and medical industries. In addition to their use as colorants, carotenoids have health functions and are also efficient in preventing cancer, macular degeneration, and cataracts (Mussagy et al. 2019).
Carotenoid synthesis in microalgal cells is accompanied by lipid biosynthesis under HL conditions or a combination of HL and ND (Rabbani et al. 1998;Paliwal et al. 2017). The carotenoid that accumulated in small lipid droplets within the chloroplast, termed as βcarotene plastoglobules, were composed of 40-50% (w/ w) triacylglycerols, 50-60% (w/w) β-carotene, a small amount of proteins and were surrounded by a monolayer of polar lipids (Benamotz et al. 1982;Davidi et al. 2015). The carotenoid was composed of two stereoisomers, all-trans β-carotene and 9-cis β-carotene, in approximately equal amounts (Davidi and Pick 2017). The two stereoisomers differ in their spectral properties and in their physical properties: 9-cis β-carotene was more lipid-soluble than all-trans β-carotene and probably acts as a solvent for the all-trans isomer in β-carotene plastoglobules, since the latter tends to crystalize at high concentrations (Davidi and Pick 2017). In addition, some abiotic stress was reported to activate the expression of some key genes of carotenoid biosynthesis, such as phytoene desaturase, lycopene β-cyclase, 4-hydroxy-3methylbut-2-enyl diphosphate reductase, and phytoene synthase Ramos et al. 2008;Ramos et al. 2009). .As an alternative to activation of genes of the carotenoid biosynthetic pathway, Rabbani et al. (1998) suggested that synthesized lipid globules can sequester carotenoids and is the driving force for βcarotene overproduction. Nevertheless, carotenoids were not produced when D. salina was cultured in the presence of fatty acid synthesis inhibitors (such as sethoxydim and cerulenin), namely, the metabolisms of carotenoids and fatty acids were inseparable and that carotenoid synthesis was positively correlated with lipid accumulation (Rabbani et al. 1998). However, the amount of carotenoids produced in microalgal D. salina (CCAP 19/18) cells is positively correlated with the synthesis of fatty acids oleic acid (C18:1) and palmitic acid (C16:0), and has little association with lipids under HL and ND conditions (Lamers et al. 2010;Mendoza et al. 1999). Thus, the relationship between carotenoid and fatty acid metabolism is controversial and remains to be elucidated in microalgae.
In this paper, we investigated the growth curves and the β-carotene, lutein, lipid, and fatty acid contents of the green microalga D. salina Y6 under HL, ND, and HS conditions in order to investigate the relationship between carotenoid production and fatty acid metabolism in this species.
For high-light (HL) treatment, the light intensity was 200 μmol m −2 s −1 . For nitrogen-depleted (ND) treatment, the culture medium described above without KNO 3 and NH 4 NO 3 was used. For high-salt (HS) treatment, the microalgal cells were cultured in DsMG medium with a high salt content (NaCl, 175.5 g/L). The remaining culture conditions for HL, ND, and HS treatments were the same as described above. The cell densities at day 0, 2, 4, 6, 8, 10, 12, 14, 16, 18, and 20 were measured, and microalgal cell samples were taken on day 12 (exponential phase) and day 20 (stationary phase) for analysis of cellular dry weight, lipid content, fatty acid composition and contents, and β-carotene and lutein contents. The harvested microalgal cells were stored at −80°C prior to use. Three biological replicates were included for each sample.

Growth measurement
To monitor cellular growth, cell numbers were counted by using a hemocytometer (Improved Neubauer, USA). Cell dry weight was measured according to the method described by Lee et al. (2013) and Wu et al. (Wu et al. 2020;Wu et al. 2019) with the following adjustments.5 -10 mL of microalgae suspension was filtered through a preheated (105°C, 24 h), pre-weighed glass microfiber filter (Whatman GF/C, 47 mm, UK). The filters were washed twice each with 20 mL of 0.5 M ammonium bicarbonate. The filters were weighed after drying at 105°C for 24 h to reach a constant weight. Cell dry weight (DW, g/L) was calculated using Eq. (1): where "w a " and "w b " were the weight of the filters at the end and start of cultivation, respectively, and "v" was the volume of the microalgae suspension filtered.

