Skip to main content
NCBI home page
Plant Physiology logoLink to Plant Physiology
. 2019 Oct 16;181(4):1468–1479. doi: 10.1104/pp.19.01129

Isoforms of Acyl-CoA:Diacylglycerol Acyltransferase2 Differ Substantially in Their Specificities toward Erucic Acid1

Kamil Demski a,2,3, Simon Jeppson b,2, Ida Lager b, Agnieszka Misztak a, Katarzyna Jasieniecka-Gazarkiewicz a, Małgorzata Waleron a, Sten Stymne b, Antoni Banaś a,4
PMCID: PMC6878005  PMID: 31619508

Despite highly similar amino acid sequences, isoforms of Acyl-CoA:Diacylglycerol Acyltransferase2 from Brassica napus form two distinct groups according to their affinity toward erucic acid.

Abstract

In most oilseeds, two evolutionarily unrelated acyl-CoA:diacylglycerol acyltransferase (DGAT) enzymes, DGAT1 and DGAT2, are the main contributors to the acylation of diacylglycerols in the synthesis of triacylglycerol. DGAT1 and DGAT2 are both present in the important crop oilseed rape (Brassica napus), with each type having four isoforms. We studied the activities of DGAT isoforms during seed development in microsomal fractions from two oilseed rape cultivars: edible, low-erucic acid (22:1) MONOLIT and nonedible high-erucic acid MAPLUS. Whereas the specific activities of DGATs were similar with most of the tested acyl-CoA substrates in both cultivars, MAPLUS had 6- to 14-fold higher activity with 22:1-CoA than did MONOLIT. Thus, DGAT isoforms with different acyl-CoA specificities are differentially active in the two cultivars. We characterized the acyl-CoA specificities of all DGAT isoforms in oilseed rape in the microsomal fractions of yeast cells heterologously expressing these enzymes. All four DGAT1 isoforms showed similar and broad acyl-CoA specificities. However, DGAT2 isoforms had much narrower acyl-CoA specificities: two DGAT2 isoforms were highly active with 22:1-CoA, while the ability of the other two isoforms to use this substrate was impaired. These findings elucidate the importance, which a DGAT isoform with suitable acyl-CoA specificity may have, when aiming for high content of a particular fatty acid in plant triacylglycerol reservoirs.

Many plants utilize triacylglycerols (TAG) as their primary energy storage in seeds. TAG are predominantly found in seeds and mesocarp but appear also in flower petals and pollen grains (Stymne and Stobart, 1987; Li-Beisson et al., 2013). As a rich source of reduced carbon and essential fatty acids, TAG are vital in both human and animal nutrition. In the oleochemical industry, TAG and their fatty acids are used (e.g. in paints, lubricants, and biopolymer production; Metzger and Bornscheuer, 2006; Durrett et al., 2008; Dyer et al., 2008, Bates and Browse, 2012). TAG nutritional value and chemical properties are dependent on the acyl groups they contain. Oil crops such as canola rapeseed (Brassica napus), sunflower (Helianthus annuus), and soybean (Glycine max) contain TAG with dominantly five different fatty acids of 16 and 18 carbons in length and with up to three double bonds. Nonetheless, the Plantae kingdom holds several hundred different fatty acids in seed TAG (Ohlrogge et al., 2018). Some of these more unusual fatty acids have unique industrial uses (Ohlrogge et al., 2018). Erucic acid (cis-Δ13 22:1 fatty acids, hereafter abbreviated as 22:1) is found in many plant oils; high-erucic acid rapeseed seed oil is one of them. Erucic acid is mainly used as a precursor for the slipping agent erucamide (Ecke et al., 1995) but has been eliminated in modern cultivars suitable for consumption through plant breeding (canola type or low-erucic acid rapeseed).

The production and demand for plant oils are expected to increase significantly (Alexandratos and Bruinsma, 2012) due to growing world population and expanded use for nonfood purposes. Research in optimizing plant oil yield and composition may prove essential for worldwide food security and global industrial development.

The final synthesis of TAG occurs in the endoplasmic reticulum. In the glycerol 3-phosphate or Kennedy pathway (Kennedy, 1961), TAG formation occurs through subsequent acylations of sn-1, sn-2, and sn-3 positions of the glycerol backbone by transfers of acyl groups from an acyl-CoA donor (Ohlrogge and Browse, 1995; Gurr et al., 2002). The final enzymatic step in the Kennedy pathway, the acylation of the sn-3 position of diacylglycerol (DAG), is mediated by acyl-CoA:diacylglycerol acyltransferases (DGAT). The acylation of DAG in the formation of TAG in plants can also occur through the acyl-CoA-independent acylation of acyl groups from phosphatidylcholine (Banaś et al., 2000; Dahlqvist et al., 2000).

DGATs are key regulatory enzymes in TAG synthesis and have been shown to considerably influence carbon flow into TAG (Weselake et al., 2009; Liu et al., 2012; Bates et al., 2013). Plants express at least three nonhomologous DGAT isoenzymes: membrane-bound DGAT1 and DGAT2 and the cytosolic metalloprotein DGAT3 (Aymé et al., 2018). DGAT1 has been proven to be crucial in plant oil production in some plants. Suppression of DGAT1 in rapeseed and a mutation in the DGAT1 gene in Arabidopsis (Arabidopsis thaliana) both led to significantly decreased oil content in seeds (Routaboul et al., 1999; Lock et al., 2009). DGAT2s have been shown to have high specificity toward various unusual fatty acids (e.g. ricinoleic acid, vernolic acid, and α-eleosteric acid) in plants rich in such fatty acids in their seed TAG (Shockey et al., 2006; Burgal et al., 2008; Li et al., 2010).

Four homologous Bna.DGAT1 isoforms from rapeseed have been shown to successfully complement TAG synthesis in a TAG-deficient Saccharomyces cerevisiae mutant, and their acyl-CoA substrate specificity has been studied to some extent (Aznar-Moreno et al., 2015; Caldo et al., 2015; Greer et al., 2015, 2016). However, there are no reports regarding the properties of rapeseed Bna.DGAT2 isoforms.

Disclosed herein is our research of Bna.DGAT1 and Bna.DGAT2 in rapeseed oil biosynthesis with focus on acyl-CoA substrate specificity. Modern rapeseed cultivars suitable for human consumption (canola type) lack erucic acid in the seed oil due to extensive plant breeding utilizing the natural variation of erucic acid level of individual plants in cultivated variety Liho (Stefansson and Hougen, 1964). We compared in vitro activities of Bna.DGAT in microsomal preparations of developing seeds in one cultivar that is devoid of erucic acid (MONOLIT) with one (MAPLUS) having 52% erucic acid in its seed oil. Whereas Bna.DGAT specific activities were similar to most tested acyl-CoA in both cultivars, MAPLUS had 6- to 14-fold higher activity with 22:1-CoA than MONOLIT. Subsequently, this finding led us to discover that rapeseed Bna.DGAT2 isoforms have major differences in acyl-CoA substrate specificity toward erucoyl-CoA. Two of them showed very little activity with 22:1-CoA, whereas the other two Bna.DGAT2 isoforms showed close to highest specificity for 22:1-CoA, despite highly similar amino acid sequences between the two groups. The implication of the findings for breeding rapeseed and other oil crops for altered oil qualities is discussed.

