Rapid induction of lipid droplets in Chlamydomonas reinhardtii and Chlorella vulgaris by Brefeldin A

Abstract

Algal lipids are the focus of intensive research because they are potential sources of biodiesel. However, most algae produce neutral lipids only under stress conditions. Here, we report that treatment with Brefeldin A (BFA), a chemical inducer of ER stress, rapidly triggers lipid droplet (LD) formation in two different microalgal species, Chlamydomonas reinhardtii and Chlorella vulgaris. LD staining using Nile red revealed that BFA-treated algal cells exhibited many more fluorescent bodies than control cells. Lipid analyses based on thin layer chromatography and gas chromatography revealed that the additional lipids formed upon BFA treatment were mainly triacylglycerols (TAGs). The increase in TAG accumulation was accompanied by a decrease in the betaine lipid diacylglyceryl N,N,N-trimethylhomoserine (DGTS), a major component of the extraplastidic membrane lipids in Chlamydomonas, suggesting that at least some of the TAGs were assembled from the degradation products of membrane lipids. Interestingly, BFA induced TAG accumulation in the Chlamydomonas cells regardless of the presence or absence of an acetate or nitrogen source in the medium. This effect of BFA in Chlamydomonas cells seems to be due to BFA-induced ER stress, as supported by the induction of three homologs of ER stress marker genes by the drug. Together, these results suggest that ER stress rapidly triggers TAG accumulation in two green microalgae, C. reinhardtii and C. vulgaris. A further investigation of the link between ER stress and TAG synthesis may yield an efficient means of producing biofuel from algae.

Figure 1. BFA treatment induces lipid droplet (LD) formation in Chlamydomonas reinhardtii strain CC-503.

(A) Images of Nile red-stained LDs in BFA-treated cells. Images were acquired after cells were incubated for 4 h with BFA at a concentration of 75 µg mL⁻¹. DMSO was used as a solvent control (Con). Bar = 10 µm. Nile red fluorescence was viewed using either the 465–495 nm excitation/515–555 nm emission channel (FITC, green) or the 540/25 nm excitation/605/55 nm emission channel (TRITC, red). (B) Time-dependent effect of BFA treatment on the fluorescence intensity of LDs stained with Nile red. The fluorescence was quantified using a fluorescence spectrophotometer with a 488-nm excitation filter and a 565-nm emission filter. AU = arbitrary units. (C) Growth of the Chlamydomonas culture treated with BFA, measured by reading at OD₇₅₀. (D) Quantification of Nile red fluorescence in individual BFA-treated cells by flow cytometry. We examined 20,000 cells subjected to two different conditions (DMSO solvent control, and BFA at 75 µg mL⁻¹). AU = arbitrary units.

Figure 2. Dose-dependence and growth phase-dependence of the effect of BFA treatment on LD formation in Chlamydomonas reinhardtii CC-503.

(A) Fluorescence intensity of LDs stained with Nile red in cells treated with different BFA concentrations (25, 50, and 75 µg mL⁻¹) for the indicated durations. (B) Nile red fluorescence intensity of LDs in cells treated at different stages of cell culture with either 75 µg mL⁻¹ BFA (closed squares) or DMSO (solvent control, Con; open triangles). Cells were grown in normal TAP medium, and at 24-h intervals, the number of cells in culture (closed circles) was analyzed using a hemocytometer, and the 8-h treatment with BFA was started. AU = arbitrary units.

Figure 3. Changes in lipid composition of BFA-treated Chlamydomonas reinhardtii CC-503.

Cells were grown to the mid-log phase, treated with BFA (75 µg mL⁻¹) or DMSO (solvent control, Con) for 8 h, and then lipids were analyzed. (A) Accumulation of TAGs on a TLC plate as revealed by staining with primuline. The bands in the box are TAGs with the same Rf value as soybean TAGs. (B) BFA-treated cells accumulate higher amounts of TAGs than control cells. (C) Comparison of fatty acid compositions of TAGs isolated from BFA-treated and control (Con) cells. In (B) and (C), three replicates were averaged, and the SEs are shown. Significant differences, as determined by Student’s t-test, are indicated by asterisks (*P<0.05, **P<0.01). (D) Comparison of the major lipid classes between BFA-treated and control (Con) cells. Averages from two replicate experiments and their standard deviations are shown. PI, phosphatidylinositol; PE, phosphatidylethanolamine; PG, phosphoglyceride; DGTS, diacylglyceroltrimethylhomoserine; SQDG, sulfoquinovosyl-diacylglycerol; MGDG, monogalactosyldiacylglycerol; and DGDG, digalactosyldiacylglycerol.

