Mol Genet Genomics DOI 10.1007/s00438-015-1011-0
ORIGINAL PAPER
Effects of tung oilseed FAD2 and DGAT2 genes on unsaturated fatty acid accumulation in Rhodotorula glutinis and Arabidopsis thaliana Yicun Chen · Qinqin Cui · Yongjie Xu · Susu Yang · Ming Gao · Yangdong Wang
Received: 17 July 2014 / Accepted: 12 February 2015 © Springer-Verlag Berlin Heidelberg 2015
Abstract Genetic engineering to produce valuable lipids containing unsaturated fatty acids (UFAs) holds great promise for food and industrial applications. Efforts to genetically modify plants to produce desirable UFAs with single enzymes, however, have had modest success. The key enzymes fatty acid desaturase (FAD) and diacylglycerol acyltransferase (DGAT) are responsible for UFA biosynthesis (a push process) and assembling fatty acids into lipids (a pull process) in plants, respectively. To examine their roles in UFA accumulation, VfFAD2 and VfDGAT2 genes cloned from Vernicia fordii (tung tree) oilseeds were conjugated and transformed into Rhodotorula glutinis and Arabidopsis thaliana via Agrobacterium tumefaciens. Realtime quantitative PCR revealed variable gene expression levels in the transformants, with a much higher level of VfDGAT2 than VfFAD2. The relationship between VfFAD2 expression and linoleic acid (C18:2) increases in R. glutinis
(R2 = 0.98) and A. thaliana (R2 = 0.857) transformants was statistically linear. The VfDGAT2 expression level was statistically correlated with increased total fatty acid content in R. glutinis (R2 = 0.962) and A. thaliana (R2 = 0.8157) transformants. With a similar expression level between single- and two-gene transformants, VfFAD2-VfDGAT2 cotransformants showed a higher linolenic acid (C18:3) yield in R. glutinis (174.36 % increase) and A. thaliana (14.61 % increase), and eicosatrienoic acid (C20:3) was enriched (17.10 % increase) in A. thaliana. Our data suggest that VfFAD2-VfDGAT2 had a synergistic effect on UFA metabolism in R. glutinis, and to a lesser extent, A. thaliana. These results show promise for further genetic engineering of plant lipids to produce desirable UFAs.
Communicated by S. Hohmann.
Introduction
Y. Chen and Q. Cui contributed equally to the paper.
Interest in the production of reproducible biomass has increased as a result of heightened environmental awareness, particularly in relation to the impact of fossil energy. Oil-rich plants are increasingly attractive as a source of popular products such as lipids, natural chemicals, and fuels (Singh 2010). Vernicia fordii Hemsl. (Euphorbiaceae), also known as the tung tree or tung oil tree, is a promising industrial oil crop in China. Tung oil, the major product of tung tree oilseeds, is widely used in paints, highquality printing, medicine, plasticizers, and chemical reagents, and possesses properties required in a raw material for biodiesel (Park et al. 2008; Shang et al. 2010). Approximately 80,000–100,000 tons of tung tree oil are produced per year in China (Zhan et al. 2012), representing 70–80 %
Y. Chen · Q. Cui · S. Yang · M. Gao · Y. Wang (*) State Key Laboratory of Tree Genetics and Breeding, Chinese Academy of Forestry, Beijing 100091, People’s Republic of China e-mail:
[email protected] Y. Chen e-mail:
[email protected] Y. Chen · Q. Cui · S. Yang · M. Gao · Y. Wang Institute of Subtropical Forestry, Chinese Academy of Forestry, Fuyang, Hangzhou 311400, Zhejiang, People’s Republic of China Y. Xu Hubei Forestry Academy, Wuhan 430075, Hubei, People’s Republic of China
Keywords FAD2 · DGAT2 · Unsaturated fatty acid · Transgenic
13
of the world market. However, the output of tung oil has not kept pace with international market demands. Studying the molecular mechanism of tung oil biosynthesis will help enable the bioengineering of improved varieties to produce a greater yield of high-value oils. Tung oil is primarily made up of unsaturated fatty acids (UFAs). Fatty acid desaturase 2 (FAD2) is the main enzyme responsible for desaturation in UFA biosynthesis, and has been studied extensively in plants and fungi (Lacombe et al. 2009; Radovanovic et al. 2014). In V. fordii, FAD2 (VfFAD2) is responsible for conjugated fatty acid biosynthesis (Dyer et al. 2002). Diacylglycerol acyltransferases (DGATs) catalyze the final step in plant lipid assembly by transferring an acyl group from acyl-coenzyme A to diacylglycerol (DGAT) via the classical Kennedy pathway (Bouvier-Navé et al. 2000). DGATs have been identified in many species, including Jatropha curcas (Xu et al. 2011), Olea europaea (Banilas et al. 2011), and Ricinus communis (Cagliari et al. 2010). Jako et al. (2001) demonstrated that seed-specific over-expression of Arabidopsis DGAT cDNA resulted in enhanced oil deposition and increased average seed weight. The expression of NcDGAT2 from Neurospora crassa in maize improved its kernel oil production by 26 % (Oakes et al. 2011). V. fordii expresses DGAT1, DGAT2, and DGAT3; of these, DGAT2 is speculated to be the main contributor to tung oil (Shockey et al. 2006; Cao et al. 2013). Therefore, FAD2 and DGAT2 are considered key enzymes in fatty acid biosynthesis and seed oil assembly, respectively. Vanhercke et al. (2013) described fatty acid biosynthesis as a ‘push’ process and the assembly of fatty acids into lipids as a ‘pull’ process. However, few reports have described the interaction between fatty acid biosynthesis (push) and fatty acid accumulation in lipids (pull). Vanhercke et al. (2013) reported that the transient co-expression of WRI1 and DGAT1 in Nicotiana benthamiana had a synergistic effect on triacylglycerol biosynthesis. To investigate the effect of the interaction between the VfFAD2 and VfDGAT2 pathways on fatty acid accumulation, we transformed Arabidopsis thaliana and the oleaginous yeast Rhodotorula glutinis with VfFAD2 conjugated to VfDGAT2. Our results show the effect of VfFAD2-VfDGAT2 compared with the individual genes in a transgenic yeast and model plant, and should aid in the efforts to engineer crops and model species to produce valuable UFAs.