RESULTS

TAG Accumulation during Seed Development in MONOLIT and MAPLUS

The TAG contents and composition of the low-erucic acid cv MONOLIT and the high-erucic acid cv MAPLUS were investigated from four developmental stages as described in Supplemental Table S1 and depicted in Figure 1A. TAG content in seeds of both cultivars increased throughout seed development at a similar rate (Fig. 1B), even though MAPLUS had a somewhat higher content of TAG per seed, 1,860 versus 1,410 nmol seed−1, at stage 4. Seeds of both cultivars accumulate TAG at the fastest rate between developmental stage 2 and stage 3. More than half of the TAG present in stage 4 was synthesized during this rapid TAG accumulation.

Figure 1.

Figure 1.

Open in a new tab

Developing seeds of MONOLIT and MAPLUS cultivars at four different stages (A) were quantified on their TAG content (B; seed weight is accounted for in Supplemental Fig. S1). MAPLUS was also quantified on its erucic acid content (C). MONOLIT contained no detectable erucic acid. Three biological replicates were made for each development stage from three different plants from both cultivars. Error bars in B and C indicate sd between biological replicates (n = 3). FA, Fatty acids.

The fatty acid composition of the accumulated TAG was also determined at the four different developmental stages (Supplemental Fig. S1). The proportion of erucic acid in MAPLUS TAG increased from stage 1 to stage 4 from 8% to 52% (Fig. 1C). As with TAG accumulation, 22:1 content increased at the fastest rate between the second and third stages of seed development (Fig. 1C). During this time, 63% of the total erucic acid content in seeds was synthesized. The erucic acid content in cv MONOLIT was below the detection limit at all stages.

In Vitro Acyl-CoA Specificities of Bna.DGAT in Microsomal Fractions of Developing Seeds Differ between MONOLIT and MAPLUS

In order to measure in vitro activity of Bna.DGAT during seed development, seed microsomal fractions were isolated from each developmental stage of both MONOLIT and MAPLUS. A series of in vitro assays were conducted using exogenous substrates ([14C]acyl-CoAs and di-6:0-DAG) in the microsomal fractions. Various acyl-CoAs were used to measure the changes in overall Bna.DGAT specific activity as well as changes in substrate specificities. Of the tested acyl-CoAs, 18:3-CoA was the most preferred acyl donor for seed microsomal fractions derived from both cultivars (Supplemental Fig. S2). Both cultivars showed similar specific activities of Bna.DGAT activity with 18:3-CoA over the four developmental stages, with the highest activity at stage 3 (Fig. 2A). This correlates well with the rapid increase in TAG content between stages 2 and 3. The Bna.DGAT activity for most of the other included acyl donors in this study exhibited a similar pattern, with the highest specific activities at stages 3 and 4 for both MONOLIT and MAPLUS (Supplemental Fig. S2). Noteworthy, however, was the drastic difference in specificity toward 22:1 between the cultivars: Bna.DGAT in the high-erucic acid cv MAPLUS produced 6- to 14-fold more TAG utilizing 22:1-CoA than Bna.DGAT in the low-erucic acid cv MONOLIT (Fig. 2B; Supplemental Fig. S2B). Since Bna.DGAT in microsomal fractions of both cultivars yielded similar activities with 18:3-CoA and other acyl-CoAs, yet exhibited significant differences toward 22:1-CoA, it may indicate that Bna.DGAT isoforms with different acyl specificities are differently active during seed development in the two cultivars.

Figure 2.

Figure 2.

Open in a new tab

DGAT activity with 18:3-CoA (A) and 22:1-CoA (B) in low-erucic acid (MONOLIT) and high-erucic acid (MAPLUS) cultivars throughout seed development. The activity was measured in microsomal fractions obtained from three replicates from three different plants for each development stage and cultivar. Error bars indicate the sd between biological replicates (n = 3).

Bna.DGAT1 Isoforms Show Similar Acyl-CoA Specificities

Of the four known Bna.DGAT1 isoforms, BnaA.DGAT1.a, BnaC.DGAT1.a, BnaC.DGAT1.b, and BnaA.DGAT1.b (Aznar-Moreno et al., 2015; Greer et al., 2016), the first three were confirmed to be expressed in both studied cultivars. We used the published BnaA.DGAT1.b sequence for further experiments, since we were unable to detect it in either cultivar. BnaA.DGAT1.a, BnaC.DGAT1.a, and BnaC.DGAT1.b were found in several (up to four) slightly deviating sequences in each Bna.DGAT1 variant in both MAPLUS and MONOLIT. To assay all these variants would be an insurmountable amount of work. Therefore, here we describe one variant of each BnaA.DGAT1.a, BnaC.DGAT1.a, and BnaC.DGAT1.b, which were cloned from a mixture of cDNA from MONOLIT and MAPLUS.

The four Bna.DGAT1 isoforms were individually overexpressed in the TAG-deficient yeast strain, H1246 (Sandager et al., 2002), for further characterization. All the Bna.DGAT1 isoforms complemented the TAG deficiency of H1246 (Fig. 3A). Microsomal preparations of the transformed yeast were used to investigate the individual DGAT1 acyl donor specificity with [14C]glycerol-labeled di-6:0-DAG as acyl acceptor and with 16:0-CoA, 18:0-CoA, 18:1-CoA, 18:2-CoA, 18:3-CoA, or 22:1-CoA as acyl donor.

Figure 3.

Figure 3.

Open in a new tab

Complementation of TAG synthesis in S. cerevisiae transformants with cloned Bna.DGAT1 (A) and Bna.DGAT2 (B) isoforms. Both photographs present a thin-layer chromatography (TLC) plate after separation of yeast extracts in hexane:diethyl ether:acetic acid (70:30:1). TAG positioning is marked by the black frames. Both images were converted to black and white while applying a blue filter and brightened by approximately 30% in Adobe Photoshop CC 2015. Lanes in A are as follows: 1, BY4742, positive control, wild strain, able to synthesize TAG; 2 to 4 represent S. cerevisiae transformants with pYES-DEST52 containing the particular Bna.DGAT1 isoform: 2, BnaC.DGAT1.a; 3, BnaA.DGAT1.a; 4, BnaC.DGAT1.b; 5, BnaA.DGAT1.b; 6, H1246, S. cerevisiae strain unable to synthesize TAG transformed with nonrecombinant pYES-DEST52. Lanes in B are as follows: 1, BY4742, positive control, wild strain, able to synthesize TAG; 2, H1246, S. cerevisiae strain unable to synthesize TAG transformed with nonrecombinant pYES-DEST52; 3 to 7 represent S. cerevisiae transformants with pYES-DEST52 containing the particular Bna.DGAT2 isoform: 3, BnaA.DGAT2.b; 4, BnaA.DGAT2.c MONOLIT; 5, BnaA.DGAT2.c MAPLUS; 6, BnaA.DGAT2.d; 7, BnaA.DGAT2.e.

