Plastid Glycerol-3-phosphate Acyltransferase Enhanced Plant Growth and Prokaryotic Glycerolipid Synthesis in Brassica napus

Plastid-localized glycerol-3-phosphate acyltransferase (ATS1) catalyzes the first-step reaction in glycerolipid assembly through transferring an acyl moiety to glycerol-3-phosphate (G3P) to generate lysophosphatidic acid (LPA), an intermediate in lipid metabolism. The effect of ATS1 overexpression on glycerolipid metabolism and growth remained to be elucidated in plants, particularly oil crop plants. Here, we found that overexpression of BnATS1 from Brassica napus enhanced plant growth and prokaryotic glycerolipid biosynthesis. BnATS1 is localized in chloroplasts and an in vitro assay showed that BnATS1 had acylation activity toward glycerol 3-phosphate to produce LPA. Lipid profiling showed that overexpression of BnATS1 led to increases in multiple glycerolipids including phosphatidylglycerol (PG), monogalactosyldiacylglycerol (MGDG), phosphatidylcholine (PC), and phosphatidylinositol (PI), with increased polyunsaturated fatty acids. Moreover, increased MGDG was attributed to the elevation of 34:6- and 34:5-MGDG, which were derived from the prokaryotic pathway. These results suggest that BnATS1 promotes accumulation of polyunsaturated fatty acids in cellular membranes, thus enhances plant growth under low-temperature conditions in Brassica napus.


BnATS1 Is Expressed in Various Tissues and BnATS1 Is Localized in Chloroplasts
To investigate the temporal and spatial distribution of BnATS1 mRNA in Brassica napus, total RNA was extracted from various tissues and used for analysis by quantitative real-time PCR. BnATS1 (BnaA08g06960D) is expressed in various tissues including leaves, stems, roots, flowers, flower buds, siliques, developing seeds, developing silique walls, and mature seeds with expression being highest in young leaves and lowest in mature seeds ( Figure 1A). The results suggest that BnATS1 is highly expressed in green tissues, particularly in actively growing leaf tissues. To further explore the subcellular localization and activity of BnATS1, the full-length BnATS1 coding sequence (CDS) was cloned from cultivar (cv.) Westar (Brassica napus L.) by reverse transcription PCR using mRNA from leaves as a template, and the resultant CDS was fused with GFP at the C-terminus and then transiently expressed in the epidermal cells of tobacco leaves by Agrobacterium infiltration. Green fluorescent BnATS1-GFP was overlaid with the red fluorescence of chloroplasts (Figure 2), suggesting that BnATS1 is localized to chloroplasts and may be involved in the prokaryotic pathway. To test whether BnATS1 encodes a GPAT, BnATS1 was expressed in E. coli cells in an enzymatic activity assay. The resulting protein was able to catalyze the acylation of G3P using 16:0-CoA as a substrate to produce LPA ( Figure 1B). . Figure 2. BnATS1 was localized in chloroplasts. BnATS1-GFP was transiently expressed in tobacco leaf cells under the control of the 35S promoter. Green fluorescent signal of BnATS1-GFP was overlaid with red auto-fluorescence produced by chloroplasts observed using a confocal laser scanning microscope. Bars = 10 μm.

Overexpression of BnATS1 Enhanced Plant Growth during Vegetative Stage
BnATS1 is highly expressed in young leaves, suggesting a role in actively growing tissues. To investigate whether BnATS1 is involved in plant growth, BnATS1 was overexpressed in cv. Westar (Brassica napus L.) plants under the control of the CaMV 35S promoter. More than 20 independent transgenic lines were obtained and BnATS1-overexpression (OE) plants exhibited a similar phenotype with enhanced vegetative growth in the field growth conditions, in which the temperature was approximately 7-10 °C (night)/15-20 °C (day) in fall and 0-3 °C (night)/7-12 °C (day) in winter ( Figure 3A). We randomly selected three lines of them for functional characterization in detail. The BnATS1 transcript level in leaves of OE6, OE16, and OE22 plants was 9-to 33-fold higher than that of the wild type (WT) ( Figure 3B). The plant height of BnATS1-OE was approximately 30% greater relative to the WT at the vegetative stage ( Figure 3C). The leaf length and width in BnATS1-OE plants were increased by approximately 34% and 29%, respectively, as compared with the WT plants ( Figure 3D,E). The fresh weight of the OE6, OE16, and OE22 lines was increased by 31%, 50%, and 38%, respectively, as compared with the WT at the vegetative stage  . Figure 2. BnATS1 was localized in chloroplasts. BnATS1-GFP was transiently expressed in tobacco leaf cells under the control of the 35S promoter. Green fluorescent signal of BnATS1-GFP was overlaid with red auto-fluorescence produced by chloroplasts observed using a confocal laser scanning microscope. Bars = 10 μm.

