ABSTRACT
To improve the economic viability of microalgal biodiesel, it will be essential to optimize the productivity of fuel molecules such as triacylglyceride (TAG) within the microalgal cell. To understand some of the triggers required for the metabolic switch to TAG production, we studied the effect of the carbon supply (acetate or CO2) in Chlamydomonas reinhardtii (wild type and the starchless sta6 mutant) grown under low N availability. As expected, initial rates of TAG production were much higher when acetate was present than under strictly photosynthetic conditions, particularly for the sta6 mutant, which cannot allocate resources to starch. However, in both strains, TAG production plateaued after a few days in mixotrophic cultures, whereas under autotrophic conditions, TAG levels continued to rise. Moreover, the reduced growth of the sta6 mutant meant that the greatest productivity (measured as mg TAG liter−1 day−1) was found in the wild type growing autotrophically. Wild-type cells responded to low N by autophagy, as shown by degradation of polar (membrane) lipids and loss of photosynthetic pigments, and this was less in cells supplied with acetate. In contrast, little or no autophagy was observed in sta6 mutant cells, regardless of the carbon supply. Instead, very high levels of free fatty acids were observed in the sta6 mutant, suggesting considerable alteration in metabolism. These measurements show the importance of carbon supply and strain selection for lipid productivity. Our findings will be of use for industrial cultivation, where it will be preferable to use fast-growing wild-type strains supplied with gaseous CO2 under autotrophic conditions rather than require an exogenous supply of organic carbon.
INTRODUCTION
Microalgae have the potential to produce significant amounts of neutral lipids, such as triacylglyceride (TAG), which can be used as a source of biodiesel (1). In species such as Chlorella vulgaris (2), Nannochloropsis oculata (3), and Chlamydomonas reinhardtii (4–7), TAGs can accumulate to levels up to 40 to 50% of cell dry weight. To initiate the desired TAG production, a nutrient limitation regime is usually required, such as removal of nitrogen (N) from the growth medium (8). Nutrient limitation also results in a substantial increase in the amount of starch (6, 7, 9), but in mutant strains of C. reinhardtii that are unable to synthesize starch, excess photosynthate cannot be allocated to this storage molecule, so levels of TAG per cell much higher than those for the wild type (WT) have been reported (10–12). For example, there is a 30-fold increase in the number of TAG lipid droplets in the C. reinhardtii sta6 mutant, which is deficient in ADP-glucose pyrophosphorylase (13). In all cases, however, such TAG production comes at a cost, as the growth rate of the cultures can also be substantially decreased by nutrient limitation (6, 10–12). This situation can be avoided to some extent by reducing, rather than removing, the N from the growth medium, as optimal TAG production in C. vulgaris was seen when cells were grown in low-N-containing medium rather than in the complete absence of N (2). Nevertheless, in order to improve the economic cost of algal biofuels, it is essential to understand how changes in the availability of nutrients to the algal cells can influence the triggers that underpin TAG metabolism and production, so as to optimize productivity.
Studies of lipid production in microalgae have shown that, as well as changes in N availability, addition of organic carbon to the medium can result in the cells producing larger amounts of lipids and rates of growth than N limitation alone (14, 15). For example, by supplementing medium with 1% acetate, C. vulgaris can increase its total lipid content from 33 to 36% of its dry weight when grown under N limitation (16). These and other observations have led to the presumption that TAG biosynthesis is responsive to the C/N ratio that cells experience, such that there is a switch from growth when the ratio is normal to TAG production when it is unbalanced by reducing the N or by increasing the C, or both.
Under N limitation, it has been postulated that autophagy takes place, such that internal (probably thylakoid) membranes are broken down and used to provide the precursors for TAG biosynthesis. This conclusion is supported by studies that show that cells grown in N-depleted medium had decreased concentrations of polar lipids and chlorophyll (6, 7, 9, 12, 17). In fact, the majority of studies on TAG accumulation in C. reinhardtii have included fixed carbon in the form of acetate in the medium (7, 12, 13, 18), under which conditions photosynthetic activity is reduced (12). The utilization of acetate for cellular metabolism is likely to be via the glyoxylate cycle and gluconeogenesis, but for lipid synthesis, it may well be as a result of direct incorporation of acetyl coenzyme A (acetyl-CoA) into fatty acids, as has been shown for higher-plant chloroplasts (19–21). The supply of acetate to C. reinhardtii could therefore provide a preferential source of acetyl-CoA for de novo fatty acyl synthesis over that from pyruvate derived from photosynthesis.