The specific activity of the DGAT1 enzymes varied considerably between isoforms (Table 1). Acyl donor specificities of the Bna.DGAT1 isoforms are presented as percentages of its activity with 16:0-CoA, the preferred acyl donor for all the Bna.DGAT1 isoforms except for BnaA.DGAT1.a, which utilized 18:1-CoA somewhat better (Fig. 4). All four Bna.DGAT1 isoforms exhibited otherwise very similar acyl-CoA specificities. 18:1-CoA was the second most preferred acyl-CoA by the other isoforms (Fig. 4). 22:1-CoA was among the least preferred acyl-CoA donors for all studied DGAT1 isoforms (Fig. 4).

Table 1. Specific activities of Bna.DGAT1 isoforms with 16:0-CoA (n = 3 technical replicates).

DGAT1 Isoform Specific Activity
pmol min−1 mg−1 microsomal protein ± sd
BnaA.DGAT1.a 650 ± 16
BnaC.DGAT1.a 545 ± 25
BnaA.DGAT1.b 3,918 ± 348
BnaC.DGAT1.b 323 ± 29
Open in a new tab

Figure 4.

Figure 4.

Open in a new tab

Acyl-CoA specificities of Bna.DGAT1 isoforms with various acyl-CoAs. A, BnaA.DGAT1.a. B, BnaC.DGAT1.a. C, BnaA.DGAT1.b. D, BnaC.DGAT1.b. Activities are expressed as percentages of activity with 16:0-CoA (presented by the white bar in each chart). Error bars indicate the sd between technical replicates (n = 3).

Bna.DGAT2 Isoforms Have Very Different Acyl-CoA Specificities toward 22:1-CoA

Four putative Bna.DGAT2 isoforms were found by searching the National Center for Biotechnology Information (NCBI) database. The identified sequences were used to design primers and identify candidate amplicons from MONOLIT and MAPLUS cDNA (Supplemental Table S2). The isolated Bna.DGAT2.b and Bna.DGAT2.e isoforms were identical to previously published isoforms for both MONOLIT and MAPLUS. Bna.DGAT2.d from both cultivars had three identical amino acid substitutions in the same positions when compared with published sequences. Only Bna.DGAT2.c differed between the two cultivars (those two variants are defined as Bna.DGAT2.c MONOLIT and Bna.DGAT2.c MAPLUS). Since earlier studies of Arabidopsis DGAT2 expression in yeast showed that codon optimization improved activity (Aymé et al., 2014), sequences of the five discovered Bna.DGAT2 variants were codon optimized for yeast and expressed in the H1246 yeast strain. All Bna.DGAT2 isoforms were able to partially complement the TAG deficiency of H1246, albeit only a very small amount of TAG was present in BnaA.DGAT2.e transformant cells (Fig. 3B).

Microsomal fractions were isolated from transformed yeast and DGAT2 specificities were assayed with the same assay conditions as for DGAT1. Although all isoforms showed microsomal activities, the absolute activity with the different acyl-CoAs differed widely, with BnaA.DGAT2.b being far the most active isoform when overexpressed in yeast (Table 2). Out of all tested acyl-CoAs, 18:3-CoA was the most efficient substrate among the DGAT2s (Fig. 5). The activities of the other tested acyl-CoAs are, therefore, are presented as the percentage of their respective 18:3 activity. Bna.DGAT2 isoforms showed a distinct pattern in which the isoforms grouped in pairs of similar acyl-CoA specificity (BnaA.DGAT2.b with two BnaA.DGAT2.c variants and BnaA.DGAT2.d with BnaA.DGAT2.e). BnaA.DGAT2.b and two BnaA.DGAT2.c variants manifested a high specificity toward 18:3-CoA (Fig. 5). Their activities with other acyl-CoAs never reached 35% of their activities with 18:3-CoA, and only very low activities could be observed with 22:1-CoA (Fig. 5). This specificity differed starkly from the two other isoforms, BnaA.DGAT2.d and BnaA.DGAT2.e, which had equally high activity with 22:1-CoA as with 18:3-CoA (Fig. 5). Albeit BnaA.DGAT2.d and BnaA.DGAT2.e still efficiently accepted 18:3-CoA, the other investigated acyl donors were accepted to a higher degree compared with BnaA.DGAT2.b and BnaA.DGAT2.c MAPLUS (Fig. 5). Similar results were obtained with transformants expressing the previously published Bna.DGAT2 variants (Supplemental Table S3; Supplemental Fig. S3). Interestingly, the observed differences in acyl-CoA specificity between the different groups of Bna.DGAT2 correlate to a higher amino acid similarity between BnaA.DGAT2.b and the two BnaA.DGAT2.c variants (almost 100% for MONOLIT and 99% for MAPLUS) and between BnaA.DGAT2.d and BnaA.DGAT2.e (97%). Amino acid sequence similarity between the pairs is between 79% and 80% (Supplemental Table S4).

Table 2. Specific activities of Bna.DGAT2 isoforms with 18:3-CoA, acyl-CoA specificities of which are shown in Figure 5 (n = 3 technical replicates).

DGAT2 Isoform Specific Activity
pmol min−1 mg−1 microsomal protein ± sd
BnaA.DGAT2.b 149,455 ± 3,943
BnaA.DGAT2.c MONOLIT 9,182 ± 643
BnaA.DGAT2.c MAPLUS 29,697 ± 2,914
BnaA.DGAT2.d 19,606 ± 1,586
BnaA.DGAT2.e 186 ± 1
Open in a new tab

Figure 5.

Figure 5.

Open in a new tab

Acyl-CoA specificities of Bna.DGAT2 isoforms with various acyl-CoAs. A, BnaA.DGAT2.b. B, BnaA.DGAT2.c MONOLIT. C, BnaA.DGAT2.c MAPLUS. D, BnaA.DGAT2.d. E, BnaA.DGAT2.e. Activities are expressed as percentages of activity with 18:3-CoA (presented by the white bar in each chart). Error bars indicate the sd between technical replicates (n = 3).

