Overexpression of endoplasmic reticulum omega-3 fatty acid desaturase gene improves chilling tolerance in tomato

Plant Physiology and Biochemistry 47 (2009) 1102–1112

Contents lists available at ScienceDirect

Plant Physiology and Biochemistry journal homepage: www.elsevier.com/locate/plaphy

Research article

Overexpression of endoplasmic reticulum omega-3 fatty acid desaturase gene improves chilling tolerance in tomato Chao Yu a, Hua-Sen Wang b, Sha Yang a, Xian-Feng Tang a, Ming Duan a, Qing-Wei Meng a, * a b

College of Life Science, State Key Laboratory of Crop Biology, Shandong Agricultural University, Tai’an 271018, PR China Zhejiang Forestry University, Hangzhou 310029, PR China

a r t i c l e i n f o

a b s t r a c t

Article history: Received 3 March 2009 Accepted 17 July 2009 Available online 30 July 2009

An endoplasmic reticulum-localized tomato omega-3 fatty acid desaturase gene (LeFAD3) was isolated and characterized with regard to its sequence, response to various temperatures and function in transgenic tomato plants. Northern blot analysis showed that LeFAD3 was expressed in all organs tested and was markedly abundant in roots. Meanwhile, the expression of LeFAD3 was induced by chilling stress (4  C), but inhibited by high temperature (40  C). The transgenic plants were obtained under the control of the cauliflower mosaic virus 35S promoter (35S-CaMV). Northern and western blot analyses confirmed that sense LeFAD3 was transferred into tomato genome and overexpressed. Level of linolenic acids (18:3) increased and correspondingly level of linoleic acid (18:2) decreased in leaves and roots. After chilling stress, the fresh weight of the aerial parts of transgenic plants was higher than that of the wild type (WT) plants, and the membrane system ultrastructure of chloroplast in leaf cell and all the subcellular organelles in root tips of transgenic plants kept more intact than those of WT. Relative electric conductivity increased less in transgenic plants than that in WT, and the respiration rate of the transgenic plants was notably higher than that of WT. The maximal photochemical efficiency of PSII (Fv/Fm) and the O2 evolution rate in WT decreased more than those in transgenic plants under chilling stress. Together with other data, results showed that the overexpression of LeFAD3 led to increased level of 18:3 and alleviated the injuries under chilling stress. Ó 2009 Elsevier Masson SAS. All rights reserved.

Keywords: Chilling tolerance Endoplasmic reticulum-localized omega-3 fatty acid desaturase Gene expression Tomato (Lycopersicon esculentum) Trienoic fatty acids

1. Introduction Low temperature is one of the major environmental factors that limit plant growth and geographical distribution. Chillingsensitive plant species, including important vegetable crops such as cucumber, tomato, and sweet pepper, suffer chilling injury when temperature is below 12  C. Since Lyons and Chapman [16] considered that chilling stress could impair membrane permeability by the transition of membrane lipids from a liquidcrystalline phase to a gel phase, many experiments have suggested that chilling tolerance is related to the composition and structure of plant membrane lipids. The changes of unsaturated fatty acids

Abbreviations: DAs, dienoic fatty acids; DGDG, digalactosyldiglycerol; ER, endoplasmic reticulum; LeFAD3, Lycopersicon esculentum omega-3 fatty acid desaturase gene; LeFAD3, Lycopersicon esculentum omega-3 fatty acid desaturase; MGDG, monogalactosyldiacylglycerol; PC, phosphatidylcholine; PE, phosphatidylethanolamine; PFD, photon flux density; PG, phosphatidylglycerol; PVDF, polyvinylidene fluoride; SDS-PAGE, sodium dodecyl sulfate polyacrylamide gel electrophoresis; SQDG, sulfoquinovosyldiglycerol; TAs, trienoic fatty acids; 18:1, oleic acid; 18:2, linoleic acid; 18:3, linolenic acid. * Corresponding author. College of Life Science, Shandong Agricultural University, Tai’an 271018, P.R. China. Tel.: þ86 538 8249606; fax: þ86 538 8226399. E-mail address: [email protected] (Q.-W. Meng). 0981-9428/$ – see front matter Ó 2009 Elsevier Masson SAS. All rights reserved. doi:10.1016/j.plaphy.2009.07.008

