Conserved Function of ACYL–ACYL CARRIER PROTEIN DESATURASE 5 on Seed Oil and Oleic Acid Biosynthesis between Arabidopsis thaliana and Brassica napus

Previous studies have shown that several ACYL–ACYL CARRIER PROTEIN DESATURASE (AtAAD) members in Arabidopsis thaliana are responsible for oleic acid (C18:1) biosynthesis. Limited research has been conducted on another member, AtAAD5, and its paralog BnAAD5 in the closely related and commercially important plant, Brassica napus. Here, we found that AtAAD5 was predominantly and exclusively expressed in developing embryos at the whole seed developmental stages. The aad5 mutation caused a significant decrease in the amounts of oil and C18:1, and a considerable increase in the content of stearic acid (C18:0) in mature seeds, suggesting that AtAAD5 functioned as an important facilitator of seed oil biosynthesis. We also cloned the full-length coding sequence of BnAAD5-1 from the A3 subgenome of the B. napus inbred line L111. We showed that ectopic expression of BnAAD5-1 in the A. thaliana aad5-2 mutant fully complemented the phenotypes of the mutant, such as lower oil content and altered contents of C18:0 and C18:1. These results help us to better understand the functions of AAD members in A. thaliana and B. napus and provide a promising target for genetic manipulation of B. napus.


INTRODUCTION
Seed fatty acids (FAs) and FA-derived complex lipids not only provide nutrients for humans and livestock (Li et al., 2006;Graham, 2008), but also serve as raw materials for industries and biofuel production (Durrett et al., 2008;Lu et al., 2011). Biosynthesis of seed oil is under the control of multiple genes, and occurs in plant cells in three steps (Baud et al., 2008;Itabe, 2010;Chapman and Ohlrogge, 2012). The first step is the production of pyruvate and other substances during glycolysis. Catabolysis of pyruvate and other substances leads to the FA precursor acetyl-CoA, which results in biosynthesis of C16-18 FAs in plastids. In the second step, FA derivatives are formed at acyl chains. FA formation occurs in the cytoplasm after chain elongation and desaturation of most C16-18 FAs from the first step. Finally, triacylglycerols are formed to store the new oil in oil bodies.
In the current study, we found that AtAAD5 is specifically expressed in the embryo during seed development in A. thaliana. We demonstrated that AtAAD5 promotes oil and C18:1 biosynthesis in A. thaliana seeds. We also cloned and functionally characterized B. napus AAD5-1 (BnAAD5-1), showing that it exhibits a conserved role with AtAAD5 in regulating seed FA accumulation when expressed in A. thaliana.

Plant Materials and Growth Conditions
The Col-0 ecotype was used as the wild type A. thaliana control, and the mutants were aad5-1 (SALK_129779C), and aad5-2 (SALK_035968C) in the Col-0 background. All A. thaliana plants were grown at 22 • C with a 16 h/8 h light/dark photoperiod, which has been reported in detail previously . The B. napus inbred line L111 was maintained in the greenhouse of South Campus, Northwest A&F University, China. T-DNA mutant were genotyped using specific primers (Supplementary Table S1).
Gene Cloning of BnAAD5-1 from B. napus Primers were designed to amplify the BnAAD5-1 gene based on the full-length coding domain sequence of BnAAD5-1 (GenBank Number XP_013735719.1). Developing seeds were used as a source of total RNA to synthesize template complementary DNA (cDNA). Seeds were collected from the B. napus inbred line L111 15 days after pollination. We used the pMD18-T vector for cloning (TaKaRa Bio, Dalian, China), and eight single colonies were picked randomly and sequenced by Sangon Biotechnology (Shanghai, China). Cloning primers are listed in Supplementary  Table S1.

Plasmid Construction
To obtain the construct of 35S:BnAAD5-1, the amplified fulllength coding regions of BnAAD5-1 were digested with Xma I and Spe I and then were cloned into pGreen-35S; this was driven by the 35S promoter. To construct pAtAAD5:GUS, the 824-bp AtAAD5 genomic region including a 283-bp promoter region, ATG, and a 538-bp region downstream of the ATG start codon in sequence was amplified and then cloned into pHY107 (Liu et al., 2007). Plasmid construction primers are listed in Supplementary  Table S1.

Generation of A. thaliana Transgenic Plants
The pAtAAD5:GUS and 35S:BnAAD5-1 constructs were transformed into Agrobacterium tumefaciens GV3101 and were used to transform A. thaliana wild type and aad5-2 plants, respectively, and the floral dip method was utilized (Clough and Bent, 1998). We used Basta R selection and genotyping to confirm that plants were transgenic until T3 homozygous lines were obtained.