Materials and Methods Plants, yeast, vectors and media Tung trees were collected from the National Tung Tree Gene Pool (constructed in 1979) in Dongfanghong Forest
13
Mol Genet Genomics
Farm, Jinhua, Zhejiang Province, China. Chinese Academy of Forestry issued the permit for the sample collection in the pool. Wild-type A.thaliana ecotype Columbia was used in this study. The oleaginous yeast R. glutinis was obtained from the China General Microbiological Culture Collection Center (CGMCC number 2.704). The media used were Luria–Bertani broth, yeast extract peptone dextrose medium, induction medium, selective medium (SM) (20 g/L glucose and 6.7 g/L yeast extract, pH 6.0), and lipid fermentation medium (LFM). The media were sterilized by autoclaving at 115 °C for 15 min. The vectors, pBI121 and pCAMBIA1301, were preserved in our laboratory. Our study does not require an ethics statement. RNA extraction and gene isolation Total RNA from tung tree oilseeds and A. thaliana leaves and seeds was extracted using an EASYspin Plant RNA Mini Kit (Aidlab Biotech, Beijing, China). Total RNA from wild-type and transformed R. glutinis was extracted using an RN10-EASY Spin Yeast Rapid Extraction Kit (Aidlab Biotech). RNA integrity was confirmed using a UV–Vis Spectrophotometer Q5000 (Quawell, San Jose, CA, USA). The full-length open reading frames (ORFs) of VfFAD2 (1152 bp; GenBank Accession number: AF525534) and VfDGAT2 (969 bp; GenBank Accession number: DQ356682) were amplified from tung tree seed cDNA using the specific primers detailed in Table 1. Co‑introduction of VfFAD2–VfDGAT2 into R. glutinis and A. thaliana using Agrobacterium tumefaciens‑mediated transformation The foot-and-mouth disease virus 2A (FMDV-2A) sequence was used to conjugate VfFAD2 with VfDGAT2. The recombinant expression vector contained a unique ORF for VfFAD2– FMDV2A–VfDGAT2, with an initiation codon at the beginning of the VfFAD2 sequence and a stop codon at the end of the VfDGAT2 sequence. FMDV-2A is cleaved in eukaryotes. The expression vector pBI121–VfDGAT2 was generated by subcloning VfDGAT2 into pBI121 at the BamH I/Sac I restriction sites. Similarly, VfFAD2 was sub-cloned into pBI121 at the Xba I/Sac I restriction sites to create the expression vector pBI121–VfFAD2. All vectors were transformed into A. tumefaciens EHA105 cells. The R. glutinis transformants were primarily selected on SM I plates using streptomycin (60 μg/ mL), cephalosporin (200 μg/mL), and kanamycin (100 μg/ mL), prior to further selection on SM II plates containing cephalosporin (300 μg/mL) and kanamycin (150 μg/mL). The primary 21 A. thaliana transgenic lines were produced, and T1 plantlets grown to maturity, and T2 seeds harvested. Homozygous lines were identified in T3 generation. Gene expression was identified using the specific primers shown in Table 1.