Overexpression of BnATS1 Enhanced Plant Growth during Vegetative Stage
BnATS1 is highly expressed in young leaves, suggesting a role in actively growing tissues. To investigate whether BnATS1 is involved in plant growth, BnATS1 was overexpressed in cv. Westar (Brassica napus L.) plants under the control of the CaMV 35S promoter. More than 20 independent transgenic lines were obtained and BnATS1-overexpression (OE) plants exhibited a similar phenotype with enhanced vegetative growth in the field growth conditions, in which the temperature was approximately 7-10 °C (night)/15-20 °C (day) in fall and 0-3 °C (night)/7-12 °C (day) in winter ( Figure 3A). We randomly selected three lines of them for functional characterization in detail. The BnATS1 transcript level in leaves of OE6, OE16, and OE22 plants was 9-to 33-fold higher than that of the wild type (WT) ( Figure 3B). The plant height of BnATS1-OE was approximately 30% greater relative to the WT at the vegetative stage ( Figure 3C). The leaf length and width in BnATS1-OE plants were increased by approximately 34% and 29%, respectively, as compared with the WT plants ( Figure 3D,E). The fresh weight of the OE6, OE16, and OE22 lines was increased by 31%, 50%, and 38%, respectively, as compared with the WT at the vegetative stage

Overexpression of BnATS1 Enhanced Plant Growth during Vegetative Stage
BnATS1 is highly expressed in young leaves, suggesting a role in actively growing tissues. To investigate whether BnATS1 is involved in plant growth, BnATS1 was overexpressed in cv. Westar (Brassica napus L.) plants under the control of the CaMV 35S promoter. More than 20 independent transgenic lines were obtained and BnATS1-overexpression (OE) plants exhibited a similar phenotype with enhanced vegetative growth in the field growth conditions, in which the temperature was approximately 7-10 • C (night)/15-20 • C (day) in fall and 0-3 • C (night)/7-12 • C (day) in winter ( Figure 3A). We randomly selected three lines of them for functional characterization in detail. The BnATS1 transcript level in leaves of OE6, OE16, and OE22 plants was 9-to 33-fold higher than that of the wild type (WT) ( Figure 3B). The plant height of BnATS1-OE was approximately 30% greater relative to the WT at the vegetative stage ( Figure 3C). The leaf length and width in BnATS1-OE plants were increased by approximately 34% and 29%, respectively, as compared with the WT plants ( Figure 3D,E). The fresh weight of the OE6, OE16, and OE22 lines was increased by 31%, 50%, and 38%, respectively, as compared with the WT at the vegetative stage ( Figure 3F). However, the leaf number of BnATS1-OE plants was comparable to that of WT plants ( Figure 3G). Thus, enhanced fresh weight was primarily due to increased leaf size in BnATS1-OE plants.  Figure 3F). However, the leaf number of BnATS1-OE plants was comparable to that of WT plants ( Figure 3G). Thus, enhanced fresh weight was primarily due to increased leaf size in BnATS1-OE plants.

The Effect of BnATS1 on Phospholipid Metabolism
The results from the subcellular localization and activity assay suggest that BnATS1 catalyzes the first step of glycerolipid assembly to produce LPA in the prokaryotic pathway. To get insight into how BnATS1 affects plant growth, we profiled polar glycerolipid contents and lipid species from leaves of two representative BnATS1-OE lines, OE6 and OE16, compared with WT plants at the vegetative growth stage. Lipids from leaves were quantitatively analyzed by electron-spray ionization tandem mass spectrometry (ESI-MS/MS). PA content in leaves was very low at 0.129 nmol/mg DW, representing 0.19% of the total polar glycerolipids examined in WT leaves of cv. Westar (Brasscia napus), and total PA content in BnATS1-OE leaves was not significantly different from that of WT plants ( Figure 4). The molecular species including 34:3-, 34:2-, 36:6-, and 36:5-PA are major constituents of PA. The contents of most PA species in BnATS1-OE plants were not significantly different from that of the WT except higher 36:6-and 36:5-PA in OE6 and a lower 34:3-PA in OE16 relative to the WT ( Figure 5). The results implicate that PA is rapidly transferred to DAG and CDP-DAG for other glycerolipid synthesis.
PG is the only major phospholipid in thylakoid membranes, which is mainly produced in the prokaryotic pathway in plants [4]. Lipid profiling showed that PG content was 6.8% of the total polar glycerolipids examined in WT leaves of cv. Westar. The major molecular species of PG were 34:4-, 34:3-, and 34:2-PG. Overexpression of BnATS1 led to an increased PG content compared with the WT (Figure 4). Increased PG in BnATS1-OE leaves resulted from increases in 34:3-, 34:2-, and