have been considered to be one of the factors in metabolic adaptation of higher plants to temperature stress [21,23]. It is commonly observed that there has be very high level of fatty acid unsaturation in the higher plants. The membrane lipids are composed of dienoic fatty acids (DAs) and trienoic fatty acids (TAs), which make up about 70% of the total fatty acids in the leaf lipids and 55–70% of the lipids in non-photosynthetic tissues such as roots [8]. Fatty acids, especially TAs are predominantly found in the membrane glycerolipids with phosphatidylcholine being the major component of the endoplasmic reticulum (ER) and monogalactosyldiacylglycerol (MGDG), the most abundant lipid in the chloroplast endomembranes in vegetative cells. There are two pathways that have been described in higher plants, namely the prokaryotic and the eukaryotic pathways for polyunsaturated fatty acid biosynthesis. However, the two pathways are not mutually exclusive [22]. In fact, linolenic acid (18:3) can be synthesized on ER phosphatidylcholine (eukaryotic pathway), or can be synthesized from the sequential desaturation of oleic acid (18:1) to 18:3 via linoleic acid (18:2) on chloroplast MGDG (prokaryotic pathway). And the synthesis of hexadecatrienoic acids (16:3) from its saturated precursor appears to reside specifically in plastid-localized omega-3 desaturases. Lipid transfer between the compartments results in contributions from both sets of omega-3

C. Yu et al. / Plant Physiology and Biochemistry 47 (2009) 1102–1112

fatty acid desaturases to the total cellular content of 18:3. The omega-3 fatty acid desaturase is a key enzyme for the formation of TAs and catalyzes the desaturation of lipid-linked DAs. There are three distinct desaturases, which are active at the final step of converting 18:2 to 18:3, and that were identified in Arabidopsis: two plastidial enzymes, FAD7 and FAD8 [14,30] and one ER-type enzyme, FAD3 [1]. Each of these enzymes is encoded by a single gene. It is widely accepted that FAD8 has a predominant role in lipid desaturation at low temperature whereas FAD7 appears to be more important in regulating the level of trienoic fatty acids during leaf maturation [11]. FAD3 is located in the ER and is a membrane-bound enzyme, which converts 18:2 to 18:3 in non-photosynthetic tissues. FAD3 gene was expressed in both roots and leaves in tobacco [7], which had a temperature-dependent post-transcriptional regulation in root tips. However, the level of FAD3 transcripts in flax seeds peaked after flowering, and transcripts were not detectable in leaves and roots [34]. The level of TaFAD3 mRNA in root tips grown at 10  C was slightly higher than in those grown at 30  C [12]. It was reported that abscisic acid (ABA) induced the expression of FAD3 in rapeseed [40]. But ABA did not induce the ER-type desaturase in mung bean [37]. To explain that conflicts people described the successful construction of transgenic Arabidopsis displaying spatial and temporal FAD3 expression assayable in vivo as luciferase bioluminescence [19]. The up- and down-regulation of transcript level of FAD3 was useful to modify the 18:3 content in tobacco [7]. FAD3 genes have been introduced in several plants [4,17,39]. Overexpressing the tobacco FAD3 can accumulate high level of TAs in both the explastid lipids of leaf and root tissues [28]. Overexpression of FAD3 and FAD7, which led to increased levels of 18:3 in roots, seeds and leaves, resulted in plants characterized by better maintenance of membrane fluidity and increased tolerance to cold and chilling [23]. It was reported that FAD3þ cell suspensions exhibited cold resistance [17]. Nevertheless, there was no significant change in the membrane properties and adaptability to temperatures in the Arabidopsis fad3 mutant which has reduced TA level in the extraplastid lipids [21], which was also seen in transgenic tobacco with increased extraplastid TA level [7]. It appears that such TA-mediated alleviation of chilling injuries is limited to certain plant species. The photoinhibition of PSII could occur under chilling stress [15], and D1 protein of PSII reaction center was thought to be the target of photoinhibition when plants were exposed to high or mediated irradiance. The extent of the unsaturation of the fatty acid in thylakoid membrane is important for the protection of photosynthetic machinery at low temperature. Although the expression pattern of FAD3 had been investigated in model plants, such as tobacco and Arabidopsis, the expession and function of ER-type fatty acid desaturase gene in vegetable crops in response to low temperature stress is still unclear. Tomato is an important chilling-sensitive vegetable. To understand the physiological effects and functional mechanism of fatty acids at low temperatures, we isolated and characterized ER-type LeFAD3 from tomato and introduced it into the same plant. Results showed that LeFAD3 expression was temperature-dependent regulated. Overexpression of ER-type LeFAD3 increased significantly the content of 18:3 in roots and leaves and enhanced the chilling tolerance. Our results pointed to the potential of exploiting FAD3 overexpression as a tool to ameliorate chilling tolerance. 2. Materials and methods 2.1. Plant materials and treatments Seeds of tomato cultivar (Lycopersicon esculentum cv. Zhongshu 4) were used and germinated between moistened filter paper at 25  C