Bna.DGAT1 and Bna.DGAT2 Isoforms Present Different Patterns of Expression throughout Seed Development

Since two groups of Bna.DGAT2 isoforms with different acyl-CoA specificities toward 22:1 were identified, differences in expression of the Bna.DGAT2 isoforms between the two cultivars could potentially be a contributing factor to the differences seen in DGAT activities toward 22:1-CoA observed in microsomal preparations from developing seeds of MONOLIT and MAPLUS. To test this hypothesis, we proceeded to determine the expression patterns of the eight known Bna.DGAT isoforms through the four assigned stages of seed development in both cultivars. The isoforms’ expression is presented relative to the ACT7 and UBC21 housekeeping genes (Fig. 6). BnaA.DGAT1.b expression could not be detected in either MONOLIT or MAPLUS seeds at any stage of development; in line with that, we were not able to clone this isoform from cDNA. The largest change in expression through seed growth occurred in the case of BnaA.DGAT1.a and BnaA.DGAT1.c. Expression of these two isoforms was 4.5- to 7.5-fold higher in the third and fourth stages of seed development in MONOLIT and 2- to 5.5-fold higher in MAPLUS, in comparison with expression in the first two developmental stages. BnaC.DGAT1.b and Bna.DGAT2 isoforms’ expression patterns through seed development did not exhibit such drastic changes (Fig. 6). Small individual changes in the relative expression of BnaA.DGAT2 isoforms were discovered, but the isoforms’ expression decreased as a general trend in the late developmental stages for MAPLUS whereas it remained on an even level in MONOLIT. There seems to be no correlation between the increased DGAT activity toward 22:1-CoA in seed microsomes from MAPLUS (Fig. 1C) seen in the latter stages of development and the relative expression of the two isoforms with specificity toward 22:1-CoA (BnaA.DGAT2.d and BnaA.DGAT2.e). The results do, however, reveal that expression patterns of particular Bna.DGAT isoforms may differ between rapeseed cultivars.

Figure 6.

Figure 6.

Open in a new tab

Relative expression of rapeseed DGAT isoforms throughout seed growth in both studied cultivars, MONOLIT (A and C) and MAPLUS (B and D), in reference to the expression of housekeeping genes ACT7 (A and B) and UBC21 (C and D) as determined by reverse transcription quantitative PCR (RT-qPCR). Comparison of particular isoforms’ expression was performed at four stages of seed development utilizing one-way ANOVA. Error bars indicate sd between biological replicates (n = 3). Different lowercase letters (a, b, c, and d) denote statistical significance (P < 0.05, Tukey’s posthoc test).

B. napus Inherited One 22:1-CoA-Specific DGAT2 Isoform Each from Brassica rapa and Brassica oleracea

In order to assess the evolutionary relations of the Bna.DGAT2 isoforms, we conducted phylogenetic analysis (Fig. 7). The results show that BnaA.DGAT2.d is most closely related to a DGAT2 isoform in B. rapa, while the other 22:1-CoA-specific BnaA.DGAT2.e is more closely related to B. oleracea. Those evolutionary connections suggest that each of the parental plants of B. napus possessed one DGAT2 gene with very high specificity for 22:1-CoA. The situation is similar for the Bna.DGAT2 isoforms presenting low activity with 22:1-CoA, where one of them is more closely related to B. rapa and the other to B. oleracea. It is noteworthy that all Bna.DGAT2 genes are found in chromosomes belonging to the AA genome, despite their relationship to both ancestral parents.

Figure 7.

Figure 7.

Open in a new tab

Phylogenetic tree showing Bna.DGAT2 isoforms’ predicted evolutionary relations with known DGAT2 sequences from B. oleracea and B. rapa. Evolutionary analysis shows that rapeseed inherited one highly 22:1-CoA-specific Bna.DGAT2 isoform (Bna.DGAT2.d/Bna.DGAT2.e) and one highly 18:3-CoA-specific Bna.DGAT2 isoform (Bna.DGAT2.b/Bna.DGAT2.c MONOLIT and Bna.DGAT2.c MAPLUS) from each of its ancestors.

DISCUSSION

Different Bna.DGAT Isoforms Contribute Differently to Acyl-CoA Specificity from MONOLIT and MAPLUS

Whereas the Bna.DGAT activities toward 18:3-CoA were similar in microsomal preparations of developing seeds from the low-erucic acid rape cultivar (MONOLIT) and the high-erucic acid cultivar (MAPLUS), the activity toward 22:1-CoA was sixfold to 14-fold higher in microsomal preparations from the high-erucic acid cultivar. In order to investigate if this difference could be due to different activities of Bna.DGAT isoenzymes with different substrate specificities in the two cultivars, we characterized acyl donor specificity of the four Bna.DGAT1 and the four Bna.DGAT2 enzymes found in the rapeseed genome. Bna.DGAT assays were performed in microsomal preparations of yeast strain H1246, lacking the capacity to synthesize TAG, by expressing the genes and assaying them with various long-chain acyl-CoAs as acyl donors and [14C]glycerol-labeled di-6:0-DAG as an artificial acyl acceptor.

All four Bna.DGAT1 enzymes showed similar and broad acyl donor specificity and also accepted 22:1-CoA, albeit at substantially lower rates compared with 16:0-CoA and 18:1-CoA, which were the best substrates. The substrate specificities of rapeseed DGAT1 isoforms expressed in yeast have previously been reported, although 22:1-CoA was not included in these assays (Aznar-Moreno et al., 2015; Greer et al., 2015, 2016). In the report by Aznar-Moreno et al. (2015), only 16:0-CoA and 18:1-CoA were tested, with 16:0-CoA giving the highest activity. However, the relative activity of 16:0-CoA to 18:1-CoA acylation varied considerably depending on the composition of exogenous long-chain DAG added in the assays. Greer et al. (2016) showed, similar to us, a broad acyl-CoA specificity of all Bna.DGAT1 isoforms. In their report, all isoforms showed much higher relative activities with 18:2-CoA and 18:3-CoA than in our study; similar results were shown with purified BnaC.DGAT1.a isoform (Caldo et al., 2015). It should also be noted that not all the Bna.DGAT1 variants cloned from cDNA from MONOLIT and MAPLUS were assayed due to the huge amount of work required to do this. Although all assayed Bna.DGAT1 variants had similar acyl-CoA preferences, we cannot exclude that the not tested Bna.DGAT1 variants contributed to the different specificities toward 22:1-CoA seen in the Bna.DGAT assays of developing seeds between the two cultivars.

In contrast to the broad substrate specificity of Bna.DGAT1 isoforms, the Bna.DGAT2 isoforms showed a narrow range of substrate specificities that fell into two groups, represented by two enzymes each. One group, BnaA.DGAT2.b and BnaA.DGAT2.c (both MONOLIT and MAPLUS), was very specific for 18:3-CoA, with much lesser activity with other acyl-CoAs and very little activity with 22:1-CoA. The other group, BnaA.DGAT2.d and BnaA.DGAT2.e, also showed high activity with 18:3-CoA but had broader acyl-CoA specificities and, most notably, had about the same activity with 22:1-CoA as with 18:3-CoA. The narrow substrate specificities of the rapeseed Bna.DGAT2s are in line with reports of a specialized role for DGAT2 in the accumulation of oil with unusual fatty acids, such as α-eleosteric acid, vernolic acid, and ricinoleic acid (Shockey et al., 2006; Burgal et al., 2008; Li et al., 2010). A DGAT2 isoform (LuDGAT2-3) expressed in linseed was shown to have, similar to BnaA.DGAT2.b and BnaA.DGAT2.c, 20-fold higher activity with 18:3-CoA than with 18:1-CoA and was postulated to contribute to the high 18:3 content in linseed TAG (Xu et al., 2018). It is tempting to speculate that the rapeseed Bna.DGAT2s play a similar role in channeling erucic acid (BnaA.DGAT2.d and BnaA.DGAT2.e) as well as linolenic acid (BnaA.DGAT2.b and BnaA.DGAT2.c) into the rapeseed TAG.

