Light attenuates lipid accumulation while enhancing cell proliferation and starch synthesis in the glucose-fed oleaginous microalga Chlorella zofingiensis

The objective of this study was to investigate the effect of light on lipid and starch accumulation in the oleaginous green algae Chlorella zofingiensis supplemented with glucose. C. zofingiensis, when fed with 30 g/L glucose, synthesized lipids up to 0.531 g/g dry weight; while in the presence of light, the lipid content dropped down to 0.352 g/g dry weight. Lipid yield on glucose was 0.184 g/g glucose, 14% higher than that cultured with light. The light-mediated lipid reduction was accompanied by the down-regulation of fatty acid biosynthetic genes at the transcriptional level. Furthermore, light promoted cell proliferation, starch accumulation, and the starch yield based on glucose. Taken together, light may attenuate lipid accumulation, possibly through the inhibition of lipid biosynthetic pathway, leading to more carbon flux from glucose to starch. This study reveals the dual effects of light on the sugar-fed C. zofingiensis and provides valuable insights into the possible optimization of algal biomass and lipid production by manipulation of culture conditions.


Results
Growth characteristics of glucose-fed C. zofingiensis with or without light. C. zofingiensis was cultured in the medium containing 30 g/L glucose, illuminated either with or without light. As indicated by Fig. 1, C. zofingiensis reached stationary growth phase after six days of cultivation under both culture conditions. Notably, in the presence of light, C. zofingiensis grew faster with a specific growth rate of 0.502 h −1 , which is slightly higher than that without light (0.486 h −1 ) (Fig. 1A). Accordingly, C. zofingiensis with light gave a higher maximum dry cell weight (Fig. 1A). C. zofingiensis tended to turn yellow-orange-red during cultivation period (Fig. 1B), attributed to the synthesis and accumulation of secondary carotenoids, astaxanthin in particular 19,24 . It is worth noting that, C. zofingiensis also exhibited difference in the color of cultures between both conditions (Fig. 1B). The more intensive color of the light cultures resulted from the biosynthesis of light-induced chlorophylls.
Light attenuates lipid accumulation while stimulating starch biosynthesis. In green microalgae, lipid and starch biosyntheses share the common carbon precursors, though the regulation of carbon partitioning into these two biosynthetic pathways is not well understood 25,26 . Glucose-fed C. zofingiensis cultures maintained a basal lipid level during the first 3 days of cultivation; thereafter, the lipid built up rapidly and reached the maximal content of 53% of dry weight on day 7 ( Fig. 2A). The algal cultures provided with light followed the similar pattern of lipid accumulation, but the lipid content was greatly attenuated as compared to the light-free cultures, e.g., 28% of dry weight on day 7, which is 48% less than the cultures without light. In contrast, there was greater starch accumulation in the cultures with light than in the cultures without light, and the starch content of the former remained higher than the latter during the whole cultivation period (Fig. 2B). Interestingly, starch content started to accumulate ahead of lipid accumulation ( Fig. 2A,B). Yield on glucose can reflect the carbon flux allocation. Notably, opposite trend of flux to lipid and starch was observed in cells with and without light (Fig. 2C,D). As compared to the cultures with light, C. zofingiensis without light had higher lipid yield on glucose (Fig. 2C), though both of them dropped after cells started to divide. On the contrary, starch yield on glucose with light was higher than that without light during whole culture period (Fig. 2D). Different from lipid yield on glucose, starch yield remained relatively stable. Figure 3A shows the time course of algal cell density grown with or without light. Dark-grown C. zofingiensis exhibited almost no change in cell number until day 6; in contrast, in the presence of light the alga started to divide on day 2 and reached up to be 15-fold higher in cell density, indicating that cell proliferation was greatly enhanced by light. Consistently, our cell cycle analysis data by flow cytometry also demonstrated that light promoted the algal cell mitosis (Fig. 3B). G1/G0 phase stands for cell in diploid or stationary form, which represents newborn cell after division. Obviously, light cultures showed a much earlier peak in G1/G0 phase than dark cultures (labeled as M1 in Fig. 3B).

Light-induced lipid reduction is accompanied by accelerated cell proliferation.