The Effect of BnATS1 on Phospholipid Metabolism
The results from the subcellular localization and activity assay suggest that BnATS1 catalyzes the first step of glycerolipid assembly to produce LPA in the prokaryotic pathway. To get insight into how BnATS1 affects plant growth, we profiled polar glycerolipid contents and lipid species from leaves of two representative BnATS1-OE lines, OE6 and OE16, compared with WT plants at the vegetative growth stage. Lipids from leaves were quantitatively analyzed by electron-spray ionization tandem mass spectrometry (ESI-MS/MS). PA content in leaves was very low at 0.129 nmol/mg DW, representing 0.19% of the total polar glycerolipids examined in WT leaves of cv. Westar (Brasscia napus), and total PA content in BnATS1-OE leaves was not significantly different from that of WT plants ( Figure 4). The molecular species including 34:3-, 34:2-, 36:6-, and 36:5-PA are major constituents of PA. The contents of most PA species in BnATS1-OE plants were not significantly different from that of the WT except higher 36:6-and 36:5-PA in OE6 and a lower 34:3-PA in OE16 relative to the WT ( Figure 5). The results implicate that PA is rapidly transferred to DAG and CDP-DAG for other glycerolipid synthesis.
PG is the only major phospholipid in thylakoid membranes, which is mainly produced in the prokaryotic pathway in plants [4]. Lipid profiling showed that PG content was 6.8% of the total polar glycerolipids examined in WT leaves of cv. Westar. The major molecular species of PG were 34:4-, 34:3-, and 34:2-PG. Overexpression of BnATS1 led to an increased PG content compared with the WT (Figure 4). Increased PG in BnATS1-OE leaves resulted from increases in 34:3-, 34:2-, and 34:1-PG, which were partially coincided with those major species of PA ( Figure 5). The results suggest that increased PG may be derived from PA mediated by BnATS1 overexpression.
PC is a major phospholipid in cellular membranes and is originally derived from PA via the CDP-choline pathway, and PC can be also produced via the PE methylation pathway [1]. PC content was approximately 15%, whereas other phospholipids such as PE, PI, and phosphatidylserine (PS) were only 8.2%, 4.5%, and 0.4%, respectively, of the total polar glycerolipids examined in WT leaves of cv. Westar (Figure 4). PC in BnATS1-OE plants was increased by 34-43%, compared with the WT (Figure 4). LysoPC (LPC) content in BnATS1-OE was also higher than that of the WT. Molecular profiling revealed that PC was predominantly composed of 34: 3 Figure 5). In addition, PI content in BnATS1-OE plants was also higher than that of the WT, which was caused by elevations in 34:3-and 34:2-PI (Figures 4 and 5). By comparison, overexpression of BnATS1 did not lead to changes in PE and PS levels as compared to the WT (Figure 4). The results suggest that increased BnATS1 expression also promotes PC and PI production with increased polyunsaturated fatty acid species. 34:1-PG, which were partially coincided with those major species of PA ( Figure 5). The results suggest that increased PG may be derived from PA mediated by BnATS1 overexpression. PC is a major phospholipid in cellular membranes and is originally derived from PA via the CDP-choline pathway, and PC can be also produced via the PE methylation pathway [1]. PC content was approximately 15%, whereas other phospholipids such as PE, PI, and phosphatidylserine (PS) were only 8.2%, 4.5%, and 0.4%, respectively, of the total polar glycerolipids examined in WT leaves of cv. Westar (Figure 4). PC in BnATS1-OE plants was increased by 34-43%, compared with the WT (Figure 4). LysoPC (LPC) content in BnATS1-OE was also higher than that of the WT.  Figure 5). In addition, PI content in BnATS1-OE plants was also higher than that of the WT, which was caused by elevations in 34:3-and 34:2-PI (Figures 4 and 5). By comparison, overexpression of BnATS1 did not lead to changes in PE and PS levels as compared to the WT ( Figure 4). The results suggest that increased BnATS1 expression also promotes PC and PI production with increased polyunsaturated fatty acid species.