It is notable that the two rapeseed Bna.DGAT2 isoforms that had high activity toward 22:1-CoA had 97% amino acid sequence within the group and 78% to 80% similarity to the two Bna.DGAT2 enzymes that lacked specificity toward this substrate. Thus, only minor changes in the amino acid sequence in DGAT2s could have drastic effects on acyl specificity.

The Differences in Specificity toward 22:1-CoA Do Not Correlate with the Expression of Different Isoforms in MONOLIT and MAPLUS Developing Seeds

Based on the found specificities of the Bna.DGAT isoforms, it can be postulated that BnaA.DGAT2.d and/or BnaA.DGAT2e, having high activity with 22:1-CoA, are more active in MAPLUS than in MONOLIT during seed development. However, this is not reflected in the relative expression of the different isoforms as judged by RT-qPCR throughout seed development. The observed isoforms’ expression patterns do indicate differences in expression between the two studied cultivars, but expression patterns of BnaA.DGAT2d and BnaA.DGAT2.e do not correlate with the higher seed microsomal Bna.DGAT specificity toward 22:1-CoA in later stages of seed development (stages 3 and 4). At those stages, all Bna.DGAT2 isoforms were about equally expressed in MONOLIT as at the two previous development stages, whereas they were less expressed in MAPLUS. It should be noted, however, that only small amounts of the total TAG is deposited between stage 3 and stage 4. Transcriptome data from developing rape seeds showed high and increasing expression of Bna.DGAT1 during seed development, whereas Bna.DGAT2 expression was declining and was much lower than Bna.DGAT1 expression at a late stage of seed development (Troncoso Ponce et al., 2011). Our RT-qPCR data also show increases of BnaA.DGAT1.a and BnaC.DGAT1.a in the latter two stages of seed development. The differences between Bna.DGAT2 expression may originate in different cultivars being used in both studies. Although expression patterns of Bna.DGAT2 with high specificity for 22:1-CoA in MAPLUS do not correlate with the high Bna.DGAT activity seen for this acyl-CoA in seed microsomes, translational and posttranslational regulation, as well as different catalytic activities of the different isoforms, are likely to be more important for overall Bna.DGAT activities. It is noteworthy in this context that the specific activity of the different Bna.DGAT isoforms in microsomal preparations from yeast expressing the genes showed huge variation, despite high amino acid similarities. Significant differences between activities of rapeseed Bna.DGAT1 isoforms expressed in yeast were previously reported (Greer et al., 2015, 2016) and have been attributed to the different enzyme levels (Greer et al., 2015). However, BnaA.DGAT1.a, which was reported to have highest specific activity toward 16:0-CoA of all Bna.DGAT1 isoforms in the study by Greer et al. (2016), had in our assays only 15% of the activity of BnaA.DGAT1.b with this substrate. These differences might be due to different yeast transformation events and not to the enzyme per se. However, it is unclear if the huge variation in specific activities of the different Bna.DGAT isoforms as measured in microsomal fractions of yeast cells (up to 800-fold) correlate with different catalytic activities of the enzymes in vivo in plants. If this is the case, expression profiles of the different isoforms will give little or no information of the actual overall Bna.DGAT specificities in vivo.

Optimized DGAT Specificities Are Essential for Avoiding Bottlenecks in Oil Accumulation

The breeding of low-erucic acid rapeseed cultivars from high-erucic acid cultivars meant a drastic change in the fatty acid composition of the seed oil. Erucic acid constitutes up to 50% of all fatty acid in the oil in the high-erucic cultivars but is virtually absent in modern edible rapeseed cultivars. Erucic acid can be categorized as one of several hundreds of unusual fatty acids reported to occur in seed oil. Like erucic acid, many of these unusual fatty acids are, more or less, excluded from membrane lipids but can occur in a very high amounts in seed oil in certain plant species and families (Dyer et al., 2008). The biosynthesis of oils with a high amount of some of these unusual fatty acids, such as α-eleosteric acid, vernolic acid, and ricinoleic acid, has been studied in some detail. It has been shown that specialized DGAT2s play a key role in channeling the unusual acyl groups into TAG in those plants (Shockey et al., 2006; Burgal et al., 2008; Li et al., 2010). Coexpression of these DGAT2s in transgenic plants expressing enzymes responsible for the synthesis of the unusual fatty acids led to a significant increase in accumulation of them in the seed oil. It is therefore likely that the two rapeseed Bna.DGAT2 genes with high activity toward erucic acid will allow the plant to efficiently channel this unusual fatty acid into TAG.

The big differences in Bna.DGAT specificities toward erucic acid in developing seeds of the high-erucic acid cv MAPLUS compared with the low-erucic acid cv MONOLIT indicates that breeding has led to different shifts in the activities of the different Bna.DGAT isoforms in the two cultivars. Most likely, these shifts are due to the selection pressure for increased oil content within breeding programs. The importance of specialized DGAT2 in avoiding bottlenecks in oil deposition in plants accumulating high amounts of unusual fatty acids is clearly demonstrated in transgenic plants producing ricinoleic acid. Expression of the castor bean oleate hydroxylase in Arabidopsis led to the low accumulation of ricinoleic acid in the seed oil and a substantial decrease in the amount of oil (Burgal et al., 2008). The lower oil content could be attributed to a bottleneck in the utilization of DAG containing ricinoleic acid in TAG synthesis, which caused a feedback down-regulation of fatty acid biosynthesis (Bates et al., 2014). Coexpression of castor bean DGAT2 (having high specificity for ricinoleoyl groups) in the hydroxylase-expressing background both increased the amount of hydroxy fatty acid in TAG and restored the oil content to nearly wild-type levels (Burgal et al., 2008). If a plant that accumulates high amounts of unusual fatty acids, like erucic acid, loses the capacity to synthesize this fatty acid, it will likely also suffer from suboptimal specificities of their DGATs, which could affect oil quantity. It is tempting to speculate that the substantial drop experienced in oil content in early low-erucic acid rapeseed cultivars (Caballero et al., 2003) could partly be due to such suboptimal Bna.DGAT specificities for the novel fatty acid quality. It is likely that further breeding to restore oil content has caused changes in the specificities of the overall Bna.DGAT activity to better fit the novel fatty acid profile. This is reflected in the differences reported here in Bna.DGAT2 specificities toward 22:1-CoA between the high- and low-erucic acid cultivars.