Morphological Observation of Mature Seeds
Mature A. thaliana seeds were randomly selected from major inflorescences, specifically from siliques in the basal region, and photographed using an OLYMPUS SZ 61 stereomicroscope.

Seed FA Measurement
Mature A. thaliana seeds for FA determination were collected from siliques in the basal region of the major inflorescences of 16 individual plants sown in different pots arranged in a randomized block design. Seed FA determination was conducted as previously described (Poirier et al., 1999;Chen et al., 2012). In brief, seeds were infused into the methanol solution containing 1 M HCl at 80 • C for 2 h, which would convert FAs into the corresponding methyl esters. Then, FA methyl esters were extracted with the hexane, and were subsequently quantified by a gas chromatograph (GC-2014; Shimadzu).

Gene Expression Analysis
Total RNA samples were isolated from A. thaliana young siliques or B. napus developing seeds with the MiniBEST Plant RNA Extraction Kit (TaKaRa) and their corresponding cDNA samples were biosynthesized with PrimerScript RT (TaKaRa). Reverse transcription-PCR (RT-PCR) and quantitative RT-PCR (qRT-PCR) were conducted for three biological replicates. SYBR Green Master Mix (TaKaRa) was utilized for qRT-PCR analysis. The A. thaliana house-keeping gene AtEF1aA4 was regarded as an internal control. Primers used for the RT-PCR and qRT-PCR analyses are listed in Supplementary Table S1.

Analysis of AtAAD5 Expression Pattern
Previous RT-PCR results showed that AtAAD5 was widely expressed in A. thaliana tissues, including leaves, stems, roots, flowers, and siliques (Kachroo et al., 2007). To better investigate the spatiotemporal expression pattern of AtAAD5, we obtained 19 independent lines of pAtAAD5:GUS from a wild type background. GUS staining patterns were similar among most of the lines; therefore, one representative line was used for GUS staining analysis. The result showed that AtAAD5 was expressed in several tissues, including hypocotyl vascular bundles ( Figure 1A), root tips ( Figures 1A,B), cotyledons ( Figure 1B), and young expanding true leaves ( Figure 1B). Notably, AtAAD5 was highly present in developing embryos at different stages (Figures 1G-L). However, no GUS staining was observed in other tissues, such as expanded true leaves ( Figure 1C (Figures 1F-L). These results suggested that AtAAD5 controls seed traits mainly occurring in the A. thaliana embryo at the whole seed developmental stages (Baud et al., 2002;Fait et al., 2006;Graham, 2008;Baud and Lepiniec, 2009).

AtAAD5 Promotes Oil and Oleic Acid Biosynthesis in Seeds
AtAAD5 was previously screened by Kachroo et al. (2007) for T-DNA insertion mutants, but they did not obtain homozygous lines. In this study, we successfully obtained two T-DNA insertion mutants SALK_129779C and SALK_035968C from the Col-0 ecotype in the 5 untranslated region and the exon of AtAAD5, respectively, from the Arabidopsis Biological Resources Center (ABRC), which were designated aad5-1 and aad5-2, respectively (Figure 2A). The genotyping PCR result indicated the presence of the two homozygous mutants (Figure 2B). The RT-PCR result showed that the N-and C-terminal AtAAD5 transcripts were not detected in aad5-1 and aad5-2 mutants, respectively ( Figure 2C). More PCR product was amplified by C-terminal primers compared to N-terminal primers for the Col-0 RNA samples (Figure 2C), suggesting that the PCR amplification efficiency of the C-terminal primers is higher than that of the N-terminal primers. Notably, the C-terminal AtAAD5 transcript in aad5-1 was almost as strong as the wild type, which needs further investigation (Figure 2C).
To explore the biological function of seed FA accumulation, we used mature seeds from wild type and aad5 plants to determine the contents of major FAs. The result showed that the seed oil content was much lower in aad5 mutants than in the wild type seeds (Figures 3A,B). In aad5 seeds, there was a significant increase in the amount of C18:0 and a significant decrease in the C18:1 content (Figure 3C), suggesting that AAD5 plays a role in the desaturation of C18:0-ACP. However, we did not observe clear differences among morphological traits of seeds, including color of the seed coat, the size of the seed, or the dry weight of the seed (Supplementary Figure S1) between mature seeds of wild type and aad5 plants. These results suggested that AtAAD5 promotes seed oil and oleic acid biosynthesis in the A. thaliana embryo.
We performed a phylogenetic analysis to investigate the evolutionary relationship between BnAAD5-1 and 33 AAD5 proteins from 11 oil-producing plant species. The analysis indicated that BnAAD5-1 is most related to the three AAD5 sequences, including BrAAD5 (XP_009134697.1) from B. rapa, BoAAD5 (XP_013630756.1) from B. oleracea, and AtAAD5 from A. thaliana (Figure 4B).