Overexpression of BnATS1 Enhanced Prokaryotic Galactolipids
Galactolipids, such as MGDG and DGDG, are mostly abundant in green tissues, particularly in leaves of plants. MGDG and DGDG were 49.6% and 15.1%, respectively, of the total polar glycerolipids examined in WT leaves of cv. Westar (Figure 4). Lipid profiling showed that molecular species of MGDG were distinguishable from those in DGDG. MGDG in leaves were predominantly composed of 34:6-MGDG (18:3/16:3-MGDG), comprising 75% of the total MGDG species, whereas DGDG was mainly composed of 36:6-DGDG (18:3/18:3-DGDG), comprising 63% of the total DGDG species in WT leaves ( Figure 5). The results suggest that MGDG is mainly assembled through the prokaryotic pathway in Brassica napus. Moreover, overexpression of BnATS1 led to a significant increase in MGDG. The MGDG contents in OE6 and OE16 leaves were increased by 45% and 20%, respectively, as compared with WT plants. The increased MGDG in BnATS1-OE plants was caused predominantly by the elevations in 34:6-34:5-, and 36:6-MGDG ( Figure 5). Although overexpression of BnATS1 did not affect the total DGDG content, 34:3-DGDG in the OE6 line was higher than that of the WT (Figure 5). These results suggest that enhanced thylakoid lipids in BnATS1-OE plants are due to enhanced galactolipid synthesis via the prokaryotic pathway mediated by BnATS1.

Overexpression of BnATS1 Enhanced Prokaryotic Galactolipids
Galactolipids, such as MGDG and DGDG, are mostly abundant in green tissues, particularly in leaves of plants. MGDG and DGDG were 49.6% and 15.1%, respectively, of the total polar glycerolipids examined in WT leaves of cv. Westar (Figure 4). Lipid profiling showed that molecular species of MGDG were distinguishable from those in DGDG. MGDG in leaves were predominantly composed of 34:6-MGDG (18:3/16:3-MGDG), comprising 75% of the total MGDG species, whereas DGDG was mainly composed of 36:6-DGDG (18:3/18:3-DGDG), comprising 63% of the total DGDG species in WT leaves ( Figure 5). The results suggest that MGDG is mainly assembled through the prokaryotic pathway in Brassica napus. Moreover, overexpression of BnATS1 led to a significant increase in MGDG. The MGDG contents in OE6 and OE16 leaves were increased by 45% and 20%, respectively, as compared with WT plants. The increased MGDG in BnATS1-OE plants was caused predominantly by the elevations in 34:6-34:5-, and 36:6-MGDG ( Figure 5). Although overexpression of BnATS1 did not affect the total DGDG content, 34:3-DGDG in the OE6 line was higher than that of the WT (Figure 5). These results suggest that enhanced thylakoid lipids in BnATS1-OE plants are due to enhanced galactolipid synthesis via the prokaryotic pathway mediated by BnATS1.