CONCLUSION

The four rapeseed Bna.DGAT2s have been biochemically characterized. It was shown that two of these enzymes (in three variants) have high specificity for 18:3-CoA but low activity toward 22:1-CoA and that two Bna.DGAT2 enzymes have high specificity for this substrate, although 18:3-CoA is also a good substrate. Whereas Bna.DGAT2 isoforms showed high activity with 18:3-CoA, Bna.DGAT1 isoforms showed low activities with this substrate. Since microsomal preparations from developing rapeseed of both cultivars also showed high activity with 18:3-CoA, this suggests that Bna.DGAT2 isoforms play a much greater role than hitherto assumed in rapeseed oil accumulation. Furthermore, the marked differences in the Bna.DGAT specificities toward 22:1-CoA between the low- and high-erucic acid cultivars suggest that the activity of the Bna.DGAT2 isoforms with high activity with 22:1-CoA has been suppressed by breeding of low-erucic acid cultivars. Suboptimal DGAT specificities compared with fatty acid profiles produced in the seed can lead to decreased oil content. Therefore, characterizing DGAT enzymes on their specificities and manipulating their activities to fit the profile of the fatty acids produced might be a useful strategy for maintaining high oil in oilseeds bred for a changed oil composition. Since we have shown that small changes in amino acid sequences can have significant effects on the acyl specificity of DGATs, polymorphism in encoding genes might also prove to affect oil content and be a target for breeding oilseeds for higher oil content.

MATERIALS AND METHODS

Plant Material and Growth Conditions

Brassica napus seeds of both low-erucic acid (MONOLIT; accession no. 500864) and high-erucic acid (MAPLUS; accession no. 502809) cultivars were obtained from the Plant Breeding and Acclimatization Institute at the National Research Institute of Poland. Plants were first cultivated in a growth chamber at 23°C with a photoperiod of 16 h of light (120 µmol photons m−2 s−1)/8 h of dark and relative humidity of 60%. Both studied cultivars are winter rapeseeds; therefore, after developing five to six leaves, the young plants were vernalized for 6 weeks at 6°C before further cultivation at 23°C. Seeds of cv MONOLIT and cv MAPLUS were grouped into four equally intervalled developmental stages based on their appearance, weight, and days after flowering (Fig. 1; Supplemental Table S1).

Chemicals

Acyl-CoA and [14C]acyl-CoA substrates were synthesized from unlabeled (Larodan Fine Chemicals) or 14C-labeled (Perkin-Elmer and Moravek) fatty acids and CoA (Sigma-Aldrich) according to a modified method described by Sánchez et al. (1973). [14C]Glycerol-labeled tri-6:0-TAG was prepared by chemical acylation from the trifluoroacetic anhydride of [14C]glycerol (Perkin-Elmer) according to Kanda and Wells (1981). The tri-6:0-[14C]glycerol-TAG was partially lipase treated with Rhizomucor miehei lipase (Sigma-Aldrich). The lipase products were separated on TLC plates (silica gel 60; Merck), and the sn-1,2-rac-6:0-[14C]glycerol was eluted from the silica with methanol:chloroform (2:1, v/v) and extracted into chloroform (Bligh and Dyer, 1959).

Bna.DGAT Cloning

The Bna.DGAT1 isoforms (except for BnaA.DGAT1.b) were isolated using pooled cDNA derived from developing rapeseeds of both cultivars using gene-specific primers (Supplemental Table S5). The Bna.DGAT2 isoform amino acid sequences were first confirmed via amplification from cDNA isolated from each cultivar. The Bna.DGAT2 isoforms (and BnaA.DGAT1.b, whose expression could not be traced in cDNA from either cultivar) were codon optimized for expression in Saccharomyces cerevisiae using the Integrated DNA Technologies codon optimization tool. The isoforms were cloned by Gateway cloning into pYES-derived plasmid pDEST52 behind a GAL1 promoter via pDONR221 following the manufacturer’s instructions.

Plant Microsomal Preparations and Enzyme Assays

Microsomal preparations from MONOLIT and MAPLUS seeds where obtained from each developmental stage using a method previously described by Stymne and Stobart (1984). Microsomal protein content was determined using the Pierce BCA Protein Assay Kit (Thermo Fisher Scientific). Aliquots of microsomal isolates were subsequently used in in vitro assays designed to determine overall Bna.DGAT acyl-CoA specificity. Each enzyme assay included 22 µg of microsomal protein, 5 nmol of exogenous di-6:0-DAG as an acyl acceptor, and 5 nmol of [14C]acyl-CoA as acyl donors. Assays were performed in 50 mm, pH 7.2, HEPES buffer in a total volume of 100 µL with shaking (1,250 rpm) for 30 min. Radioactive TAG formed by acylation of di-6:0-DAG was determined as described above for yeast microsomal assays. DGAT activity with 18:3-CoA increased nearly linearly up to 30 min of incubation with microsomal preparations of MONOLIT seeds (Supplemental Fig. S4).

Yeast Microsomal Preparation Assays of Bna.DGAT Isoforms

The TAG-deficient S. cerevisiae strain H1246 (MATα are1-Δ::HIS3, are2-Δ::LEU2, dga1-Δ::KanMX4, lro1-Δ::TRP1 ADE2; Sandager et al., 2002) was transformed with the pDEST52-GAL1::Bna.DGAT constructs. Recombinant yeast cells were grown at 30°C in synthetic uracil dropout medium supplemented with 2% (w/v) Gal for 48 h. Harvested yeast cells were washed and homogenized, and microsomal membranes were recovered through ultracentrifugation at 100,000g as described by Lager et al. (2013). The total protein content of the microsomal membranes was determined as described for plant microsomes above.

Bna.DGAT activities of the different isoforms were measured in assays using 5 nmol of radiolabeled sn-1,2-rac-6:0-[14C]glycerol as acyl acceptor and 5 nmol of an acyl-CoA as acyl donor. The assays were carried out in a buffer consisting of 0.05 m HEPES, pH 7.2, 5 mm MgCl2, 1 mg mL−1 BSA, and 22 µg of yeast microsomal protein (final volume of 100 µL) for 5 to 30 min at 30°C while shaking at 1,250 rpm. The reaction time was optimized for each isoform to give nearly linear activity during the assay time with the best acyl-CoA donor (Supplemental Figs. S5 and S6).

After the incubations, all lipids were extracted into chloroform using the method first described by Bligh and Dyer (1959). An aliquot was taken for liquid scintillation counting to measure the amount of radioactivity. The remaining lipids were separated on TLC plates (silica gel 60; Merck) in hexane:diethyl ether:acetic acid (70:30:1), and the relative amounts of the different radioactive lipids were determined with electronic autoradiography (Instant Imager; Packard Instruments). The absolute amount of TAG was then calculated from the percentage of radioactive TAG and the total amount of radioactivity in the chloroform phase, as determined by liquid scintillation.

qPCR of Bna.DGAT Isoform Expression in Seeds

In order to extract RNA from seeds of each development stage of each cultivar, three seeds were pooled and flash frozen in liquid nitrogen. RNA was extracted using PureLink Plant RNA reagent (Invitrogen). To remove genomic DNA, the RNA was incubated with dsDNase (Thermo Fisher Scientific) and cDNA was synthesized using the Maxima First Strand cDNA Synthesis Kit for RT-qPCR (Thermo Fisher Scientific), both according to the manufacturer’s instructions.