There was an increase in the per cell weight observed during the early culture period, followed by a significant drop close to the initial value (Fig. 3C). Overall, C. zofingiensis without light maintained greater (up to 5 times) per cell weight than that with light. Similar to the pattern in per cell weight, the per cell lipid content of C. zofingiensis without light showed a drastic increase and reached the maximal value of 250 pg/cell after 5 days of cultivation, which is 5-fold higher than the light-illuminated cultures (Fig. 3D). In addition, BODIPY 505/515, a fluorescent lipophilic dye for the neutral lipids' staining 27 , was employed to monitor the in vivo dynamic changes of lipids in C. zofingiensis cells (Fig. 3E). The green Scientific RepoRts | 5:14936 | DOi: 10.1038/srep14936 signals represent the staining of neutral lipids, predominantly in the form of TAGs. In accordance with the per cell weight (Fig. 3C) and lipid content (Fig. 3D), dark-grown C. zofingiensis exhibited a drastic increase in both cell size and florescent staining with a peak value obtained on day 4, while the algal culture with light reached the maximum on day 2, followed by a gradual decline in cell size and staining due likely to the accelerated cell division (Fig. 3E).
Light alters the transcriptional expression of fatty acid biosynthetic genes. It is commonly agreed that green algae follow the similar lipid biosynthetic pathway as in higher plants. Among the enzymes involved in lipid biosynthesis, acetyl-CoA carboxylase (ACCase) is a rate-limiting enzyme catalyzing the first committed step for de novo fatty acid synthesis in chloroplast 28 . Chloroplastic ACCase is composed of four subunits and the expression of the genes coding the subunits is autoregulated to each other. Thus, the characterization of one subunit such as biotin carboxylase (BC) can be representative of ACCase. Stearoyl ACP desaturase (SAD) introduces the first double bond to acyl chain and plays an important role in determining the degree of saturation of fatty acids 28 . To investigate the effect of light on fatty acid biosynthesis, the transcript levels of SAD and BC in C. zofingiensis were determined using a real time-PCR approach. In the dark-grown C. zofingiensis cells, an increase in the steady-state mRNA level of both SAD and BC was observed and the mRNA levels reached their maximum on day 4, much higher than the maximum values in light-grown cells on day 3 (Fig. 4A,B). This is well consistent with the data that dark cells showed a sharp increase in per cell lipid content on day 4 (Fig. 3C). The introduction of light to dark culture after 3 days of cultivation exerted a negative effect and attenuated the SAD and BC transcripts dramatically compared to dark-grown cells on day 4. It's worth noting, however, that the expression of both genes increased sharply on day 5 and then decreased. When the light-illuminated cells transferred to dark, both SAD and BC transcriptional levels exhibited significantly higher than those in light-grown cells from day 4 to 6. Overall, C. zofingiensis accumulated more SAD and BC transcripts in dark than under light, which may explain why dark cells accumulated more lipids than light cells.
Culture conditions have little effect on fatty acid profiles. The quality of biodiesel is largely determined by its fatty acid composition 29 . GC-MS was employed to analyze the fatty acid profiles in C. zofingiensis under different culture conditions. The algal cells produced fatty acids mainly in the form of C18:1 (32.2%-35.8%), C18:2 (18.2%-20.1%), and C16:0 (16.1%-18.5%), which together account for more than 66% of total fatty acids, regardless of the culture conditions (Fig. 5). Although the total lipid contents varied greatly, no significant difference was observed in the fatty acid composition under the tested conditions (Fig. 5).

