Identification and characterization of a plastidial ω-3 fatty acid desaturase EgFAD8 from oil palm (Elaeis guineensis Jacq.) and its promoter response to light and low temperature

In higher plants, ω-3 fatty acid desaturases are the key enzymes in the biosynthesis of alpha-linolenic acid (18:3), which plays key roles in plant metabolism as a structural component of both storage and membrane lipids. Here, the first ω-3 fatty acid desaturase gene was identified and characterized from oil palm. The bioinformatic analysis indicated it encodes a temperature-sensitive chloroplast ω-3 fatty acid desaturase, designated as EgFAD8. The expression analysis revealed that EgFAD8 is highly expressed in the oil palm leaves, when compared with the expression in the mesocarp. The heterologous expression of EgFAD8 in yeast resulted in the production of a novel fatty acid 18:3 (about 0.27%), when fed with 18:2 in the induction culture. Furthermore, to detect whether EgFAD8 could be induced by the environment stress, we detected the expression efficiency of the EgFAD8 promoter in transgenic Arabidopsis treated with low temperature and darkness, respectively. The results indicated that the promoter of EgFAD8 gene could be significantly induced by low temperature and slightly induced by darkness. These results reveal the function of EgFAD8 and the feature of its promoter from oil palm fruits, which will be useful for understanding the fuction and regulation of plastidial ω-3 fatty acid desaturases in higher plants.


Introduction
Linolenic acid (18:3) plays important roles in plant metabolism as a structural component of storage and membrane lipids, and as a precursor of signaling molecules involved in plant development and stress response [1,2]. For human, alpha-linoenic acid (18:3 ω-3) is one of essential fatty acids, which are necessary for health and must be acquired from diet [3]. PLOS  exist in wild-type yeast cells [38]. Recently, the in vivo function of a DGAT2 gene from oil palm mesocarp has been verified in yeast and transgenic Arabidopsis, exhibiting a substrate preference towards unsaturated fatty acids [39]. In the present work, a chloroplast ω-3 fatty acid desaturase (EgFAD8) and its promoter from oil palm has been isolated and characterized. The expression pattern of EgFAD8 in leaves and mesocarp at five developmental stages has been detected by Real-time quantitative PCR. The in vivo function of EgFAD8 has been verified by overexpression in Saccharomyces cerevisiae (INVSc1) with exogenous linoleic acid (18:2). To detect whether it could be induced by low temperature or the change of environment, we detected the efficiency of the EgFAD8 promoter in transgenic Arabidopsis treated with low temperature and darkness, respectively. These results revealed the function of EgFAD8 and the feature of its promoter from high plants, which will be useful for understanding the role of ω-3 fatty acid desaturases from fruits grown in the tropical areas.

Materials
Ethics statement. No specific permissions were required for these locations/activities. We confirmed that the field studies did not involve endangered or protected species.

Bioinformatic analysis
Nucleotide and amino acid sequence analyses used the BLAST program and open reading frame (ORF) finder from NCBI website (https://www.ncbi.nlm.nih.gov/). Amino acid alignment of fatty acid desaturases was performed by ClustalX 2.1 program with default setting [40]. A phylogenetic tree was constructed using the neighbor-joining method in MEGA5 [41]. Promoter prediction and cis-acting regulatory element analysis were carried out on the Plant-CARE (http://bioinformatics.psb.ugent.be/webtools/plantcare/html/) [42].

Cloning of EgFAD8 gene
Total RNA from mesocarp was extracted by cetyltrimethylammonium bromide (CTAB) based method as described previously [43]. First-strand cDNA was synthesized from 1 μg of total RNA using FastQuant RT Kit (Tiangen, Beijing, China) according to manufacturer's instructions, and was used for gene cloning and expression analysis. The coding sequence of EgFAD8 was amplified by Phusion High-Fidelity DNA Polymerase (Thermo Fisher Scientific, USA) using primers EgFAD8-pYES2 F and R with restriction site KpnI and XbaI, respectively (Table 1). PCR was performed with an initial denaturation at 98˚C for 30s, 30 cycles at of 98 C for 10 s, 56˚C for 30s, 72˚C for 60 s and a final extension step of 72˚C for 5 min. The PCR product was cloned into the pEASY-Blunt vector (Transgen biotech, Beijing, China) and sequenced. The cloning of EgFAD8 promoter region sequence has been published in our previous study [44].