BnAAD5-1 Fully Rescues the FA Phenotype of A. thaliana aad5-2 Seeds
To further elucidate the function of BnAAD5-1 in seed FA biosynthesis, we over-expressed it in the A. thaliana aad5-2 mutant, using the construct 35S:BnAAD5-1 (Figure 5A). (B) Comparison of seed total FA content (µg/seed) between the wild type (Col-0) and aad5 plants. (C) Comparison of contents of major seed FA compositions between the wild type (Col-0) and aad5 plants. Asterisks indicate significant differences in the seed total FA content (A,B) and the contents of major seed FA compositions (C) compared to that in the wild type (two-tailed paired Student's t-test, P ≤ 0.05). DW, dry weight. Values are means ± SD (n = 5). Error bars indicate standard deviation.
A total of 23 independent T1 transgenic plants were obtained following Basta R selection, and five independent transgenic lines (aad5-2 35S:BnAAD5-1 T3) were confirmed by PCR amplification of the BnAAD5-1 gene with the specific primers 35S_Pro/BnAAD5-1_R1 ( Figure 5A; Supplementary Table S1). Expression of the BnAAD5-1 gene in these transgenic plants was measured by qRT-PCR, and was determined to be highest in the transgenic line aad5-2 35S:BnAAD5-1#10, whereas its expression was not detected in the wild type or aad5-2 plants ( Figure 5B). We observed that ectopic expression of BnAAD5-1 fully rescued aad5-2 seed phenotypes, such as lower oil content ( Figure 5C) and altered contents of C18:0 and C18:1 (Figure 5D). Although the aad5-2 35S:BnAAD5-1#8 transgenic line showed the lowest expression of BnAAD5-1, the contents of total FAs, C18:0, and C18:1 were close to those of other transgenic plants (Figures 5B-D). This indicated that BnAAD5-1 regulates seed FA accumulation in a dose-independent manner when overexpressed in A. thaliana. These results together suggested that BnAAD5-1 has a similar function to AtAAD5.