To enable economic production of algal biofuels, it would be preferable to use gaseous CO2 in autotrophic growth environments rather than to supply exogenous organic carbon (22), which in any case can exacerbate problems with bacterial contamination (23). In the study described in this paper, we sought to compare directly the effect of an organic or inorganic carbon supply (acetate or CO2) on TAG production and autophagy in C. reinhardtii wild-type and sta6 mutant cells grown under low N availability. Specifically, we hypothesized that if acetate were being directly used as a precursor for de novo fatty acid synthesis, the initial rate of TAG production would increase and there would be less cellular autophagy, especially in sta6 mutant cells, where TAG production is greater than that in wild-type cells. At the same time, we aimed to address the issue of whether it is the C/N ratio, rather than absolute C and N concentrations, that is sensed by the cells in determining whether TAG production is initiated. The potential productivity of these strains in industrial settings is also addressed by considering TAG accumulation beyond the short-term (2 days) initial increase in TAG production investigated in other studies (8, 16).
MATERIALS AND METHODS
Culturing of algae.Stocks of Chlamydomonas reinhardtii wild-type strain 12 (WT-12) nit− (derived from strain CC124 137c mt− nit1 nit2) and the C. reinhardtii sta6 mutant (also known as BAFJ5 [cw15 arg7-7 nit1 nit2 sta6-1::ARG7]) were obtained from Saul Purton (University College London) (24, 25). The parental strain of the cell wall-deficient sta6 mutant is thought to be strain 330 (mt+ arg7-7 cw15 nit1 nit2) (7, 12, 24). We chose to use WT-12 in our studies, since it is the parent to the cw15 strain and has a complete cell wall, which would be much more likely to be suitable for large-scale cultivation.
Colony growth on agar plates.Five-microliter spots of C. reinhardtii WT-12 or the sta6 mutant (normalized to cell number; Z2 particle count analyzer; Beckman Coulter Ltd., United Kingdom) stock cultures were added to either Sueoka's high-salt medium (HSM) or Tris-acetate-phosphate (TAP) solid 2% agar plates. Plates were incubated at 25°C with either continuous or diurnal (12 h of light/12 h of dark) light (120 μmol m−2 s−1) for 7 days.
Transfer of stock cultures to normal-nitrogen- or low-nitrogen-containing medium.Cells were initially grown in either Sueoka's HSM containing 9.34 mM NH4Cl (with a normal level of nitrogen, referred to as 1N HSM) (26) for autotrophic growth or TAP containing 7.01 mM NH4Cl (1N TAP) for mixotrophic growth (27). Cultures were placed in a closed incubator (Infors HT Multitron incubator; Basel, Switzerland) and shaken at 120 rpm with continuous light (120 μmol m−2 s−1) at 25°C. Flasks containing HSM were aerated with 5% CO2 via a sterile filter. At an optical density at 600 nm (OD600; Thermo Spectronic UV1; Thermo Scientific, Hemel Hempstead, United Kingdom) of approximately 0.8, the cultures were centrifuged at 25°C and 600 × g for 10 min (Avanti J-26 XP; Beckman Coulter). Pellets were washed with sterile water to remove excess medium salts, recentrifuged as described above, and then resuspended in the same volume of 4 different media: HSM containing 9.34 mM NH4Cl (1N HSM), HSM containing 0.934 mM NH4Cl (with 1/10 of the normal level of nitrogen, referred to as 0.1N HSM), TAP containing 7.01 mM NH4Cl (1N TAP), or TAP containing 0.701 mM NH4Cl (0.1N TAP). Cultures were grown as described above for 10 days from transfer (day 0), with individual flasks being taken for analysis every other day. For each flask, OD600, OD750, and dry cell mass were determined as described by Stephenson et al. (2), with the rest of the sample being analyzed for the cellular constituents as described below.