After optimization and confirmation of linearity of the primer amplification (all primers used had a primer efficiency between 1.9 and 2), we conducted qPCR experiments using isoform-specific primers (Supplemental Table S6) in Quant Studio 3 (Thermo Fisher Scientific) according to the protocol found in Maxima SYBR Green/ROX qPCR Master Mix (Thermo Fisher Scientific). Later, all Bna.DGAT results were normalized to Bna.Act7 and Bna.UBC21. The Pfaffl method was used to calculate the relative expression of the Bna.DGAT genes (Pfaffl, 2001). Differences between a particular isoform expression level throughout seed development were subjected to one-way ANOVA with Tukey’s posthoc test (P < 0.05; Fisher, 1925; Tukey, 1949). All obtained amplicons were also sequenced to confirm the primer specificity, due to the high nucleotide sequence similarity between the studied isoforms.

Phylogenetic Analysis

A maximum likelihood phylogenetic tree of Bna.DGAT2 isoforms in relation to Brassica rapa and Brassica oleracea was constructed in MEGA X (Kumar et al., 2018) with 1,000 bootstraps (Felsenstein, 1985). The nucleotide sequences were received from the NCBI, and the coding sequence for each protein was translated to an amino acid sequence (Supplemental Data Set S1). The amino acid sequences were aligned using ClustalW, and the phylogenetic tree was constructed using the JTT matrix-based model (Jones et al., 1992).

Accession Numbers

Sequence data from this article can be found in the GenBank/EMBL data libraries under the following accession numbers: BnaA.DGAT1.a, gene identifier 106352950; BnaC.DGAT1.a, gene identifier 106402120; BnaA.DGAT1.b, gene identifier 106375527; BnaC.DGAT1.b, gene identifier 106436339; BnaA.DGAT2.b, NCBI reference sequence XM_013878945.2; BnaA.DGAT2.c, NCBI reference sequence XM_013788938.2; BnaA.DGAT2.d, NCBI reference sequence XM_013890141.2; and BnaA.DGAT2.e, NCBI reference sequence XM_013881688.2.

Supplemental Data

The following supplemental materials are available.

ACKNOWLEDGMENTS

We thank the Plant Breeding and Acclimatization Institute, National Research Institute of Poland, for providing us with B. napus seeds of both cultivars.

Footnotes

1

This work was supported by the National Science Centre, Poland (project no. 2014/13/N/NZ9/00873, PRELUDIUM 7) and by the Swedish Foundation for Strategic Research, Vinnova (Trees and Crops for the Future) and Formas (to S.J. and I.L.).