DISCUSSION
The increase and optimization of FA composition in oilproducing plant seeds is the most important objective for breeders. Several of the seven AtAAD genes, including AtSSI2, AtAAD1, AtAAD2, AtAAD3, and AtAAD4, have been functionally identified for seed FA biosynthesis in A. thaliana (Kachroo et al., 2007;Bryant et al., 2016). However, little is known about the role of AtAAD5, and its paralog BnAAD5 in seed FA accumulation. Our results provide two major lines of evidence for a conserved and important role for AAD5 in mediating total FAs accumulation in seeds and C18:1 accumulation in the embryo in both A. thaliana and B. napus.
First, the aad5 mutation resulted in a considerable increase in the amounts of oil and C18:0, and a significant decrease in the C18:1 content in mature seeds (Figure 3). The expression of AtAAD5 was stably observed in developing embryos, but not in the endosperm and seed coat, during the whole seed developmental stages (Figure 1). C18:1 FA mainly exists in the forms of C18:1 9 and C18:1 11 in the A. thaliana embryo and endosperm plus seed coat, respectively, and C18:1 11 FA only accounts for less than 1 mol% of total FAs in the embryo (Bryant et al., 2016). The previous study showed that AtAAD5 preferentially desaturates C18:0-ACP substrate at the C9 position (Kachroo et al., 2007). These results together suggested that AtAAD5 plays an important role in controlling the conversion of C:18-ACP to C18:1 9 in the A. thaliana embryo. Traits of seeds including color of the coat, size of the seed, and weight of the seed were not altered in aad5 mutants, which is consistent with the fact that AtAAD5 was not expressed in seed coat and endosperm (Figure 1F-L). Intricate regulatory networks control FA accumulation in seeds. These networks also require coordinated development of three distinct seed tissues: embryo, endosperm, and seed coat. Therefore, the disruption of the structural gene AtAAD5 might disturb seed embryo development, causing lower seed oil accumulation (Figures 3A,B). AtAAD5 and AtAAD1 are most closely related based on phylogenetic analyses of the AtAAD family; they are 82% identical at the amino acid level (Kachroo et al., 2007). Error bars indicate standard deviation. (C) Quantitative comparison of total FA content between the wild type (Col-0), aad5-2, and aad5-2 35S:BnAAD5-1 seeds. Asterisks indicate statistically significant differences in total FA content of aad5-2 seeds compared to that of wild type seeds (two-tailed paired Student's t-test, P ≤ 0.05). Values are means ± SD (n = 5). Error bars indicate standard deviation. (D) Quantitative comparison of FA compositions of C18:0 and C18:1 between the wild type (Col-0), aad5-2, and aad5-2 35S:BnAAD5-1 seeds. Asterisks indicate statistically significant differences in contents of FA compositions of aad5-2 seeds compared to that of wild type seeds (two-tailed paired Student's t-test, P ≤ 0.05). Values are means ± SD (n = 5). Error bars indicate standard deviation.
Consistently, AtAAD1 and AtAAD5 showed similar functions on the conversion of C18:0-ACP to C18:1 9 in the A. thaliana embryo ( Figure 3C; Kachroo et al., 2007). It is worth mentioning that AtAAD1 negatively affects C18:2 biosynthesis (Kachroo et al., 2007), whereas AtAAD5 has no significant effect on the accumulation of C18:2 and other major seed FAs except for C18:0 and C18:1 ( Figure 3C). These results indicated that the two genes have some differences in the regulation of seed FA biosynthesis in the A. thaliana embryo.
Second, ectopic expression of BnAAD5-1 cloned from the A3 subgenome of the B. napus inbred line L111 in the aad5-2 mutant fully rescued altered seed FA contents of the mutant (Figure 5). This strongly suggested that BnAAD5-1 exhibits a conserved role with AtAAD5 in regulating seed FA accumulation when expressed in A. thaliana. However, no obvious differences were observed in the contents of oil, C18:0, and C18:1 in seeds between aad5-2 BnAAD5-1 overexpressors and the wild type control (Figure 5). This indicates that the alterations caused by reduced expression of AtAAD5 and increased expression of BnAAD5-1 in A. thaliana do not simply mirror each other. Arabidopsis thaliana and B. napus are both part of Cruciferae, and there are three A. thaliana loci in the B. rapa, B. oleracea, and B. nigra genomes (Kowalski et al., 1994;Osborn et al., 1997;Lagercrantz, 1998;Haberer et al., 2006). Brassica rapa and B. oleracea hybridize to create B. napus (Parkin et al., 1995;Osborn et al., 1997). During B. napus evolution, there was a high frequency of rearrangement, fusion, and deletion of chromosomes (Lagercrantz, 1998), which led to, on average, 2-8 paralogs in the B. napus genome for each gene locus in A. thaliana (Osborn et al., 1997;Cavell et al., 1998). Here we found a single copy of AtAAD5 in the A. thaliana genome as expected, and seven putative BnAAD5 paralogs in the B. napus genome ( Figure 4A). Our previous study showed that BnTOP1α-1 from the inbred L111 line has lost 4 amino acid stretches, compared with BnTOP1α-1 (XP_013685667.1) from ZS11, which collectively correspond to 130 amino acids . However, the cloned BnAAD5-1 from the inbred L111 line has the same sequence as BnAAD5-1 (XP_013735719.1) from ZS11 at the protein level (Figure 4A), and is most related to BrAAD5 (XP_009134697.1), BoAAD5 (XP_013630756.1), and AtAAD5 ( Figure 4B). Saturated FA quantity in B. napus has been increased by seed-specific antisense repression of one BrAAD gene from B. rapa (Knutzon et al., 1992). These results indicate that AAD5 might be conserved during evolution of the cruciferous species (A. thaliana, B. rapa, B. oleracea, and B. napus).
In summary, this study is the first to identify that an AtAAD member, AtAAD5, is responsible for converting C18:0-ACP to C18:1 and promoting oil accumulation in the A. thaliana embryo.
In addition, we showed that BnAAD5-1 has a conserved function with AtAAD5 in regulating seed FA accumulation when it is expressed in A. thaliana. Brassica napus is grown as a crop primarily for its seed oil. The identification and manipulation of key B. napus genes controlling seed oil and FA accumulation are of fundamental importance for agricultural production. These results suggest that BnAAD5-1 can be used as a promising target to genetically manipulate B. napus and other oil-producing plants to improve the amounts of seed oil, C18:0, and C18:1.

AUTHOR CONTRIBUTIONS
CJ and DL carried out the experiments. CJ and CG analyzed the data. KL, SQ, SD, ZL, JG, and JW assisted with doing the experiments. MC conceived and designed the experiments. MC and CJ wrote the manuscript. DL, CG, and JH helped to draft the manuscript and revise the manuscript. All authors read and approved the final manuscript.

ACKNOWLEDGMENT
The T-DNA insertion mutants used in this study, SALK_129779C and SALK_035968C, were distributed by ABRC.