Figure 4. BFA-induced LD formation is independent of the presence of an acetate or nitrogen source in the medium.

(A) Nile red fluorescence from CC-503 cells grown in normal TAP (+N, +Ac) medium, transferred to medium with or without acetate, and incubated in continuous light. After an 8-h incubation, Nile red fluorescence was measured for control (Con: DMSO solvent control) and BFA-treated cells using a fluorescence spectrophotometer equipped with a 488-nm excitation filter and a 565-nm emission filter. AU = arbitrary units. (B) Nile red fluorescence of CC-503 cells grown in nitrogen-replete or nitrogen-deficient medium for 1 day, and then treated with BFA for 8 h. AU = arbitrary units. (C) Total TAGs from control (Con) or BFA-treated cells starved in nitrogen-deficient medium for 2 days. DMSO and BFA treatment lasted for 3 h. In (b) and (c), three replicates were averaged, and the SEs are shown. Significant differences, as determined by Student’s t-test, are indicated by asterisks (*P<0.05, **P<0.01). (D) Fatty acid composition of the TAGs extracted from the solvent control (Con) and BFA-treated cells. Both samples had been starved of nitrogen for one day. Averages from three replicate experiments are shown. No statistically significant differences were found.

Figure 5. Expression levels of ER stress marker genes in BFA- and DTT-treated Chlamydomonas reinhardtii CC-503.

Cells in mid-log phase culture were treated with 6 mM DTT dissolved in TAP solution for 2 h, or with BFA (75 µg mL⁻¹) dissolved in DMSO for 8 h. Control cells were treated with TAP solution or DMSO, respectively. Transcript levels of BiP homologs (g1475 and Cre02.g080600) and a SAR1 homolog (Cre11.g468300) were analyzed, and fold changes compared with the expression level in the control samples are presented. The averages and standard errors from two independent experiments are shown. Significant differences, as determined by Student’s t-test, are indicated by asterisks (*P<0.05, **P<0.01).

Figure 6. BFA treatment induces LD formation in Chlorella vulgaris.

(A) A 4 h BFA (75 µg/ml) treatment increases Nile red fluorescence (left) without altering cell density (right). Nile red fluorescence resulting from LDs was quantified using a fluorescence spectrophotometer. AU = arbitrary units. Cells represented by the right-most two bars of each panel were sonicated for 30 s at the 40 watt (W) setting of the sonicator (Vibra Cell™, VC 130PB). Averages from three replicate experiments, and their standard errors are shown. (B) Images of Chlorella cells treated with 75 µg mL⁻¹ BFA and stained with Nile red. Bar = 10 µm. DMSO was used as a solvent control (Con).

Figure 7. A hypothetical model explaining BFA-induced LD formation.

1. BFA inhibits trafficking of vesicles that deliver proteins and lipids to their final destinations (Lippincott-Schwartz et al., Cell, 60 (1990) 821–836). 2. BFA-induced perturbation of vesicle trafficking results in retrograde transport and Golgi-ER aggregation (Lippincott-Schwartz et al., Cell, 60 (1990) 821–836). 3. Failure to deliver proteins and lipids to their final destination causes proteins to accumulate (Lippincott-Schwartz et al., Cell, 60 (1990) 821–836), and disrupts sterol homeostasis (Stephen M. et al., The International Journal of Biochemistry & Cell Biology, 39 (2007) 1843–1851). 4. A portion of the lipids from the disturbed membranes is converted into TAGs, which form LDs. 5. The LDs thus formed may provide a site for sequestering immature proteins, thereby protecting the cell from further damage (Mérigout et al., FEBS letters, 518 (2002) 88–92; Fei et al., Biochem. J, 424 (2009) 61–67). Since the effect of the drug is rapid, the drug may be a useful trigger for recycling intracellular membranes to TAGs, a lipid form more easily processed to fuel than membrane lipids.