Extraction and analysis of total cellular lipids.All chemicals were purchased from Sigma-Aldrich, unless stated otherwise. Lipid extraction was based on a modified Bligh and Dyer method (28), as outlined in the work of Horst et al. (29). To extract the total lipids, cells from 5 ml of culture were pelleted (by centrifugation at 600 × g and 25°C) for 15 min. The supernatant was removed, 10 ml of chloroform-methanol (2:1, vol/vol) was added to the tube, and the tube was vortexed (15 s) and sonicated for 30 min. Prior to extraction, samples were spiked with 1 mg ml−1 C15:0 fatty acid. Five milliliters of chloroform-methanol (2:1, vol/vol) and 5 ml of deionized water were then added to separate the two phases. The lower chloroform phase was dried in a solvent evaporator (45°C; GeneVac EZ-2; SP Scientific, Ipswich, United Kingdom) and resuspended in 200 μl n-heptane (27). To test whether lipase activity was influencing the lipid profiles, especially the free fatty acids (FFAs), a lipase inhibitor (100 μl isopropanol) was added to the pelleted samples of WT-12 and sta6 mutant cultures (day 8 of 0.1N treatment in TAP) during storage at −80°C (30).
TAGs, polar lipids, and FFAs in the crude total lipid extract were analyzed by gas chromatography (GC)-flame ionization detection (FID) with a Varion Select Biodiesel for glycerides GC metal column (10 m by 0.32 mm; film thickness, 0.1 μm; part number CP9076; Agilent Technology), as described by Horst et al. (29). Free fatty acids were identified by coretention of fatty acid standards and quantified using standard curves derived from C17:0 heptadecanoic acid (0.075 to 1.2 mg ml−1). Polar lipids were identified by coretention of monogalactosyl diglyceride, digalactosyl diglyceride, dioleyl phosphatidyl, phosphatidic acid, phosphatidic glycerol, and sulfoquinovosyl diglyceride (Lipid Products, South Nutfield, United Kingdom) and quantified using monogalactosyl diglyceride (0.25 to 1.0 mg ml−1), as this is reported to be the dominant polar lipid in thylakoid membranes (31). TAGs were quantified using glyceryl tripalmitate (see Fig. S1 in the supplemental material).
Preparation, identification, and quantification of FAMEs.The total lipid fractions of the samples were converted to fatty acid methyl esters (FAMEs; acid catalyzed) as described by Stephenson et al. (2). The FAMEs were separated and identified using gas chromatography (Trace GC Ultra; Thermo Scientific) with a Zebron ZB-Wax capillary GC column (30 m by 0.25 mm; film thickness, 0.25 μm; Phenomenex, United Kingdom). The injection volume was 1 μl with a 35:1 split ratio, the injector temperature was 230°C, and helium was used as the carrier gas at a constant flow rate of 1.2 ml min−1. The following gradient was used: initial oven temperature, 60°C for 2 min and then an increase to 150°C at 15°C min−1 and to 230°C at 3.4°C min−1. The detector temperature was 250°C. FAMEs were identified by coelution with a FAME standard mix (grain fatty acid methyl ester mix; catalog no. 47801; Sigma-Aldrich) and were quantified using standard curves derived from C17:0 methyl esters.
Analysis of free fatty acids.Fifteen microliters of the total lipid extract was spotted onto a glass-backed silica gel 60 plate (20 by 20 cm; Merck). The free fatty acids, polar lipids, and TAGs were separated using a 100-ml running solvent of hexane-diethyl ether-acetic acid (70:30:2, vol/vol/vol) in a thin-layer chromatography (TLC) developing tank. Once run, the plate was dried under a stream of N2 gas before being sprayed with 0.2% (wt/vol) 8-anilino-1-naphthalenesulfonic acid in methanol. Free fatty acids were visualized under UV light (366 nm), scraped off the plate, and converted to FAMEs as described above.
FAMEs, and underivatized free fatty acids in the crude total lipid extract, were further analyzed and identified by gas chromatography-mass spectrometry (GC-MS; Trace GC Ultra-DSQII EI MS; Thermo Scientific) with a Zebron ZB-FFAP GC column (30 m by 0.25 mm; film thickness, 0.25 μm; Phenomenex, United Kingdom). The injection volume was 1 μl with a 10:1 split ratio (split flow, 12 ml min−1), and the same GC conditions described above (without FID) were used. The mass spectrometry conditions in the positive mode were as follows: ion source, 250°C; mass range, 45 to 650 Da; scan rate, 1,783 atomic mass units/s. Free fatty acids were identified by coretention with standards (Sigma) and the use of the National Institute of Standards and Technology (NIST; v2.0; http://www.nist.gov/srd/nist1a.cfm) and MassBank (32) mass spectral search libraries (the mass spectra of the identified peaks are provided in Supplemental Dataset S1 in the supplemental material).