Overexpression of BnATS1 Up-Regulated the Expression of Genes Related to Lipid Anabolism
Lipid profiling showed that overexpression of BnATS1 promoted simultaneously the accumulation of multiple lipids, such as PG, MGDG, PC, and PI. It is possible that overexpression of BnATS1 may affect the expression of genes involved in different lipid metabolic processes. To test this possibility, RNA was extracted from the leaves of BnATS1-OE and WT plants and analyzed by real-time PCR. The β-ketoacyl-ACP reductase (KAR) is required for de novo FA synthesis [35], and the KAR expression level in BnATS1-OE leaves was two-fold higher than that of the WT (Figure 6). ATS2 catalyzes the sn-2 acylation of the glycerol backbone to produce PA in the prokaryotic pathway following ATS1 action [33], and the ATS2 transcript level in BnATS1-OE leaves was increased two-fold compared with the WT (Figure 6). Moreover, overexpression of BnATS1 led to up-regulation of genes, such as PEAMT (phosphoethanolamine N-methyltransferase) and AAPT1 (aminoalcoholphosphotransferase 1), involved in PC synthesis from DAG, and LPCAT (lysophosphatidylcholine acyltransferase) in PC-acyl editing. In addition, overexpression of BnATS1 also promoted the expression of PGPS (phosphatidylglycerolphosphate synthase) involved in PG synthesis. The PGPS expression was significantly up-regulated in OE6 and slightly increased in OE16 as compared with the WT (Figure 6). Taken together, the results suggest that overexpression of BnATS1 is able to promote synchronously the expression of genes involved in multiple lipid anabolic processes, thus coordinates with other enzymes to enhance glycerolipid synthesis in Brassica napus.
Lipid profiling showed that overexpression of BnATS1 promoted simultaneously the accumulation of multiple lipids, such as PG, MGDG, PC, and PI. It is possible that overexpression of BnATS1 may affect the expression of genes involved in different lipid metabolic processes. To test this possibility, RNA was extracted from the leaves of BnATS1-OE and WT plants and analyzed by real-time PCR. The β-ketoacyl-ACP reductase (KAR) is required for de novo FA synthesis [35], and the KAR expression level in BnATS1-OE leaves was two-fold higher than that of the WT (Figure 6). ATS2 catalyzes the sn-2 acylation of the glycerol backbone to produce PA in the prokaryotic pathway following ATS1 action [33], and the ATS2 transcript level in BnATS1-OE leaves was increased two-fold compared with the WT (Figure 6). Moreover, overexpression of BnATS1 led to up-regulation of genes, such as PEAMT (phosphoethanolamine N-methyltransferase) and AAPT1 (aminoalcoholphosphotransferase 1), involved in PC synthesis from DAG, and LPCAT (lysophosphatidylcholine acyltransferase) in PC-acyl editing. In addition, overexpression of BnATS1 also promoted the expression of PGPS (phosphatidylglycerolphosphate synthase) involved in PG synthesis. The PGPS expression was significantly up-regulated in OE6 and slightly increased in OE16 as compared with the WT (Figure 6). Taken together, the results suggest that overexpression of BnATS1 is able to promote synchronously the expression of genes involved in multiple lipid anabolic processes, thus coordinates with other enzymes to enhance glycerolipid synthesis in Brassica napus. Figure 6. Overexpression of BnATS1 enhanced expression of genes related to the lipid anabolic process. Total RNA was extracted from leaves of 40-day-old plants under field growth conditions. The relative expression levels of genes were analyzed by real-time PCR normalized to the β-Actin expression level. Values are means ± SD (n = 3 separate samples). * and ** denote significance at p < 0.05 and p < 0.01, respectively, compared with WT plants based on Student's t test. Values are means ± SD (n = 3 separate samples). * and ** denote significance at p < 0.05 and p < 0.01, respectively, compared with WT plants based on Student's t test.