References

  1. Alexandratos N, Bruinsma J (2012) World Agriculture towards 2030/2050. http://www.fao.org/docrep/016/ap106e/ap106e.pdf
  2. Aymé L, Arragain S, Canonge M, Baud S, Touati N, Bimai O, Jagic F, Louis-Mondésir C, Briozzo P, Fontecave M, et al. (2018) Arabidopsis thaliana DGAT3 is a [2Fe-2S] protein involved in TAG biosynthesis. Sci Rep 8: 17254. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Aymé L, Baud S, Dubreucq B, Joffre F, Chardot T (2014) Function and localization of the Arabidopsis thaliana diacylglycerol acyltransferase DGAT2 expressed in yeast. PLoS ONE 9: e92237. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Aznar-Moreno J, Denolf P, Van Audenhove K, De Bodt S, Engelen S, Fahy D, Wallis JG, Browse J (2015) Type 1 diacylglycerol acyltransferases of Brassica napus preferentially incorporate oleic acid into triacylglycerol. J Exp Bot 66: 6497–6506 [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Banaś A, Dahlqvist A, Ståhl U, Lenman M, Stymne S (2000) The involvement of phospholipid:diacylglycerol acyltransferases in triacylglycerol production. Biochem Soc Trans 28: 703–705 [PubMed] [Google Scholar]
  6. Bates PD, Browse J (2012) The significance of different diacylgycerol synthesis pathways on plant oil composition and bioengineering. Front Plant Sci 3: 147. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Bates PD, Johnson SR, Cao X, Li J, Nam JW, Jaworski JG, Ohlrogge JB, Browse J (2014) Fatty acid synthesis is inhibited by inefficient utilization of unusual fatty acids for glycerolipid assembly. Proc Natl Acad Sci USA 111: 1204–1209 [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Bates PD, Stymne S, Ohlrogge J (2013) Biochemical pathways in seed oil synthesis. Curr Opin Plant Biol 16: 358–364 [DOI] [PubMed] [Google Scholar]
  9. Bligh EG, Dyer WJ (1959) A rapid method of total lipid extraction and purification. Can J Biochem Physiol 37: 911–917 [DOI] [PubMed] [Google Scholar]
  10. Burgal J, Shockey J, Lu C, Dyer J, Larson T, Graham I, Browse J (2008) Metabolic engineering of hydroxy fatty acid production in plants: RcDGAT2 drives dramatic increases in ricinoleate levels in seed oil. Plant Biotechnol J 6: 819–831 [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Caballero B, Finglas P, Toldra F (2003) Encyclopedia of Food Sciences and Nutrition, Ed 2 Academic Press, Cambridge, MA [Google Scholar]
  12. Caldo KMP, Greer MS, Chen G, Lemieux MJ, Weselake RJ (2015) Purification and properties of recombinant Brassica napus diacylglycerol acyltransferase 1. FEBS Lett 589: 773–778 [DOI] [PubMed] [Google Scholar]
  13. Dahlqvist A, Ståhl U, Lenman M, Banaś A, Lee M, Sandager L, Ronne H, Stymne S (2000) Phospholipid:diacylglycerol acyltransferase: An enzyme that catalyzes the acyl-CoA-independent formation of triacylglycerol in yeast and plants. Proc Natl Acad Sci USA 97: 6487–6492 [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Durrett TP, Benning C, Ohlrogge J (2008) Plant triacylglycerols as feedstocks for the production of biofuels. Plant J 54: 593–607 [DOI] [PubMed] [Google Scholar]
  15. Dyer JM, Stymne S, Green AG, Carlsson AS (2008) High-value oils from plants. Plant J 54: 640–655 [DOI] [PubMed] [Google Scholar]
  16. Ecke W, Uzunova M, Weißleder K (1995) Mapping the genome of rapeseed (Brassica napus L.). II. Localization of genes controlling erucic acid synthesis and seed oil content. Theor Appl Genet 91: 972–977 [DOI] [PubMed] [Google Scholar]
  17. Felsenstein J. (1985) Confidence limits on phylogenies: An approach using the bootstrap. Evolution 39: 783–791 [DOI] [PubMed] [Google Scholar]
  18. Fisher RA. (1925) Statistical Methods for Research Workers. Genesis Publishing, Edinburgh [Google Scholar]
  19. Greer MS, Pan X, Weselake RJ (2016) Two clades of type-1 Brassica napus diacylglycerol acyltransferase exhibit differences in acyl-CoA preference. Lipids 51: 781–786 [DOI] [PubMed] [Google Scholar]
  20. Greer MS, Truksa M, Deng W, Lung SC, Chen G, Weselake RJ (2015) Engineering increased triacylglycerol accumulation in Saccharomyces cerevisiae using a modified type 1 plant diacylglycerol acyltransferase. Appl Microbiol Biotechnol 99: 2243–2253 [DOI] [PubMed] [Google Scholar]
  21. Gurr MI, Harwood JL, Frayn KN (2002) Lipid Biochemistry: An Introduction, Ed 5 John Wiley & Sons, Hoboken, NJ [Google Scholar]
  22. Jones DT, Taylor WR, Thornton JM (1992) The rapid generation of mutation data matrices from protein sequences. Comput Appl Biosci 8: 275–282 [DOI] [PubMed] [Google Scholar]
  23. Kanda P, Wells MA (1981) Facile acylation of glycerophosphocholine catalyzed by trifluoroacetic anhydride. J Lipid Res 22: 877–879 [PubMed] [Google Scholar]
  24. Kennedy EP. (1961) Biosynthesis of complex lipids. Fed Proc 20: 934–940 [PubMed] [Google Scholar]
  25. Kumar S, Stecher G, Li M, Knyaz C, Tamura K (2018) MEGA X: Molecular Evolutionary Genetics Analysis across computing platforms. Mol Biol Evol 35: 1547–1549 [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Lager I, Yilmaz JL, Zhou XR, Jasieniecka K, Kazachkov M, Wang P, Zou J, Weselake R, Smith MA, Bayon S, et al. (2013) Plant acyl-CoA:lysophosphatidylcholine acyltransferases (LPCATs) have different specificities in their forward and reverse reactions. J Biol Chem 288: 36902–36914 [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Li R, Yu K, Hatanaka T, Hildebrand DF (2010) Vernonia DGATs increase accumulation of epoxy fatty acids in oil. Plant Biotechnol J 8: 184–195 [DOI] [PubMed] [Google Scholar]
  28. Li-Beisson Y, Shorrosh B, Beisson F, Andersson MX, Arondel V, Bates PD, Baud S, Bird D, DeBono A, Durrett TP, et al. (2013) Acyl-lipid metabolism The Arabidopsis Book 11: e0161, doi:10.1199/tab.0161 [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Liu Q, Siloto RM, Lehner R, Stone SJ, Weselake RJ (2012) Acyl-CoA:diacylglycerol acyltransferase: Molecular biology, biochemistry and biotechnology. Prog Lipid Res 51: 350–377 [DOI] [PubMed] [Google Scholar]
  30. Lock YY, Snyder CL, Zhu W, Siloto RMP, Weselake RJ, Shah S (2009) Antisense suppression of type 1 diacylglycerol acyltransferase adversely affects plant development in Brassica napus. Physiol Plant 137: 61–71 [DOI] [PubMed] [Google Scholar]
  31. Metzger JO, Bornscheuer U (2006) Lipids as renewable resources: Current state of chemical and biotechnological conversion and diversification. Appl Microbiol Biotechnol 71: 13–22 [DOI] [PubMed] [Google Scholar]
  32. Ohlrogge J, Thrower N, Mhaske V, Stymne S, Baxter M, Yang W, Liu J, Shaw K, Shorrosh B, Zhang M, et al. (2018) PlantFAdb: A resource for exploring hundreds of plant fatty acid structures synthesized by thousands of plants and their phylogenetic relationships. Plant J 96: 1299–1308 [DOI] [PubMed] [Google Scholar]
  33. Ohlrogge J, Browse J (1995) Lipid biosynthesis. Plant Cell 7: 957–970 [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Pfaffl MW. (2001) A new mathematical model for relative quantification in real-time RT-PCR. Nucleic Acids Res 29: e45. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Routaboul JM, Benning C, Bechtold N, Caboche M, Lepiniec L (1999) The TAG1 locus of Arabidopsis encodes for a diacylglycerol acyltransferase. Plant Physiol Biochem 37: 831–840 [DOI] [PubMed] [Google Scholar]
  36. Sánchez M, Nicholls DG, Brindley DN (1973) The relationship between palmitoyl-coenzyme A synthetase activity and esterification of sn-glycerol 3-phosphate in rat liver mitochondria. Biochem J 277: 697–706 [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Sandager L, Gustavsson MH, Ståhl U, Dahlqvist A, Wiberg E, Banaś A, Lenman M, Ronne H, Stymne S (2002) Storage lipid synthesis is non-essential in yeast. J Biol Chem 277: 6478–6482 [DOI] [PubMed] [Google Scholar]
  38. Shockey JM, Gidda SK, Chapital DC, Kuan J-C, Dhanoa PK, Bland JM, Rothstein SJ, Mullen RT, Dyer JM (2006) Tung tree DGAT1 and DGAT2 have nonredundant functions in triacylglycerol biosynthesis and are localized to different subdomains of the endoplasmic reticulum. Plant Cell 18: 2294–2313 [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Stefansson BR, Hougen FW (1964) Selection of rape plants (Brassica napus) with seed oil practically free from erucic acid. Can J Plant Sci 44: 359–364 [Google Scholar]
  40. Stymne S, Stobart AK (1984) The biosynthesis of triacylglycerols in microsomal preparations of developing cotyledons of sunflower (Helianthus annuus L.). Biochem J 220: 481–488 [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Stymne S, Stobart AK (1987) Triacylglycerol biosynthesis In Stumpf PK, and Conn EE, eds, Lipids: Structure and Function. Academic Press, Cambridge, MA, pp 175–214 [Google Scholar]
  42. Troncoso Ponce MA, Kilaru A, Cao X, Durrett TP, Fan J, Jensen JK, Thrower NA, Pauly M, Wilkerson C, Ohlrogge JB (2011) Comparative deep transcriptional profiling of four developing oilseeds. Plant J 68: 1014–1027 [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Tukey JW. (1949) Comparing individual means in the analysis of variance. Biometrics 5: 99–114 [PubMed] [Google Scholar]
  44. Weselake RJ, Taylor DC, Rahman MH, Shah S, Laroche A, McVetty PBE, Harwood JL (2009) Increasing the flow of carbon into seed oil. Biotechnol Adv 27: 866–878 [DOI] [PubMed] [Google Scholar]
  45. Xu Y, Holic R, Li D, Pan X, Mietkiewska E, Chen G, Ozga J, Weselake RJ (2018) Substrate preferences of long-chain acyl-CoA synthetase and diacylglycerol acyltransferase contribute to enrichment of flax seed oil with α-linolenic acid. Biochem J 475: 1473–1489 [DOI] [PubMed] [Google Scholar]

Articles from Plant Physiology are provided here courtesy of Oxford University Press

RESOURCES