Discussion
C. zofingiensis can grow well photoautotrophically, mixotrophically, and heterotrophically. Under heterotrophical growth conditions, organic carbon sources, glucose in particular, are the sole carbon and energy sources 8,19,24 . Chlorella possesses an inducible hexose/H + active symport system that is responsible for the uptake of glucose from the medium 30,31 . In the presence of glucose, the hexose/H + symport system protein in Chlorella cells can be activated in just a few minutes 32 . Usually, the specific growth rate of glucose-fed microalgae growing with light is approximately the sum of cell growth rates under photoautotrophic and glucose-fed conditions in dark 33 . This might explain our findings that C. zofingiensis grew faster feeding on glucose in the presence of light (Fig. 1A). In green microalgae, lipid and starch are the two dominant energy storage forms and share the common carbon precursors for biosynthesis. It has been reported that photoautotrophic Chlorella is able to accumulate lipid and starch up to 60% and 45% of dry weight, respectively, depending on the algal strains and culture conditions [34][35][36][37][38] . Little attention, however, has been paid to the effect of light on carbon flux to lipid and starch in algal cells feeding on organic carbon sources such as glucose. In the present study, for the first time, we investigated the accumulation of lipid and starch in glucose-fed C. zofingiensis with or without light. Both lipid and starch contents increased, but starch accumulation preceded lipid synthesis ( Fig. 2A,B), which is consistent with the previous studies in photoautotrophically cultured C. zofingiensis 39 and Pseudochlorococcum sp 40 . Compared to the light cultures, the dark cultures tended to accumulate more lipids and less starch (Fig. 2). One possible explanation is that lipids require more reducing power than starch for production, while glucose-fed cultures in dark can generate more reducing power 17 . For example, lipid synthesis for a C18 fatty acid needs 16 NADPH molecules 33 , which is less energetically economical than starch synthesis, as the latter requires 6 NADPH molecules and 9 ATP molecules to form an 18-carbon molecule 40 . On the other hand, light can stimulate algae proliferation (Fig. 3). Cell division is an energy-consuming process and microalgae tend to accumulate sufficient energy before proliferation. The energy storage materials, lipids and starch in particular, tend to accumulate in the algal cells under stress conditions when the cell growth halts. Lenneke 41 discovered that starch and lipid accumulation in Neochloris oleoabundans occurred before mitosis. Stress relief facilitates the degradation of lipids or starch, which can provide energy for boosting the cell growth. In the present study, we noticed a sharp decrease in lipid yield on glucose but not in starch yield on glucose after cell proliferation (Fig. 2C,D), suggesting that glucose-fed C. zofingiensis cells tended to utilize lipids rather than starch to provide energy for cell division.
SAD and BC are two genes encoding key enzymes involved in fatty acid biosynthesis. It has been suggested that the control of these two genes on fatty acid biosynthesis may occur at transcriptional level in C. zofingiensis 25 . Consistent with the attenuated lipid yield on glucose after 3 days (Fig. 2C), a depressed expression level of these two genes was found in the cells provided with light, as compared   to the dark-grown cells on day 4 (Fig. 4A,B). In this context, light might down-regulate the expression of fatty acid biosynthetic genes, leading to the decrease in lipid content to provide energy for cell proliferation, which would direct more carbon flux from glucose to starch biosynthetic pathway resulting in enhanced starch accumulation.
Fatty acid composition determines the quality of biodiesel and is subject to change in different culture conditions. The fatty acids in C. zofingiensis cells consisted predominantly of 16-18 carbons and the maximum unsaturation degree was 3, which are similar to that of plant oils currently used for biodiesel production 42 . It was found that oil from C. zofingiensis in dark was more suitable than that from photoautotrophic cells for biodiesel production, as the former contained high content of oleic acid (C18:1), linoleic acid (C18:2), and palmitic oil (C16:0), which can balance oxidative stability and low-temperature properties and can promote the quality of biodiesel 8 . Notably, our result suggested that light had no significant effect on fatty acid composition in glucose-fed C. zofingiensis.
C. zofingiensis accumulates lipid efficiently in the dark supplemented with glucose. Light has a dual effect on C. zofingiensis: promoting cell proliferation and biomass yield, while on the other hand enhancing starch accumulation at the cost of lipid possibly through the inhibition of lipid biosynthetic pathway. Our results provided insights into utilizing different culture conditions for boosting biomass, lipid content, and lipid yield on glucose. O. Thirty grams of glucose was added to 1 liter of medium. The pH of the medium was adjusted to pH 6.1 prior to autoclaving. Briefly, 10 mL of liquid Kuhl medium was inoculated with cells from slants, and the alga was grown aerobically in flasks at 25 °C for 4 days with orbital shaking at 150 rpm and with continuous illumination at 50 μ mol photon m −2 ·s −1 . The cells were then inoculated at 10% (v/v) into a 250-mL Erlenmeyer flask containing 50 mL of the growth medium. Algal cells in exponential growth phase were used as seed cells for the following batch cultures.

Methods
For dark cultivation, seed cells were inoculated into 100 mL of fresh medium in 500-mL flasks at a starting cell density of 0.5 g/L and were grown in the dark at 25 °C with orbital shaking at 150 rpm. Light cultivation was conducted under continuous illumination at 50 μ mol photon m −2 ·s −1 ; the other parameters were the same as for dark culture.