Gene expression in Saccharomyces. cerevisiae
In order to generate a yeast expression construct, EgFAD8 fragments were digested with KpnI and XbaI from pEASY vector, and subcloned into pYES2 vector. The construct EgFAD8-pYES2 and the control pYES2 were separately transformed into INVSc1 cells using the polyethylene glycol/lithium acetate method [45,46]. Transformants were selected on synthetic complete medium lacking uracil (SC-U) supplemented with 2% (w/v) glucose. Three independent positive clones of each transformation were used for further expression studies.
Precultures were made in 2 mL SC-U medium with 2% (w/v) raffinose and 1% (w/v) tergitol NP-40 (Sigma, USA), and were grown overnight at 30˚C. For induction gene expression, appropriate amount of precultures were inoculated into 20 mL SC-U medium containing 2% (w/v) galactose and 50 μM linoleic acid at an OD 600 of 0.1. Yeast cultures were incubated for three days at 30˚C under continuous agitation (250 rpm). Then the cells were harvested and washed twice with sterile deionized water before used for fatty acid analysis.

Fatty acid analysis
Yeast cells were directly transmethylated by 2 mL of methanol containing 2.7% (v/v) H 2 SO 4 at 80˚C for 2 hours. Then, 3 mL of Hexane was added for recovering fatty acid methyl esters (FAMEs), and 2 mL of NaCl (2.5%, w/v) was used for phase separation. After centrifuge, the upper hexane phase was transferred to a fresh tube and applied for gas chromatography (GC) analysis.

Exogenous application of low temperature and light treatments
Ten-day-old seedlings of the T3 generation of transgenic Arabidopsis were subjected to stress treatments. For low temperature and light treatments, transgenic plants were planted at 15˚C and in dark, respectively, with other conditions are the same with the control (as mentioned in Materials). Quantitative analysis of GUS activity in each plant was carried out after treatments for 24h and 48h.

Primers Sequence
https://doi.org/10.1371/journal.pone.0196693.t001 Characterization of a plastidial ω3-fatty acid desaturase EgFAD8 and its promoter from oil palm Quantitative analysis of GUS activity in transgenic Arabidopsis was determined according to Jefferson [47]. The protein concentrations of plant extracts were measured as descried by Bradford [48]. GUS activity was calculated as pmol 4-Methylumbelliferone (4-MU) per min per microgram protein. Three replicates were performed for each sample.

Real-time quantitative PCR analysis
Total RNA from mature leaves and mesocarp at five developmental stages were extracted as described previously, and cDNAs was used for real-time quantitative PCR (RT-qPCR). EgFAD8 RT-PCR primers RT-EgFAD8 F/R (Table 1) were designed for RT-qPCR. The housekeeper gene β-actin (RT-β-actin F/R, Table 1) was used as an internal control for expression analysis. RT-qPCR was carried out using a CFX Connect™ Real-Time PCR Detection System (Bio-Rad Laboratories) and SYBR 1 premix Ex Taq™ II (Tli RNaseH Plus) Kit (Takara, Tokyo, Japan) according to the manufacturer's instructions. Expression was quantified as comparative threshold cycle (Ct) using 2 -ΔΔCt method [49]. Reactions were in triplicate including templatefree and no-reverse-transcriptase negative controls.

Cloning and bioinformatic analysis of EgFAD8
The open reading frame (ORF) cDNA sequence (1362 bp) of a ω-3 fatty acid desaturase (designated as EgFAD8) was cloned from oil palm (Elaeis guineensis Jacq.) mesocarp based on that gene sequence from Elaeis oleifera (GenBank accession: EU057620.1) using RT-PCR. The nucleotide sequence is 100% identity with that from Elaeis oleifera. It encodes a 453-amino acid polypeptide with the conserved domain of ω-3 fatty acid desaturase (amino acids 1-443), which is predicted to be chloroplastic. The protein BLAST analysis of EgFAD8 showed that it shared high homology (more than 70%) with other members of the ω-3 desaturase enzyme family. Phylogenetic analysis of EgFAD8 with various fatty acid desaturases from model plant Arabidopsis thaliana revealed that EgFAD8 clusters with two chloroplastic ω-3 fatty acid desaturases AtFAD7 and AtFAD8, as well as the endoplasmic reticulum-type ω-3 fatty acid desaturase AtFAD3 (Fig 1).