Mol Genet Genomics Fig. 1 Relative expression of VfFAD2 and VfDGAT2 in R. glutinis transformants. RgACT1 was used as a reference gene (GenBank accession number: AB248916) with specific primers (Table 1). Gene expression varied among the transformants. VfFAD2 expression was greatest in VfFAD2-1st, whereas VfDGAT2 expression was greatest in VfDGAT2-3rd
Fig. 2 Relative expression of VfFAD2 and VfDGAT2 in A. thaliana transformants. AtUBC was used as a reference gene in seeds (GenBank accession number: At5g25760) with specific primers (Table 1). Gene expression varied among the transformants. VfFAD2 expression was greatest in VfFAD2-4th, whereas VfDGAT2 expression was greatest in VfDGAT2-9th
Real‑time quantitative PCR (RT‑qPCR) For RT-qPCR, 1 µg of RNA was reverse transcribed using SuperScript Reverse Transcriptase (Life Technologies, Carlsbad, CA, USA) for gene-specific cDNA synthesis. RTqPCR was performed in 96-well plates using a 7500 RealTime PCR System (Life Technologies) and a SYBR Premix Ex Taq Kit (TaKaRa, Tokyo, Japan). The reactions were performed in an ABI7300 Real-Time PCR System (Life Technologies). The program used was as follows: initial desaturation step at 95 °C for 30 s, followed by 40 cycles of 95 °C for 5 s and 60 °C for 31 s. A melting curve was performed
after the PCR cycles to verify the specificity of amplification. Three independent experiments were performed in quadruplicate. The relative gene expression level was calculated using the delta–delta method and the 2–[ΔCPsampleΔCPcontrol] equation according to Pfaffl (2001). The specific primers used for RT-qPCR are shown in Table 1. Lipid extraction and fatty acid profile analysis Fatty acids from R. glutinis were extracted by hot methanolic HCl method after incubation in LFM at 30 °C for 60 h. Lipids were obtained from seeds of A. thaliana by soxhlet
13
Mol Genet Genomics
Table 1 Primers for gene amplification and RT-qPCR in the R. glutinis and A. thaliana transformants Gene symbol
GenBank accession number
Primer sequence (5′ → 3′) (forward/reverse)
Product length (bp)
VfFAD2
AF525534
TATCTTCAGGTTGGGCAACGA/CGATTGAACGGCCTACATTAT
1267
VfDGAT2
DQ356682
ATGGGGATGGTGGAAG/TCAAAAAATTTCAAGTTTAAGG
969
VfFAD2
AF525534
TGTCCTCCGTTCATTCTCAT/GCAAGACGGTAGAGACCAAA
585
VfDGAT2
DQ356682
CTTTCGTTCTGATCGTGCTT/TAGGCTGTGGATTTTGCTTC
542
VfFAD2
AF525534
AGCATCCGCTGGTTCTCTAA/GCAAGAACACCAGCATCAGA
212
VfDGAT2
DQ356682
TGGCTCTTTCCATTTCATCC/CGTAAGCACGATCAGAACGA
230
RgACT1
AB248916
ACTTTGAGCAGGAGATGCAG/GACATGACAATGTTGCCGTA
235
ArUBC
At5g25760
CTGCGACTCAGGGAATCTTCTAA/TTGTGCCATTGAATTGAACCC
Aractin2
NM_112764.3
GATTCAGATGCCCAGAAGTCTTGTTCC/GATTCCTGGACCTGCCTCATCATACTC
61 347
Table 2 Lipids increasing folds in R. glutinis and A. thaliana transformants with VfFAD2-VfDGAT2 R. glutinis transformants
Total fatty acids increased folds
A. thaliana transformants
Total fatty acid increased folds (%)
Vector
Vector
VfFAD2-1st
– –
VfFAD2-4th
– –
VfFAD2-2nd
–
VfFAD2-5th
–
VfFAD2-3rd
–
VfFAD2-6th
–
VfDGAT2-1st
0.55 ± 0.01
VfDGAT2-6th
0.21 ± 0.01 (6.76)
VfDGAT2-2nd
2.86 ± 0.020
VfDGAT2-7th
0.23 ± 0.00 (7.52)
VfDGAT2-3rd
6.24 ± 0.031
VfDGAT2-8th
0.17 ± 0.01 (5.65)
VfFAD2-DGAT2-1st
1.63 ± 0.016
VfFAD2-DGAT2-9th
0.17 ± 0.00 (5.62)
VfFAD2-DGAT2-2nd
2.76 ± 0.013
VfFAD2-DGAT2-11th
0.18 ± 0.01 (5.99)
VfFAD2-DGAT2-12th
0.21 ± 0.00 (6.96)
Fig. 3 Fatty acid profiles of R. glutinis transformants expressing VfFAD2 and VfDGAT2. The data are presented as the mean ± standard error (n = 3). VfFAD2-1st, VfFAD2-2nd, and VfFAD2-3rd were three R. glutinis transformants carrying the VfFAD2 gene. VfDGAT2-1st, VfDGAT2-2nd, and VfDGAT2-3rd were three R. glutinis transformants with the VfDGAT2 gene. VfFAD2-DGAT2-1th and
VfFAD2-DGAT2-2th were two R. glutinis transformants carrying VfFAD2-VfDGAT2 genes. **p