Fatty acid composition and contents
For fatty acid quantification, ca. 5 mg of lyophilized cell pellets were transferred to a screw-capped glass centrifuge tube, to which 50 μL of the C19:0 internal standard (5 mg of methyl decanoate added to 10 mL of dichloromethane) and 1 mL of 2 M NaOH-CH 3 OH solutions were added. After vigorously shaking at 300 rpm for 1 h, the mixtures were heated at 75°C in a water bath for 15 min. After cooling down to room temperature, 1 mL of 4 M HCl-CH 3 OH solution and 0.5 mL 38% (v/v) HCl were added followed by heating at 75°C for 15 min. After cooling down to room temperature once more, 1 mL of hexane was added to the extracts, followed by a brief centrifugation. Five-hundred microliters of the hexane top layer (passed through a 0.22-μm pinhole filter) from the tube was transferred into a GC vial for fatty acid methyl esters (FAMEs) analysis, which was performed by using an Agilent 7890A gas chromatograph coupled to a 5975C mass spectrometer (GC-MS). The temperature program was as follows: initial temperature of the oven was 70°C, which was held for 2 min, then increased at 25°C/min to 195°C and held for 5 min, then increased at 3°C/min to 250°C. Helium was used as the carrier gas with a flow rate of 1 mL/min. The temperature of both the injector and the detector was set at 250°C. The injection volume was 1 μL, and the split ratio was 10:1. The spectrometer was set to scan in the range of m/z = 50-500 at 70 eV with electron impact (EI) mode of ionization. The FAMEs were quantified by comparing the detected peaks of the total ion count to the total ion count of the internal standard. Duplicate samples were analyzed for each time point of each experiment (Wu et al. 2019).

β-carotene and lutein quantification
Carotenoid extraction methods were carried out as described by Jiang et al. (2004) with the following adjustments. Twelve microliters of microalgal solution was transferred into a 15-mL centrifuge tube and centrifuged (1200×g, 4°C) for 5 min. The supernatant was removed and the precipitate was washed twice with distilled water, after which 5 mL of acetone was added. The samples were sonicated for 5 min, and then centrifuged (3200×g, 4°C) for 3 min until the microalgal pellet turned white. The extract was transferred into another centrifuge tube to which 0.5 mL of 60% (v/v) KOH was added, and after shaking for a few minutes, allowed to stand until the layers separated. The supernatant extract was filtered through a 0.22-μm pinhole filter and finally stored at −20°C until further use. According to related previous studies (Aman et al. 2005;Li et al. 2006), a high-performance liquid chromatography (HPLC) (LC-20A, Shimadzu Corporation) method was used to analyze the β-carotene and lutein (Darko et al. 2000) contents. Acetonitrile:deionized water (95:5, v:v) was used as the first mobile phase to separate lutein, then acetonitrile:methanol:dichloromethane (80:5:23, v:v:v) was used as the second mobile phase quickly to separate the β-carotene. For HPLC, the detection wavelength was 450 nm, the column temperature was 28°C, the injection amount was 50 μL, and the flow rate was 0.8 mL·min −1 . The β-carotene standard was configured as a series of standard solutions of 0.3, 0.5, 1.0, 2.0, and 4.0 mg/L, whereas the lutein standard was designed as standard solutions of 0.5, 1.0, 2.5, 4.0, and 6.0 mg/L. Finally, the contents of β-carotene and lutein were determined according to the correlation equation between the carotenoid peak area and the standard curves.

Statistical analysis
Experiments were carried out with biological replicates included from three separate cultures. Samples were collected from three microalgal replicates, and data were analyzed for the standard errors. All data were depicted as mean ± standard deviations (mean ± SD) and statistically analyzed by Student's t test to investigate differences compared to the control group. A p value less than 0.01 (p < 0.01) was considered highly significantly different, p < 0.05 represents statistically different, while p > 0.05 means not significant.