Expression pattern of EgFAD8 in leaf and mesocarp during fruit development
The transcript levels of EgFAD8 in leaf and mesocarp at five different developmental stages were determined by RT-qPCR using β-actin as an internal control. The results showed that the transcript level of EgFAD8 in mesocarp was quite flat during ripening with a slight increase at the beginning (from P1 to P2, Fig 2). Interestingly, the expression level of EgFAD8 in mature leaves was at least twice that in mescoarp (Fig 2).

Heterologous expression of EgFAD8 in yeast could produce 18:3 using exogenous 18:2
The activity of EgFAD8 was determined by heterologous expression in S. cerevisiae supplemented with linoleic acid as substrate, which is absent in yeast. Yeast transformed with empty vector (pYES2) was used as negative control. The total fatty acid composition of yeast transformed with EgFAD8 or pYES2 was detected and analyzed. The result showed that yeast transformed with EgFAD8 produced a new fatty acid species 18:3 ω-3 (about 0.27 mol%) in yeast, but not presented in the negative control (Fig 3). Meanwhile, there was no obvious difference in the levels of other fatty acids between yeast transformed with EgFAD8 and the negative control pYES2 (Fig 3). Therefore, the expression of EgFAD8 in yeast could produce 18:3 using the supplemented 18:2.

The activity of EgFAD8 promoter response to growth environment treatments
Promoter prediction analysis suggested that the EgFAD8 promoter contained some stressresponsive elements, including light-responsive elements. Therefore, further work was carried  Characterization of a plastidial ω3-fatty acid desaturase EgFAD8 and its promoter from oil palm out to investigate whether it was induced under various stress conditions, such as cold and dark. The ProEgFAD8::GUS transgenic Arabidopsis T3 lines were cultivated at 15˚C and in the dark, respectively, with other conditions are the same with the control. The GUS activity of transgenic plants under the darkness or low temperature treatment was no obvious difference with that of the control after 24 hours treatment (Fig 4). However, 48 hours treatment later, the GUS activity of the transgenic plants under the darkness treatment showed obvious increase by 26% when compared with that of the control, and the transgenic plants under the low temperature treatment exhibited the extraordinarily significant increase of GUS activity by about 14-fold comparing with the control (Fig 4).