Starch and pigment analysis.Total starch was quantified using a commercial amyloglucosidase enzymatic kit (starch assay kit SA-20; Sigma-Aldrich) according to the manufacturer's instructions. Total chlorophyll and carotenoid concentrations were determined after extraction of the cell pellet with dimethylformamide using the equations by Inskeep and Bloom (33) and Wellburn (34), respectively.
Confocal imaging.Prior to imaging, 5 μl of Nile Red (50 μg ml−1 acetone) was added to 0.5 ml of culture to stain neutral lipids (TAGs). The cells were visualized under a confocal fluorescence microscope (Leica DM6000B; scan head, Leica TCS SP5) using the following settings: a ×63 (numerical aperture, 1.32) oil lens, excitation at 488 nm at 40%, and emission at 550 to 590 nm for Nile Red fluorescence and 685 to 730 nm for chlorophyll fluorescence.
Statistics.Statistical significance of values between days 0 and 10 was determined by the t-test function in the Excel program (Microsoft Office 2007).
RESULTS
Accumulation of lipids and starch in cells grown under nitrogen deficiency.To study the effect on TAG production and autophagy of different C sources—either autotrophic (HSM plus CO2) or mixotrophic (TAP, containing acetate)—C. reinhardtii WT-12 and sta6 mutant cells were grown to late exponential phase, pelleted, washed, and transferred to fresh medium with either a normal level of N (1N) or a low level of N (0.1N) under these two C regimes. Complete removal of N from the medium results in the abrupt cessation of cell growth (2), so to increase TAG productivity, reduced N is preferable (8). Over a 10-day period, Nile Red staining and confocal microscopy revealed that cells grown in low-N medium accumulated many large neutral lipid oil bodies that were not seen in algae grown in normal N (Fig. 1). These visual changes were quantified by extracting the lipid and measuring the TAG content by GC analysis, and starch content was determined by enzymatic analysis of glucose after acid hydrolysis.
Confocal microscopy images of C. reinhardtii (WT-12) cells grown in normal N or low N (1/10) medium. Cells were stained with Nile red to visualize neutral lipid (green fluorescence) and overlaid with red chlorophyll fluorescence. Images of cells at 2 days and 10 days after transfer are shown; lipid droplets are clearly visible within the cells after 10 days of growth in low N.
Figure 2 presents the results of TAG and starch levels in the cells over the time course of N deprivation. As expected, in 1N medium (Fig. 2, solid lines), little or no TAG was produced, irrespective of the C source. In cells grown mixotrophically in 0.1N TAP medium, both WT-12 and the sta6 mutant produced TAGs over the time course (Fig. 2, dashed lines). This increase was statistically significant between days 0 and 10 (P < 0.01). However, whereas in WT-12 TAG accumulation occurred over the first 2 days and then stayed more or less constant, in the sta6 mutant, the TAG level increased for a longer period and finally reached levels some 4-fold higher than those in WT-12 (161 versus 37 mg TAG g [dry weight] of cells−1). These values are in line with those of others who found higher levels of lipid in starchless mutants than in those with a normal starch content (7).
Accumulation of total triacylglycerides and starch (mg metabolite per gram [dry weight {d.wt.}] of cells) of Chlamydomonas reinhardtii (WT-12 or the sta6 mutant) grown in either normal-nitrogen (1N) or low-nitrogen (0.1N) HSM or TAP medium for 10 days. Data are means and variances of 2 to 4 replicates.
The profile of TAG production (and starch in WT-12) in cells grown autotrophically (HSM plus CO2) was quite different (Fig. 2). Under all conditions, after an initial lag of 2 to 3 days, there was continual synthesis throughout the time course (P ≤ 0.01 between days 0 and 10), and the total amount that accumulated after 10 days in the sta6 mutant was only slightly larger than that in WT-12 (109 versus 84 mg TAG g [dry weight] of cells−1). This gradual increase in TAG accumulation in WT-12 continued to a concentration of 167 mg TAG g (dry weight) of cells−1 after 15 days of growth in 0.1N HSM (data not shown).
Table 1 shows the data expressed as the amount of TAG produced per unit of time per g (dry weight) of cells or liter of culture (i.e., productivity), and it is clear that under autotrophic conditions, TAG production was more or less constant (∼10 to 15 mg TAG g [dry weight] of cells−1 day−1 from days 4 to 10), whereas under mixotrophic conditions, productivity ceased after the initial burst. However, there was an important difference between the WT-12 and sta6 strains, namely, that despite the higher level of TAG per unit dry cell weight in the sta6 mutant, the overall production of TAG was much greater in WT-12 cultures (expressed as mg TAG liter−1 day−1) than in the starchless mutant cultures.