A total of 48 samples were divided into four groups, which were cultivated in the dark, in light, in dark-to-light and in light-to-dark conditions. Samples in the latter 2 groups were transferred to the light or dark after three days of incubation. Samples from four culture condition groups were collected for testing every day.

Analysis of lipid content and starch content.
Total lipids were extracted from lyophilized cell powder according to Converti 43 with some modifications. A 100-mg mass was ground before 2 extractions with petroleum ether; the supernatants were then merged and evaporated with N 2 . The crude oil was then weighed. Lipid content was expressed as lipid weight per unit biomass.
The starch content was analyzed using the above defatted sediment according to a modified method used by Brányiková 34 . A 30% perchloric acid solution was added to 5 mg of sediment, stirred for 15 min at 25 °C and centrifuged. This procedure was repeated 3 times. The extracts were combined, and the volume was adjusted to 10 mL. Next, 2-mL aliquots of solubilized starch solution were reacted with 5 mL of concentrated sulfuric acid (98% by weight) and 1 mL of phenol (6%, w/v) at room temperature for 10 min. The absorbance was read in a spectrophotometer at 490 nm. Samples were then quantified by comparison to a calibration curve using glucose as the standard. Starch content was expressed as starch weight per unit biomass.
Determination of cell density, biomass, specific growth rate, per cell weight, per cell lipid content, lipid and starch yield on glucose. Cell density were counted using a hemocytometer.
Microalgal cells were centrifuged and filtered through a pre-dried Whatman GF/C filter (Cat No 1822-047) after 2 washes with distilled water. Next, the filter paper was dried at 80 °C in a vacuum oven for 12 h and was subsequently cooled down to room temperature before weighing. The biomass was expressed as cell dry weight. The specific growth rate (μ max ) was calculated according to:  RNA isolation and real-time RT-PCR assay. The expression levels of two genes involved in fatty acid synthesis were determined using real-time PCR according to Liu 46 . Briefly, RNA was isolated from aliquots of approximately 10 8 cells using TRIZOL reagent (Molecular Research Center, Cincinnati, OH, USA) according to the manufacturer's instructions. The concentration of total RNA was determined spectrophotometrically at 260 nm. Total RNA (1 μ g) extracted from different samples was reverse transcribed to cDNA using the SuperScript III First-Strand Synthesis System (Invitrogen, Carlsbad, CA, USA) for reverse transcription PCR (RT-PCR) primed with oligo(dT) according to the manufacturer's instructions. Real-time RT-PCR analysis was performed using 1 μ L of the RT reaction mixture in a total volume of 20 μ L with specific primers and the Platinum SYBR Green qPCR SuperMix-UDG (Invitrogen). PCR amplification was conducted using specific primers targeting BC (forward, 5′ -GTGCGATTGGGTATGTGGGGGTG-3′ and reverse, 5′ -CGACCAGGACCAGGGCGGAAAT-3′ ), SAD (forward, 5′ -TCCAGGAACGTGCCACCAAG-3′ and reverse, 5′ -GCGCCCTGTCTTGCCCTCATG-3′ ), and the internal control actin (ACT) gene(forward, 5′ -TGCCGAGCGTGAAATTGTGAG-3′ and reverse, 5′ -CGTGAATGCCAGCAGCCTCCA-3′ ). PCR was performed in a Bio-Rad iCycler IQ Multi-Color Real-Time PCR Detection System (Bio-Rad, Hercules, CA). The relative levels of the amplified mRNAs were evaluated using the 2 −ΔΔCt method 47 , using the actin gene for normalization.
Cell cycle analysis. Cell cycle was determined by flow cytometry (FCM) with prodium iodide (PI) staining. The method was according to Gerashchenko 48 with modifications. Briefly, 10 6 cells were collected and washed twice with PBS. Then methanol was added in before removing PBS. Cells in methanol were dispersed and stored in 4 °C for analysis. Upon analysis, the samples were washed with PBS and then stained with 50 lg/ml PI in the presence of 25 lg/ml RNase A in 37 °C bath for 30 min. The cell cycle distribution of 10,000 cells was recorded by a flow cytometer (BD FACS Calibur), and result was analyzed with ModFit software.