Discussion
In this study, we identified and characterized the first chloroplast ω-3 fatty acid desaturase (EgFAD8) from oil palm. It exhibited the capability of synthesizing 18:3 ω-3 when heterologously expressed in yeast with the exogenous supply of linoleic acid. However, there are three ω-3 fatty acid desaturases in Arabidopsis: the ER-localized FAD3 and two chloroplastic FAD7 and FAD8. When compared with fatty acid desaturases from Arabidopsis, EgFAD8 showed the highest similarity with the temperature-sensitive chloroplastic ω-3 fatty acid desaturase and AtFAD8 (data not shown). Moreover, the phylogenetic analysis of EgFAD8 revealed that it is grouped with AtFAD7 and AtFAD8 from Arabidopsis, but with more close relation with AtFAD8 (Fig 1). Therefore, the bioinformatic analysis of EgFAD8 indicated that the EgFAD8 gene might encode a plastidial FAD8, which could be induced by low temperature [16]. Nevertheless, more experimental evidence should be shown to support the results from sequence analysis although FAD8 sequences are highly conserved in plants [17].
Furthermore, the expression pattern in oil palm mesocarp at five developmental stages and leaves was also detected to see whether the expression level of EgFAD8 is correlated with oil deposition in oil palm mesocarp. While, we can see EgFAD8 was predominately expressed in leaves, at least twice-fold as that of mesocarp. Moreover, the expression level of EgFAD8 remains still during fruit ripening with a slight increase at the beginning. The majority of oil Characterization of a plastidial ω3-fatty acid desaturase EgFAD8 and its promoter from oil palm deposition in oil palm mesocarp occurs at the late stages (from phase 3 to phase 5) [32,35]. Thus, the expression of EgFAD8 is not correlated with oil accumulation in mesocarp. Similar patterns have been seen in the expression of FAD8 in other plant species [11,13,17]. The GmFAD8 genes from soybean were constitutively expressed in all vegetative tissues (roots, stems, mature leaves and flowers) analyzed at optimal growth temperatures [11,13]. Like EgFAD8, the plastidial PfrFAD7 genes from Perilla frutescens var. frutescens were expressed at low levels during all developmental stages of seeds, when compared with expression levels in leaves [17]. Therefore, this result is also another proof to support EgFAD8 encodes a plastidial ω-3 fatty acid desaturase, rather than the microsomal ω-3 fatty acid desaturase.
In order to study the in vivo function of EgFAD8, it has been successfully expressed in S. cerevisiae. However, there is no linoleic acid (18:2) produced in yeast. So the exogenous supply of 18:2 is needed for studying the activity of ω-3 fatty acid desaturase. The result from GC analysis indicated that the EgFAD8 transformed yeast cells could produce slight amounts of 18:3 using the exogenous 18:2. At least, this result confirmed the in vivo activity of ω-3 fatty acid desaturase encoded by EgFAD8 from oil palm. While, the amount of 18:3 produced by the EgFAD8 transformed yeast is very limited. There might be three possible reasons as follows: first, the EgFAD8 gene is likely to encode a chloroplast ω-3 fatty acid desaturase, rather than the ER-localized desaturase; thus it could not participate in the synthesis of storage oil in oil palm mesocarp. Likewise, when expressed in yeast cells, EgFAD8 might just be involved in the desaturation of galactolipids in the plastid. Second, EgFAD8 has been predicted to be temperature-sensitive desaturase by bioinformatic analysis. The yeast growth condition at 30˚C might not intrigue the large expression of transformed gene EgFAD8. Third, the exogenous supply of 18:2 might be a bit unhealthy or toxic for yeast cell. So, the coexpression of ω-3 fatty acid desaturase with ω-6 fatty acid desaturase in yeast might be a better solution than feeding with 18:2 in the yeast culture. However, the JcFAD7 from Jatropha curcas also showed the similar result when expressed in yeast, producing about 2.5% of 18:3 in transformed cells [24]. Therefore, the EgFAD8 gene from oil palm encodes a ω-3 fatty acid desaturase.
Furthermore, in order to elucidate the physiological role of EgFAD8, we also detected whether the EgFAD8 promoter could be affected by the environmental stress including lowtemperature and darkness treatments. In our previous study, the activity of EgFAD8 promoter in transgenic Arabidopsis has been characterized (S1 Fig) [44]. The GUS staining results indicated that the EgFAD8 promoter has the expected capability to drive transgene expression in all tissues tested, except with very weak expression in roots (S1 Fig). Thus, we used the ProEg-FAD8::GUS transgenic plants for the stress treatment analysis, since the EgFAD8 promoter is constitutively active in transgenic Arabidopsis. The results revealed that the GUS activity of the transgenic plants under the darkness and low temperature treatment was increased by 26% and 14-folds after 48-hour treatment when compared with that of the control, respectively (Fig  4). So the EgFAD8 gene from oil palm could be significantly induced by low temperature and slightly induced by darkness. This is consistent with the induction expression of AtFAD8 from Arabidopsis when treated at 4˚C [16,50]. Similarly, the PoleFAD8 transcript level of Portulaca oleracea L. also increased significantly after treatment at 5˚C [51]. As expected, this result also evidenced that the chloroplastic EgFAD8 from oil palm is cold-inducible. On the other hand, lots of light regulatory elements were predicted in the promoter of EgFAD8 by PlantCARE [44]. The result of the darkness treatment on transgenic plants is also consistent with the bioinformatic prediction. The darkness could slightly induce the expression of EgFAD8. While, the mechanism that the light regulates the expression of FAD8 still remains obscure. Our hypothesis is that more chloroplasts are needed for fatty acid synthesis in the dark than in the light, since the rate of fatty acid synthesis in leaves is six-fold higher in the light than in the dark [2]. Thus, to maintain the constitutive function, more chloroplasts are synthesized in the leave cells, and the expression level of genes encoding the corresponding enzymes involved in the synthesis of chloroplast might be up-regulated. FAD8 is responsible for the polyunsaturated components of chloroplast membranes [16]. However, the related mechanism needs to be elucidated.
In conclusion, we have isolated and identified a plastidial ω-3 fatty acid desaturase (EgFAD8) from oil palm. The in vivo function of EgFAD8 has been confirmed by heterologously expression in yeast, with showing the conversion of linoleic to linolenic acid. Moreover, the expression profile of EgFAD8 also revealed that it mainly involved in the synthesis of chloroplast membrane, unlike ER-localized ω-3 fatty acids that is tightly correlated with the oil accumulation in seeds or fruits [1,20,25]. Furthermore, the characterization of the EgFAD8 promoter indicated that the EgFAD8 gene could be induced by low temperature and darkness. Overall, these results will be helpful for understanding the function and regulation of plastidial ω-3 fatty acid desaturases in higher plants.