Productivity of TAGs in Chlamydomonas reinhardtii WT-12 or sta6 mutant grown in 0.1N HSM or TAP medium for 10 days
In the WT-12 strain, 0.1N TAP and 0.1N HSM conditions also led to a significant increase in starch content (P = 0.069 and P = 0.006, respectively, between days 0 and 10) (Fig. 2). When the accumulation of TAG was plotted against starch levels (Fig. 3), the two storage molecules showed a linear relationship up to ∼500 mg starch g (dry weight) of cells−1. Thereafter, in the cells growing in 0.1N HSM, the amount of TAG steadily increased, with little or no increase in starch content, suggesting that under these conditions TAG and starch production is uncoupled.
Correlation between total triacylglycerides and starch (mg metabolite per gram [dry weight] of cells) of C. reinhardtii (WT-12) grown in low-nitrogen (0.1N) HSM or TAP medium. Each point represents mean TAG and starch amounts from day 0 to day 10, shown individually in Fig. 2. Lines were fitted manually. Data are means and variances of 2 to 4 replicates.
Evidence of autophagy and de novo synthesis of fatty acyl groups for TAG production.Our criteria for assessing autophagy in cells grown in 0.1N-containing medium were (i) a decrease in chlorophyll and carotenoid levels (9, 35), (ii) a decrease in levels of polar (i.e., membrane) lipids with a concurrent increase in TAG levels (7), and (iii) no change in total fatty acid levels (pooled from membrane lipids and TAGs), since this would require de novo fatty acid synthesis (36). We measured the levels of the individual classes of metabolites in the samples over the time course, and the results are shown in Fig. 4. In WT-12 cells, chlorophyll levels declined significantly as a result of 0.1N treatment under both mixotrophic (TAP) and autotrophic (HSM) conditions (P ≤ 0.001 between days 0 and 10), whereas there was no significant change in chlorophyll over time in the sta6 mutant. A similar significant decline in carotenoid levels was seen for WT-12 (P < 0.01), though in the sta6 mutant there was a slight increase in cells grown in 0.1N TAP medium and 0.1N HSM. In fact, the level of carotenoids in WT-12 and sta6 mutant cells grown in TAP medium was lower than that in cells grown in HSM from the start (Table 2, day 0), perhaps because the supply of organic carbon suppresses biogenesis of the photosynthetic apparatus (12).
Concentrations of total chlorophyll, carotenoids, polar lipids, free fatty acids, and total FAMEs (mg metabolite per gram [dry weight] of cells) of Chlamydomonas reinhardtii (WT-12 or the sta6 mutant) grown in either normal-nitrogen (1N) or low-nitrogen (0.1N) HSM or TAP medium for 10 days. Data are means and variances of 2 to 4 replicates.
Allocation of carbon to different classes of metabolites in Chlamydomonas reinhardtii WT-12 or the sta6 mutanta
The polar lipid content decreased over time in wild-type cells grown in 0.1N HSM (P = 0.02) (Fig. 4), and although in 0.1N TAP the decline was more modest, levels were still much lower than in cells grown in 1N TAP (P = 0.028 at day 10). In contrast, in sta6 mutant cells grown in HSM, there was no change in polar lipids, but a substantial decrease was observed in TAP medium (P = 0.04 from day 0 to day 10), despite little or no change in chlorophyll.
To determine if there was de novo synthesis of fatty acids under these conditions, we then estimated the total fatty acids—i.e., the sum of fatty acids in TAGs, polar lipids, and free fatty acids—by transesterification of the lipid fraction to form fatty acid methyl esters (FAMEs), which were then quantified by GC. In the wild-type cells, there was essentially no difference in total FAME levels in cells under any condition (Fig. 4), indicating that there was no net fatty acid synthesis. In contrast, in sta6 mutant cells, there was a 2- to 5-fold increase in FAMEs under 0.1N conditions in both HSM (P = 0.014 from days 0 to 10) and TAP (although the difference was not statistically significant due to the larger variances in 0.1N samples). The identifiable fatty acyl methyl esters of the total FAME pool were predominantly 16:0 > 18:3 > 18:1 > 18:2 > 18:0 (Fig. 5). In WT-12, there were minor alterations in the composition of fatty acyl groups in cells grown in 0.1N medium. In contrast, for sta6 mutant cells, substantial changes in the profile of fatty acids were seen; specifically, there were spectacular increases in 16:0 fatty acids in sta6 mutant cells grown in 0.1N medium.