Growth curves
Interestingly, the cell density of D. salina Y6 cultures was significantly higher, with a density of 13.6 × 10 5 cells/mL observed for HL cultures by day 20, which was 10.4% higher than that under control conditions (p < 0.01) (Fig. 1). Surprisingly, DW and cellular content of HL-treated cells were lower than those of the control (p < 0.05) (Fig. 2). These results indicate that HL promotes D. salina Y6 culture growth but inhibits the accumulation of cellular biomass. Xu et al. (2018) found that D. salina strains, DF15 and UTEX 2538 still maintained high photosynthetic efficiency under high light conditions. Therefore, based on the results of this study, we speculated that this D. salina Y6 also had a relatively strong adaptability to high light, which improved photosynthetic efficiency, thereby speeding up cell division and energy storage substances (such as starch and lipid) accumulation were reduced. These phenomena deserve later scholars' in-depth study on the adaptability mechanism of this algae species to high light in the future.
Under ND, the cell density of D. salina Y6 cultures was significantly lower than that under control conditions, with a maximal cell density of 7.8 × 10 5 cells/mL observed for ND cultures by day 20, which was only 63.4% of that of the control (p < 0.01). Nevertheless, the DW of ND-treated cells reached 652 mg/L, which was 29.1% higher than that of the control (p < 0.01) by day 20. The cellular content of ND-treated cells reached 12.5 mg/10 6 cells, approximately onefold higher than that of the control by day 20 (p < 0.01) (Fig. 2), suggesting that microalgal culture growth was inhibited by ND whereas biomass accumulation per cell was promoted.
Similarly, D. salina Y6 culture growth was significantly inhibited under HS compared to the control, with HS resulting in a cell density of 10.0 × 10 5 cells/mL, which was only 81.7% of that of the control (p < 0.05) (Fig. 1).
Nevertheless, single-cell biomass (DW and cellular content) was increased under HS conditions by day 12 (p < 0.05) and 20 (p < 0.01) (Fig. 2). These results indicate that D. salina Y6 culture growth is inhibited to a certain extent under HS conditions, and that HS can promote the accumulation of single-cell biomass.
In summary, these results demonstrate that HL promotes D. salina Y6 culture growth, whereas both ND and HS severely inhibit culture growth. Nevertheless, the highest biomass accumulation (DW and cellular content per 10 6 cells) was observed under ND and HS conditions, whereas HL resulted in the lowest single-cell biomass. The cells' division of D. salina Y6 was not increased under the conditions of ND and HS stress, which caused the cells to accumulate a large amount of energy storage substances. This phenomenon also occurred in other strains/microalgae (Abomohra et al. 2020;Wu et al. 2019). This algae may had a strong adaptability under HL conditions, which was conducive to promoting cell division and reducing the accumulation of single-cell biomass.

Fatty acid composition and accumulation
Microalgae generally accumulate large amounts of lipids under abiotic-stress conditions (Paliwal et al. 2017). The lipid content of D. salina Y6 cells was about 108 μg/mg (10.8% DW) under suitable conditions that was similar to the 12.8% DW reported in a previous research (Lamers et al. 2012). However, we found that the lipid content of D. salina Y6 cells under HL treatment was only slightly increased by 6% compared to the control by day 20 of culture ( Fig. 3). Lipid content decreased significantly under ND treatment during exponential phase (p < 0.01), but no significant change in final lipid content was observed compared to the control by day 20 (Fig. 3) (p > 0.05). Moreover, under HS stress, the lipid content of microalgal cells by day 20 was 97 μg/mg, which was significantly lower than that of the control (108 μg/mg) (Fig. 3) (p < 0.05), indicating that HS stress does not promote lipid accumulation in D. salina Y6. In general, lipid accumulation increased significantly in microalgae under HL, ND, and HS conditions (Abomohra et al. 2019;Abomohra et al. 2020;Monte et al. 2020;Paliwal et al. 2017). Nevertheless, the final lipid contents under HL, ND, and HS conditions were not considerably greater than that of the control group, potentially indicating that the stress treatments did not promote lipid production in D. salina Y6. This strain of D. salina may had special enzymes to regulate the metabolic pathway of carbon flow under different stresses, and it is worth our in-depth study in the future.

β-carotene and lutein accumulation
Carotenoid biosynthesis is induced in D. salina as a protective mechanism against photoinhibition, since the accumulation of β-carotene in lipid globules absorbs excess light energy and protects the photosynthetic apparatus, especially during UV-A-mediated photoinhibition (Wu et al. 2018). Moreover, photosynthetically produced oxygen radicals trigger considerable βcarotene accumulation in D. salina (Lin et al. 2017).
Here, we found that the β-carotene content of D. salina Y6 increased under HL and ND conditions, resulting in contents of 1162.9 μg/mL and 1332.4 μg/mL by day 20, which were 31.5% and 50.6% higher than those of the control, respectively (Fig. 4) (p < 0.01). These results suggest that ND treatment more efficiently induces βcarotene synthesis in D. salina Y6. Conversely, the βcarotene content under HS conditions was lower than that of the control by day 12 and no significant difference was observed by day 20 (Fig. 4) (p > 0.05), indicating that HS does not promote β-carotene synthesis in D. salina Y6.
Lutein is an antioxidant that protects microalgal cells from the toxicity of reactive oxygen species produced under extreme environmental conditions (Chokshi et al. 2017). We observed an increase in lutein production under each of the three tested abiotic-stress conditions, especially under HL where the lutein content was significantly higher than that under the other stress conditions. For example, the lutein content under HL was 854.6 μg/ mL (1.2 fold higher than that of the control) and 1133.2 μg/mL (95.9% higher) by day 12 and 20, respectively (Fig. 5) (p < 0.01). These results indicate that HL promotes the synthesis of lutein, which is largely consistent with the findings of Fu et al. (2013). Under ND conditions, the lutein content was 594.5 μg/mL (50.6% higher than that of the control) and 744.6 μg/mL (28.7% higher), whereas under HS conditions, the lutein content was 441.8 μg/mL (11.9% higher) and 778 μg/mL (34.5%  higher) by day 12 and 20, respectively (Fig. 5). These results show that ND and HS conditions have similar effects on promoting lutein accumulation in D. salina Y6. In general, HL promoted lutein accumulation more than ND and HS.
Although D. salina Y6 cells were analyzed under the same stress conditions, the contents of β-carotene and lutein varied greatly. For example, in this study, the abiotic-stress conditions that promoted β-carotene accumulation were ND > HL > HS, whereas the conditions that promoted lutein accumulation were HL > ND ≈ HS. It is well known that different metabolic pathways are responsible for the biosynthesis of β-carotene and lutein (Lamers et al. 2010). We propose that different abioticstress conditions affect the gene expression levels and activities of enzymes functioning in these metabolic pathways. It is of interest to investigate these regulatory mechanisms in future studies.

