Metabolic engineering of Saccharomyces cerevisiae for production of Eicosapentaenoic Acid, using a novel {Delta}5-Desaturase from Paramecium tetraurelia

Abstract

Very-long-chain polyunsaturated fatty acids, such as arachidonic acid (ARA), eicosapentaenoic acid (EPA), and docosahexaenoic acid (DHA), have well-documented importance in human health and nutrition. Sustainable production in robust host organisms that do not synthesize them naturally requires the coordinated expression of several heterologous desaturases and elongases. In the present study we show production of EPA in Saccharomyces cerevisiae using glucose as the sole carbon source through expression of five heterologous fatty acid desaturases and an elongase. Novel Δ5-desaturases from the ciliate protozoan Paramecium tetraurelia and from the microalgae Ostreococcus tauri and Ostreococcus lucimarinus were identified via a BLAST search, and their substrate preferences and desaturation efficiencies were assayed in a yeast strain producing the ω6 and ω3 fatty acid substrates for Δ5-desaturation. The Δ5-desaturase from P. tetraurelia was up-to-2-fold more efficient than the microalgal desaturases and was also more efficient than Δ5-desaturases from Mortierella alpina and Leishmania major. In vivo investigation of acyl carrier substrate specificities showed that the Δ5-desaturases from P. tetraurelia, O. lucimarinus, O. tauri, and M. alpina are promiscuous toward the acyl carrier substrate but prefer phospholipid-bound substrates. In contrast, the Δ5-desaturase from L. major showed no activity on phospholipid-bound substrate and thus appears to be an exclusively acyl coenzyme A-dependent desaturase.

FIG. 1.

Simplified representation of PUFA biosynthetic pathways from stearic acid (18:0) to ARA, EPA, and DHA.

FIG. 2.

Δ5-Desaturase activities of Ptet1D5D. GC-FID chromatograms of fatty acids from yeast strains expressing Ptet1D5D (a and c) or harboring the empty plasmid pSF28 (b and d), cultivated with supplementation of 0.5 mM 20:3ω6 (a and b) or 20:4ω3 (c and d) fatty acids, respectively.

FIG. 3.

Desaturation efficiencies of Ptet1D5D, OlD5D, OtD5D, LmD5D, and MaD5D. Yeast strains expressing Ptet1D5D, OlD5D, OtD5D, LmD5D, and MaD5D genes were cultivated in the presence of 0.5 mM 20:3ω6 or 20:4ω3 fatty acids. Desaturation efficiencies {[(product)/(substrate + product)] × 100} were calculated using the percentages of substrates and products in total fatty acids. Each value represents the mean ± standard deviation of five biological replicates.

FIG. 4.

Desaturation efficiencies of Ptet1D5D, OtD5D, OlD5D, LmD5D, and MaD5D in a strain endogenously producing 20:3ω6 and 20:4ω3. The Δ5-desaturases were coexpressed with MaD9D, MaD12D, OtD6D, and MaD6E genes. Desaturation efficiencies were calculated as described in the legend of Fig. 3. Each value represents the mean ± standard deviation of three independent experiments.

FIG. 5.

Desaturation efficiencies of Ptet1D5D and Ptet2D5D in a strain endogenously producing 20:3ω6 and 20:4ω3. The Δ5-desaturases were coexpressed with MaD9D, MaD12D, OtD6D, and MaD6E genes. Desaturation efficiencies were calculated as described in the legend of Fig. 3. Each value represents the mean ± standard deviation of three independent experiments.

FIG. 6.

Distribution of 20:3ω6 and total fatty acids in the lipid fractions of a Ptet1D5D-expressing yeast strain during a time course feeding experiment. Exogenous fatty acid substrate (0.5 mM 20:3ω6) was added to the yeast culture at time zero, and samples for lipid analysis were collected at the indicated time points. (a) Absolute amount of 20:3ω6 in each lipid fraction, expressed in mg/OD unit. (b) Absolute amount of total fatty acids in each lipid fraction, expressed in mg/OD unit.

FIG. 7.

Fatty acid profile of the acyl-CoA pool of a Ptet1D5D-expressing yeast strain during a time course feeding experiment. The experiment was carried out as described in the legend of Fig. 6. Samples of 1 OD unit were collected at the indicated time points (for the 24-h samples, aliquots of 8 OD units were collected), and the acyl-CoA fraction was analyzed. *, ARA peak in the 24-h sample.

FIG. 8.

Absolute amounts of ARA in the lipid fractions of Δ5-desaturase-expressing strains during time course feeding experiments. Yeast strains expressing Ptet1D5D, OtD5D, OlD5D, LmD5D, or MaD5D were cultivated as described in the legend of Fig. 6, and samples were taken at the indicated time points. No ARA was detected in any of the lipid fractions at time points 0 or 5 min.