Concentrations of total fatty acyl methyl esters (acid catalyzed) of Chlamydomonas reinhardtii (WT-12 or the sta6 mutant) grown in either normal-nitrogen (1N) or low-nitrogen (0.1N) HSM or TAP medium for 10 days. The mean value for 16:0 fatty acids in the sta6 mutant in 0.1N TAP is 160 mg g (dry weight)−1. Data are means and variances of 2 replicates. Positions of unsaturations are given in parentheses.
Lastly, we looked at the free fatty acid (FFA) pool (Fig. 4). In WT-12 cells, there was no alteration in FFAs under any condition, and although there was some alteration in FFAs in the sta6 mutant grown in 0.1N HSM, it was not significant. In contrast, for sta6 mutant cells grown in 0.1N TAP, there was a dramatic increase in these metabolites, with levels rising from almost negligible to about 200 mg g (dry weight) of cells−1 (P = 0.015 from day 0 to 10). This increase in FFAs, detected by GC-FID, was confirmed by two independent methods, thin-layer chromatography and GC-MS. Separation of the crude lipid extract by TLC, followed by staining with 0.2% (wt/vol) 8-anilino-1-naphthalenesulfonic acid in methanol and visualization under UV light, showed accumulation in the region corresponding to FFAs (by comparison with standards) for the sta6 mutant in 0.1N TAP but not in any other treatment (data not shown). The identity was confirmed by eluting the FFAs from the TLC plate, esterification to FAMEs, and then analysis by GC-MS. The underivatized FFAs from the total crude extract were also analyzed by GC-MS. The mass profile of each peak was compared against reference standards from the NIST (http://www.nist.gov/srd/nist1a.cfm) and MassBank (www.MassBank.jp) mass spectral libraries, and FFAs were identified as 14:0, 16:0, 16:1, 18:0, 18:1, 18:2, 18:3, and 18:4 (see Supplemental Dataset S1 in the supplemental material); the 18:4 FFA was not detected using GC-FID.
To establish whether the FFA accumulation was caused by degradation of glycerolipids during extraction, a known lipase inhibitor, isopropanol, was added to representative samples during the extraction procedure. Although the absolute amount of FFAs was reduced somewhat, with a corresponding increase in the amounts of polar lipids, there was still over 100 mg FFAs per g (dry weight) of cells (∼10% of cell dry weight) in the sta6 mutant grown for 8 days in 0.1N TAP; this was 3- to 4-fold greater than the amount measured in cells grown in 1N TAP.
Carbon allocation and strain viability.The allocation of carbon (C) to each class of metabolite varied between the wild type and the sta6 mutant and between the HSM and TAP growth media (Table 2, day 10). Under 1N conditions, most of the C in WT-12 and sta6 mutant cells was in polar lipids and chlorophyll (on the basis of mg C g [dry weight] of cells−1 and mg C liter culture−1; Table 2, day 0). After 10 days of low N availability, WT-12 cells allocated most of the C to starch, followed by TAG and polar lipids, whereas sta6 mutant cells allocated most of their C to free fatty acids or polar lipids, followed by TAGs. The amount of C allocated to TAGs was the largest in sta6 mutant cells grown in TAP (122 mg C g [dry weight]−1) and the smallest in wild-type cells grown in TAP (28 mg C g [dry weight]−1). However, when expressed as mg C per liter of culture, the culture that had the largest amount of C allocated to TAG was wild-type cells grown in HSM (133 mg C liter−1) and the culture that had the smallest amount was sta6 mutant cells grown in HSM (22 mg C liter−1), essentially due to the decreased growth of the sta6 mutant cells growing under 0.1N conditions over time (Fig. 6). To evaluate further the impact of reduced growth in the sta6 mutant, we grew the cells on HSM (which has no carbon source) or TAP (which includes a carbon source) agar plates under either a constant light or a diurnal light (12 h of light/12 h of dark) regime for 7 days. As the sta6 mutant cannot accumulate starch for use as a C source for respiration in the dark period, we hypothesized that the growth of these cells would be severely affected when grown under diurnal conditions, and this was indeed what was observed. Although the sta6 mutant was able to grow on TAP under both light regimes, it did not grow well on HSM, and under diurnal conditions it did not grow at all (Fig. 7). In contrast, WT-12 was able to grow, albeit more slowly than under constant light.