Discussion
In plants, glycerolipids are assembled via two parallel pathways, the plastid-localized prokaryotic pathway and the ER-localized eukaryotic pathway. ATS1 catalyzes a committed step reaction in the prokaryotic pathway. The Arabidopsis mutant ats1 (also designated atc1) with defective ATS1 exhibits a reduction in prokaryotic thylakoid lipids [3,4]. However, the effect of overexpressing ATS1 on plant growth and lipid metabolism remained to be elucidated. Here, we found that plastid-localized BnATS1 in Brassica napus has a positive impact on plant growth, accompanied by increases in multiple glycerolipids including PG, MGDG, PC, and PI.
Our results from Brassica napus plants overexpressing BnATS1 showed that BnATS1 has a promotion on vegetative growth, particularly leaves. Indeed, BnATS1 was highly expressed in green tissues, being the highest in young leaves, suggesting its role in leaf growth. Our observation showed that BnATS1 is localized in chloroplasts, implicating its involvement in lipid assembly in the prokaryotic pathway. Fatty acids are synthesized de novo in chloroplasts [2], and this plastid-localized BnATS1 may facilitate the access of acyl groups for glycerolipid assembly. It was showed that ATS1 catalyzes the sn-1 acylation of G3P using 18:1-ACP and 16:0-ACP as substrates in Helianthus annuus [34]. Our result from the in vitro assay showed that BnATS1 is capable of using 16:0-CoA as a substrate to produce LPA, which is a precursor for PA synthesis through the acylation at the sn-2 of glycerol backbone. PA is an intermediate in glycerolipid metabolism and is rapidly transferred to other lipids [1,2,36]. PA produced in the prokaryotic pathway is responsible for synthesis of PG and galactolipids [15,16], whereas PA produced in the eukaryotic pathway is primarily involved in phospholipid synthesis, and could be transported to chloroplasts for galactolipid synthesis [1,[17][18][19]. Our results from the lipid profiling showed that most PA species in BnATS1-OE plants were not significantly higher than in the WT. Instead, overexpression of BnATS1 led to increased PG in leaves. The increased molecular species of PG were mostly similar to those species abundant in PA, implicating that increased PG in BnATS1-OE plants may be derived from PA. PG is an essential component of photosynthetic membranes and most PG (approximately 85% of total PG) in plastids is derived from the prokaryotic pathway in Arabidopsis [37]. Increased PG in BnATS1-OE plants agrees with a previous report showing that the ats1 (act1) mutant had marked reduction in PG [3].
Overexpression of BnATS1 also led to increased PC, which can be attributed to all molecular species of PC examined in leaves, as compared with the WT. PC is a major phospholipid in cellular membranes and an entry point for newly synthesized fatty acids from chloroplasts to the ER for glycerolipid assembly [40]. Our lipid profiling revealed that PC is mostly composed of polyunsaturated fatty acids such as 36:6-, 36:5-, 36:4, 36:3-, 34:3-, and 34:2-PC, and overexpression of BnATS1 increased PC with polyunsaturated acyl groups. The ratio of polyunsaturated to saturated fatty acids of PC in BnATS1-OE plants (4.7-5.5:1) was substantially higher than that of WT plants (3.9:1). It was reported that plant species with a higher content of unsaturated fatty acids exhibit more tolerance to cold stress than those species containing a higher level of saturated fatty acids [41]. Brassica napus plants frequently experience cold stress under the growth conditions in the field during winter. Increased PC with polyunsaturated acyl groups in BnATS1-OE plants may contribute to enhanced tolerance to cold stress, thus promoting plant growth under low-temperature conditions. However, increased PC in BnATS1-OE plants may not be resulting directly from increased BnATS1 expression, as PA produced in the prokaryotic pathway may be not involved in PC synthesis [2,3].
In agreement with these alterations, overexpression of BnATS1 promotes the expression of genes such as KAR, ATS2, PEAMT, AAPT1, and LPCAT involved in de novo fatty acid synthesis, PC assembly, and acyl editing. In PC synthesis from the CDP-choline pathway, phosphoethanlamine is methylated to phosphocholine by PEAMT, and then is activated to CDP-choline [1,42]. The phosphocholine group of CDP-choline is further transferred to DAG by AAPT, producing PC [1,43]. LPCAT is involved in acyl editing through re-esterification of lyso-PC [10]. Our data showed that PC was largely increased in BnATS1-OE, which agrees with the increased expression of multiple genes in the lipid anabolic pathway. These results suggest that chloroplast-localized BnATS1 provides a "pull" force to enhance fatty acid synthesis and a "push" effect on glycerolipid assembly.

Plant Materials and Growth Conditions
The seeds of canola cultivar (cv.) Westar (Brassica napus L.) were germinated in pots containing soil. Two-week-old seedlings with comparable size were transferred to pots (one plant per pot) and grown with regular watering in field conditions from late autumn through the spring season in Wuhan, China. The temperature was approximately 7-10 • C (night)/15-20 • C (day) in late autumn and 0-3 • C (night)/7-12 • C (day) in winter in Wuhan, China.

Gene Cloning, Vector Construction and Plant Transformation
A cDNA pool was synthesized from mRNA extracted from leaves of cv. Westar (Brassica napus) plants by reverse transcription using a TransScript cDNA Synthesis SuperMix Kit according to the manufacturer's instructions (TansGene Biotech, Beijing, China). The full-length BnATS1 cDNA was amplified from cDNA pools by PCR using the primers BnATS1-OE-F 5 -TCTAGAATGACTCTCACGTTTTCCTC-3 (forward) and BnATS1-OE-R 5 -GAGCTCCTAATTCCAAGGTTGTGACA-3 (reverse), and then ligated into the expression vector pBI121 after digestion with XbaI and SacI under the control of the 35S promoter. The construct containing BnATS1 was confirmed by sequencing and then introduced into Agrobacterium tumefaciens strain GV3101, which was used to infect cv. Westar hypocotyls in tissue culture to get regeneration plants based on the methods described previously [44]. The transgenic plants were verified by PCR using the pBI121 vector sequence specific primer 5 -GATGGTTAGAGAGGCTTACGCA-3 and BnATS1 specific primer BnATS1-OE-R (Supplementary Table S1).