Carotenoid and fatty acid metabolism
Dunaliella salina accumulates lipid droplets, including larger triacylglycerols, and these lipid globules are frequently synthesized under carotenogenic conditions (Bonnefond et al. 2017). A combination of β-carotene and fatty acids in the same lipid droplet has been reported for D. salina under HL-and ND-stress conditions (Pick et al. 2019). However, here, we did not found a significant positive correlation between the synthesis of lipids and βcarotene. Under HL, it was reported that the accumulation of β-carotene is correlated only with the contents of specific fatty acids (C18:1 and C16:0) and the proportions of unsaturated fatty acids (Lamers et al. 2010). It has also been proposed that oleic acid (C18:1) is the main component of the β-carotene-containing lipid droplets in D. salina (Lamers et al. 2012); however, the results of this study did not reveal a significant change in the C18:1 content of D. salina Y6 under HL stress, but C18:3n3 was increased Fig. 6 Schematic diagram of carotenoid and fatty acid metabolism in D. salina Y6. The dotted and solid lines represent indirect and direct chemical reactions, respectively; red triangles, green squares, and black ellipses represent related compounds are increased under the high light (HL), high salt (HS), and nitrogen-depleted (ND) conditions, respectively; G3-P, glyceraldehyde-3-phosphate; PEP, phosphoenolpyruvate; MVA Pathway, mevalonic acid, produced in the mevalonate pathway; MEP pathway, methylerythritol 4-phosphate pathway; IPP, isopentenyl pyrophosphate; GGPP, geranylgeranyl pyrophosphate compared to the control. He et al. (2017) found that the microalgal cell content of C18:3n3 decreased in the presence of photosynthetic inhibitors, whereas the main fatty acid C18:3n3 in D. salina was closely related to chloroplast-like synthesis. It was also reported that the microalgal cell content of C18:3n3 increased under HL, indicating that these fatty acids may play a role in resisting photooxidative stress (Bredda et al. 2019). We speculated that the accumulation of C18:3n3 in D. salina Y6 is positively correlated with carotenoid synthesis under HL. In addition, it was found that the same specific fatty acid contents were higher under stress conditions than in the control, such as C16:0 and C18:3n3 that accumulated under HL, C14:0 and C15:0 that accumulated under ND, and C14:0 and C16:0 that accumulated under HS (Fig. 6). These results suggest that the synthesis of these fatty acids is positively correlated with carotenoid accumulation under different abiotic-stress conditions in D. salina Y6. Continuing investigation of these specific fatty acids is worthwhile, for instance using comparative metabolomics and transcriptomics analysis, to reveal further details of carotenoid and specific fatty acid metabolism under different abiotic stress conditions.

Conclusions
In this study, we found that different abiotic stresses result in different metabolite profiles in the D. salina Y6 strain. Specifically, HL promotes cell biomass accumulation, whereas HS and ND inhibit cell growth. Furthermore, the production of β-carotene is induced by ND > HL > HS, whereas lutein production is induced by HL > ND ≈ HS, indicating different sensitivity of these two pigments to different stresses. Under abiotic-stress conditions, the accumulation of lipids is not directly related to β-carotene and lutein contents. Still, these specific fatty acids (C16:0, C18:3n3, C14:0, and C15:0) are positively correlated with carotenoid accumulation under different abiotic-stress conditions in D. salina Y6. Further investigation to establish the relationship between carotenoid and fatty acid metabolism using comparative metabolomics and transcriptomics analysis will be carried out in future.