The biochemistry of headgroup exchange during triacylglycerol synthesis in canola

Abstract

Many pathways of primary metabolism are substantially conserved within and across plant families. However, significant differences in organization and fluxes through a reaction network may occur, even between plants in closely related genera. Assessing and understanding these differences is key to appreciating metabolic diversity, and to attempts to engineer plant metabolism for higher crop yields and desired product profiles. To better understand lipid metabolism and seed oil synthesis in canola (Brassica napus), we have characterized four canola homologues of the Arabidopsis (Arabidopsis thaliana) ROD1 gene. AtROD1 encodes phosphatidylcholine:diacylglycerol cholinephosphotransferase (PDCT), the enzyme that catalyzes a major flux of polyunsaturated fatty acids (PUFAs) in oil synthesis. Assays in yeast indicated that only two of the canola genes, BnROD1.A3 and BnROD1.C3, encode active isozymes of PDCT, and these genes are strongly expressed during the period of seed oil synthesis. Loss of expression of BnROD1.A3 and BnROD1.C3 in a double mutant, or by RNA interference, reduced the PUFA content of the oil to 26.6% compared with 32.5% in the wild type. These results indicate that ROD1 isozymes in canola are responsible for less than 20% of the PUFAs that accumulate in the seed oil compared with 40% in Arabidopsis. Our results demonstrate the care needed when translating results from a model species to crop plants.

Keywords: ROD1; fatty acids; metabolic diversity; seed lipid metabolism; seed oil; triacylglycerol.

Figure 1.. The metabolic network of reactions contributing to triacylglycerol (TAG) synthesis in oilseeds.

See text for details of pathway possibilities. The symmetrical phosphatidylcholine:diacylglycerol cholinephosphotransferase (PDCT) reaction that is the subject of this paper is shown by the double-headed red arrow. Abbreviations: CPT, cholinephosphotransferase; DAG, diacylglycerol; DGAT, diacylglycerol acyltransferase; FAD, fatty acid desaturase; FAS, fatty acid synthase; G3P, glycerol-3-phosphate; GPAT, G3P acyltransferase; LACS, long-chain acyl-CoA synthetase; LPA, lysophosphatidic acid; LPC, lysophosphatidylcholine; LPAT, lysophosphatidic acid acyltransferase; LPCAT, LPC acyltransferase; PA, phosphatidic acid; PAP, PA phosphatase; PC, phosphatidylcholine; PDAT, phospholipid:diacylglycerol acyltransferase; PDCT, phosphatidylcholine:diacylglycerol cholinephosphotransferase; PLA, phospholipase A; PLC, phospholipase C; PLD, phospholipase D; PUFA, polyunsaturated fatty acid.

Figure 2.. Expression of BnROD1 genes during seed filling in Brassica napus.

Data are from RNA sequencing analyses of seed harvested at six stages of development. The course of oil accumulation is indicated by the orange line (right axis scale). FPKM, reads per kilobase per million reads in the database.

Figure 3.. BnROD1.A3 and BnROD1.C3 complement an Arabidopsis rod1 mutant.

Data are seed fatty acid compositions from plants of wild type (WT), the rod1–1 mutant and rod1–1 expressing either a BnROD1.A3 (a) or BnROD1.C3 (b) transgene. Means ± SE (n = 9). Statistical principal component analysis indicates that the WT and each complemented line are highly similar and very different from the rod1–1 mutant.

Figure 4.

Two BnROD1 genes encode active phosphatidylcholine:diacylglycerol cholinephosphotransferase (PDCT) in microsomes from yeast cells. Complementary DNAs of AtROD1 and four BnROD1 isogenes were expressed in yeast. Synthesis of ¹⁴C-phosphatidylcholine (¹⁴C-PC) from ¹⁴C-diacylglycerol (¹⁴C-DAG) demonstrates PDCT activity. Lipid extracts from yeast microsomes were separated by thin-layer chromatography and visualized in a phosphorimager.

Figure 5.. Expression of an RNA interference (RNAi) construct targeting BnROD1.A3 and BnROD1.C3 in seeds increases 18:1 content compared with the wild type (WT).

The fatty acid compositions of WT and two independent RNAi lines are shown. Means ± SE (n = 5). Statistical analyses (t-test; P-values shown) indicate highly significant differences in 18:1 and 18:2 between the RNAi lines and WT.

Figure 6.. Assays of phosphatidylcholine:diacylglycerol cholinephosphotransferase (PDCT) activity in yeast microsomes.

Assays of phosphatidylcholine:diacylglycerol cholinephosphotransferase (PDCT) activity in yeast microsomes for mutants of (a) BnROD1.A3 and (b) BnROD1.C3. Methods are as described in the caption to Figure 4. ¹⁴C-PC, ¹⁴C-phosphatidylcholine; ¹⁴C-DAG, ¹⁴C-diacylglycerol.

Figure 7.. BnROD1.C3 is not expressed in a Bnrod1.c3–1 mutant line.

(a) Using RNA from developing seeds, primers detect BnROD1.C3 transcript in the wild type (WT) but not the Bnrod1.c3 mutant line. (b) Amplification of the BnDGAT1 transcript indicates RNA preparations are of similar quality. The left lane in each panel shows molecular weight markers. Expected RT-PCR product sizes are 728 bp for BnRod1.C3 and 1486 bp for BnDGAT1.

Figure 8.. Increased 18:1 content in Bnrod1 mutants.

Average seed fatty acid compositions for five samples each of wild type (WT), the Bnrod1.a3–1 and Bnrod1.c3–1 mutants, and a Bnrod1.a3–1/c3–1 double mutant grown in a greenhouse. All SE values (<0.5%) were too small to show. Statistical analyses (t-test; P-values shown) indicate highly significant differences in 18:1 and 18:2 between each single mutant and WT, and between the double mutant and each single mutant.