Biomass accumulation of Chlamydomonas reinhardtii (WT-12 or the sta6 mutant) cultures grown in either normal-nitrogen (1N) or low-nitrogen (0.1N) HSM or TAP medium for 10 days expressed as dry biomass and optical density (absorbance [Abs] at 750 nm). Data are means and variances of 2 replicates.
Growth of Chlamydomonas reinhardtii (WT-12 or the sta6 mutant) colonies grown on either 2% HSM or 2% TAP agar for 7 days under constant light or diurnal light (12 h of light/12 h of dark). Each colony originated from 5-μl aliquots of cultures that were normalized to ∼9,400 cells per 5 μl. Colony sizes were between 2 and 4 mm in diameter.
DISCUSSION
Differences in carbon supply affect the initial rate of TAG accumulation.In this study, we have compared TAG production initiated by N limitation in wild-type and sta6 mutant cells of C. reinhardtii grown under autotrophic versus mixotrophic conditions. The main hypothesis was that acetate, which could be used directly as a precursor for de novo acetyl-CoA synthesis, would increase the initial rate of TAG production. Our results did indeed show that the initial production of TAG was influenced by the type of C supply. There was a 9-fold and 25-fold increase in TAG production in WT-12 and sta6 mutant cells, respectively, when cells were grown in TAP than when cells were grown in HSM (Fig. 2). However, this rate of increase was not sustained, and the level of TAG reached a maximum after 2 or 4 days of N limitation in wild-type and sta6 mutant cells, respectively. This indicates that the production of TAG in cells provided with acetate is limited by C availability. The effect of C limitation in cells grown in acetate has also been tested by Goodson et al. (37), where they observed an increase in the number and size of oil bodies in cells given an acetate boost after 2 days of N starvation. Also, the fact that in our experiments WT-12 and sta6 mutant cells in autotrophic medium with a constant supply of CO2 gradually increased their amount of TAG over time implies that continued TAG production is related to C availability in the growth medium. This is in line with the findings of Fan et al. (4), who also concluded that C availability is a key factor in the control of C partitioning between starch and TAG in C. reinhardtii, with TAG production lagging behind starch production.
Interestingly, the trends in the increase in TAG concentrations in our study were mirrored in the starch concentrations in N-limited wild-type cells, in that the initial rate of starch accumulation in cells grown in TAP was faster than the rate of starch accumulation in cells grown in HSM, but the relationship between the cellular content of the two storage molecules was linear up to a maximum of ∼500 mg starch g (dry weight)−1 (Fig. 3). Similarly, Siaut et al. (7) found that after 4 days of growth in N-depleted TAP medium, TAG increased steadily but starch remained constant.
There is less cellular autophagy of polar lipids in wild-type cells supplied with acetate but not in sta6 mutant cells.We also expected to observe less cellular autophagy of the structural lipids in thylakoid membranes in cells supplied with acetate, as the rate of de novo synthesis of fatty acyl groups would be sufficient to supply fatty acids for TAG production, especially in sta6 mutant cells, where TAG production is greater than that in wild-type cells. Metabolite changes (decreases in polar lipids and pigments, no change in total and individual fatty acids) indicated that TAG production was the result of autophagy in WT-12 cells grown in HSM, and the level of autophagy was less (in terms of the decrease in polar lipid concentrations) in cells grown in TAP. However, this was not observed in sta6 mutant cells in HSM, since there was no change in polar lipid or chlorophyll, and the total fatty acids increased, with an alteration in the profile indicating de novo fatty acid synthesis. This indicates that the TAG production was mainly from the de novo synthesis of fatty acids rather than autophagy of cellular membranes (17, 18, 36).
Interestingly, the metabolic response in sta6 mutant cells grown in 0.1N TAP did indicate some autophagy, since there was a decrease in polar lipids. This mix of autophagy and de novo synthesis of fatty acids could explain why there was a higher concentration of TAG in sta6 mutant cells grown in TAP than those grown in HSM. Our observation differs from that described by Li et al. (11), who showed comparable concentrations of TAG production in a starchless mutant grown in N-deficient HSM and TAP. However, their result was based on only 2 days of N depletion. They also suggested that the accumulation of TAG was predominantly from fatty acids synthesized de novo from photosynthetically fixed C, rather than acetate, as the production of TAG in the starchless cells was severely reduced when cells were grown in TAP with no light.