RNA Extraction and Real-Time PCR
Total RNA was isolated from various tissues of cv. Westar plants using Transzol reagent (TransGen Biotech, Beijing, China). RNA extracts were treated with DNaseI to remove contaminating DNA and used as a template to synthesize the first-strand cDNA by reverse transcription using a TransScript cDNA Synthesis SuperMix Kit according to the manufacturer's instructions (TransGen Biotech, Beijing, China). β-Actin from Brassica napus was used as an internal standard. Real-time PCR was performed as described previously [45]. The primers used are listed in Supplementary Table S1.

BnATS1 Protein Expression and Activity Assay
The full-length BnATS1 coding sequence (CDS) omitting the stop codon was amplified by PCR using primers BnATS1-PF and BnATS1-PR (Supplementary Table S1) and then was cloned into the pET28a vector at the cutting sites of EcoRI and XhoI. After sequence confirmation, the construct was introduced into the E. coli Rosetta (DE3) strain (TransGen Biotech, Beijing, China) and cultured in LB liquid medium until the OD 600 reached 0.6 for protein expression by the induction with 0.5 mM isopropyl β-D-thiogalactopyranoside (IPTG) for 20 h at 16 • C. Cells were lysed by ultrasonification in cold lysis buffer (50 mM Tris-HCl, pH 7.5, 120 mM NaCl, 1 mM DTT, 0.1% Triton-100, 10% glycerol), and followed by centrifugation at 10,000 g at 4 • C for 20 min. BnATS1 activity was assayed in 100 µL of reaction mixtures containing 250 µM HEPES buffer (pH 7.4), 8 µM 16:0-CoA, 600 µM G3P-Ca, 5 mg/mL BSA, and 20 µg crude proteins incubated at 25 • C for 5 min. The reaction was stopped by addition of 300 µL of chloroform: methanol (1:2, v/v). Resultant lipids were separated on thin layer chromatography (TLC) plates using a developing solvent (chloroform: methanol: acetic acid: H 2 O = 85: 15: 10: 3.5, v/v) and were visualized with iodine vapor [34]. The spots corresponding to LPA were quantified by the Image J software (1.48, National Institutes of Health, USA).

Subcellular Localization
The full-length BnATS1 CDS was amplified by PCR using primers BnATS1-GFP-F 5 -GGTACCA TGACTCTCACGTTTTCCTC-3 (forward) and BnATS1-GFP-R 5 -TCTAGAATTCCAAGGTTGTGACA AAG-3 (reverse) and cloned into the pCAMBIA1301s vector after digestion with KpnI and XbaI, which results in in-frame C-terminal fusion to GFP. The construct was introduced into Agrobacterium tumefaciens GV3101. The transformant was grown overnight in liquid LB media, and then centrifuged at 4000 rpm for 10 min. The cells were resuspended with the solution containing 10 mM MgCl 2 and 1 mM acetosyringone, and used for infiltrating leaves of 4-week-old tobacco (Nicotiana benthamiana) plants for transient expression driven by the 35S promoter. After infection for 2 to 3 days, BnATS1-GFP was visualized using a confocal laser scanning microscope (Leica, Biberach, Germany).

Lipid Extraction and Analysis
Total lipids were extracted according to the method described previously [46]. Briefly, leaf discs (0.5 g per sample) were sampled from 40-day-old plants (cv. Westar) and immediately immersed in 75 • C isopropanol (5 mL) containing 0.01% butylated hydroxytoluene (BHT) for 15 min to arrest enzymatic activity. After cooling to room temperature (25-28 • C), the sovlent (3.5 mL) of chloroform/water (2.5:1, v/v) was added to the sample and incubated for 1 h at room temperature by shaking. The extracts were transferred to clean glass tubes, and then the remaining leaves were re-extracted with 5 mL chloroform/methanol (2:1, v/v) containing 0.01% BHT at room temperature for several times until leaf tissues became bleached. The extracts were combined and washed twice with 1 M KCl and once with water, and dried under a stream of nitrogen gas, and then dissolved in a defined volume of chloroform. Lipids were quantitatively profiled by electron-spray ionization tandem mass spectrometry (ESI-MS/MS) based on the approach described previously [46].

Conflicts of Interest:
The authors declare no conflict of interest.