The difference in the carotenoid levels between the wild-type and sta6 mutant cells might also explain the strain-specific autophagous responses to low N availability. Perez-Perez et al. (35) demonstrated that a decrease in carotenoids can trigger autophagous processes in wild-type C. reinhardtii. This correlation was also observed in our study, where wild-type cells, with a reduction in carotenoids caused by N limitation, displayed autophagous responses, whereas the sta6 mutant cells with unaltered levels of carotenoids did not.
The fatty acid content in wild-type and sta6 mutant cells grown in low-N TAP was comparable to that measured by James et al. (18), who reported 116 mg FAMEs g (dry weight) of cells−1 and 649 mg FAMEs g (dry weight) of cells−1 in wild-type (cc-124) and starchless (BAF-J5) mutants of C. reinhardtii, respectively. However, unlike the studies by James et al. (18, 38), we were unable to detect 16:3 and 16:4 fatty acids. They, alongside Msanne et al. (6), also reported that the major fatty acids were C16 and C18. This higher concentration of total fatty acids, measured as FAMEs, in sta6 mutant cells than wild-type cells is partly due to the substantial increase in the FFA component of the former. In the wild-type cells, excess photosynthate can be stored as starch, but this is not possible in sta6 mutant cells, where the C is stored as TAG, but it would appear that under mixotrophic conditions, there is further derailing of metabolism leading to a substantial increase in FFAs. This observation is significant, as the increased availability of FFAs could be exploited for further metabolic engineering applications. Although total FAMEs have been reported to be higher in sta6 mutant cells than wild-type cells (12), to our knowledge, this is the first report of the measurement of FFAs.
Increased TAG production per cell does not relate to increased TAG production per culture.An understanding of the resources required for optimal TAG production in algal cells is of paramount importance to biofuel industries, where algal cultivation and N-fertilizer constitute the main energy and economic demands in life cycle analysis models (1, 22, 39). Our research has shown that mixotrophic growth increases the initial rate of TAG production per unit dry weight. However, when expressed as mg C per liter of culture, the most TAG was produced by the wild type grown autotrophically (133 mg C per liter of culture). The main advantage that the sta6 mutant had over the wild-type strains was that TAG was produced in the first 2 days of N limitation when cells were grown in TAP (23.6 mg TAG liter−1 day−1), whereas the maximum TAG productivity for wild-type cells was during days 8 and 10 when cells were grown in HSM (34.3 mg TAG liter−1 day−1). Our values for the wild type and the sta6 mutant, although slightly lower than those quoted by Li et al. for the sta6 mutant (50 mg TAG liter−1 day−1) (10), are within the range of TAG productivity values quoted for many algal species by Griffiths and Harrison (8). The reason for the difference in productivity between WT-12 and the sta6 mutant was essentially because of the decreased biomass production of the sta6 mutant cells over time when grown in low-N medium (Fig. 6), as already previously described by James et al. (38) and Li et al. (10). Moreover, the sta6 mutant was not able to survive under diurnal light conditions (Fig. 7), a phenotype also described in starchless mutants of Arabidopsis (40), presumably because the lack of starch stores impaired respiration during the dark period. This previously unreported observation would prevent cultivation of the sta6 mutant under outdoor conditions, which would be most likely to prevail in an industrial context.
Finally, it should be emphasized that significant TAG production was triggered only when cells were subjected to low N availability and not when the C/N balance was altered by the inclusion of fixed carbon as acetate, since little TAG was produced in 1N TAP. This has been observed in other algal species (8, 41). Moreover, higher productivity over a sustained period was observed under autotrophic conditions than under conditions with the addition of fixed carbon (Table 1). The supply of gaseous CO2 is likely to be more economically sustainable than the need to supply an exogenous organic carbon source.
ACKNOWLEDGMENTS
We thank Maria Zori and Geraldine Heath for excellent technical support.
This research was supported by funds from Shell Research Ltd. A.C.P.L. is supported by a studentship from the United Kingdom Biotechnology and Biological Sciences Research Council (BBSRC) Doctoral Training Partnership.
FOOTNOTES
- Received 24 July 2013.
- Accepted 26 December 2013.
- Accepted manuscript posted online 10 January 2014.
Supplemental material for this article may be found at http://dx.doi.org/10.1128/EC.00178-13.
- Copyright © 2014, American Society for Microbiology. All Rights Reserved.
REFERENCES