1103

for 3 d. Sprouted burgeons were then planted in 13.5 cm-diameter plastic pots (one plant per pot) filled with sterilized soil and grown at 25–30/15–20  C (day/night temperature regime) under a 14 h photoperiod (300–400 mmol m2 s1 PFD) in a greenhouse. When the sixth leaf was fully expanded (about 3-month-old seedlings), the plants were treated with temperature stress. Then the treated roots, stems, leaves, petals and fruits were immediately frozen in liquid nitrogen and stored at 80  C until use. 2.2. Isolation and sequencing of LeFAD3 Total RNA was isolated from tomato leaves using the total RNA isolation system (Promega Corporation, Madison, Wis., USA) and used for reverse-transcription polymerase chain reaction (RT-PCR) and RNA gel blot analysis. A 2 mg sample of RNA was denatured at 70  C for 5 min and 2 ml AMV reverse transcriptase (Promega) was added. The transcription reaction was mixed briefly and then incubated at 42  C for 1 h, and terminated at 85  C for 10 min. To isolate the FAD3 from tomato, a 628 bp fragment was amplified from cDNA prepared from tomato leaves. Primers FP1: 50 -T/C/A CC A/C/T AAGCA T/C TGTTGGGT-30 and FP2: 50 -G/ CAAGTCCAG/ACCACAT-30 , which contained conserved sequence were designed based on the homology to FAD3 from tobacco, Brassica napus and Arabidopsis. The cDNA amplification products were cloned into the pMD-18T vector and sequenced. The 50 - and 30 -ends of the gene were PCR-amplified from the cDNA. The 50 -RACE PCR was carried out by using the gene-specific primer: FP3: 50 -CCAACGACGGTAGAAAGAT-30 and an abridged universal amplification primer AAP according to the manufacturer’s instructions (GGBCO-BRL Kit). 30 -RACE PCR was carried out by using the gene-specific primer: FP4: 50 -ATGATTTTTGTGATGTGGCT-30 and B26. The putative full-length of LeFAD3 was carried out from the cDNA using 50 and 30 specific primers: FP5: 50 -CGCCGCCAA CAACTCATCT-30 and FP6: 50 -TTACTTATCCTTTTTACCAG-30 . The PCR amplification was as follows: initial denaturation at 94  C for 5 min, followed by 35 cycles, 94  C for 50 s, 50  C for 50 s, 72  C for 1 min, a final extension cycle of 72  C for 10 min and termination of the reaction at 4  C. All the primers were synthesized from Shanghai Bioasia Bio-engineering Limited Company. Nucleotide and deduced amino acid sequences were analyzed using DNAman version 5.2 (Lynnon Biosoft, USA). Sequence data from this article has been deposited at GenBank under accession number EU251190. 2.3. Southern blot analysis Genomic DNA (10 mg) from tomato leaves was digested with XbalI, XholI and BamHI separated by electrophoreses on an 0.8% agarose gel and blotted onto a nylon membrane by an alkaline transfer method. The full-length sequence of the LeFAD3 cDNA clone was used as a probe. Then the membrane was hybridized with the probe (above), washed with 0.1 SSC and 0.1% SDS at 65  C for 1 h and then autoradiography was performed at 80  C, respectively. 2.4. RNA gel blot analysis Twenty micrograms of total RNA were separated in a 1.2% agarose formaldehyde gel and transferred to nylon membrane as described by Sambrook et al. [25]. RNA was fixed on the membrane by a cross-linking with UV-light. Pre-hybridization was performed at 65  C for 12 h. The full-length sequence of the LeFAD3 cDNA clone was used as a probe and labeled with [a-32P]-dCTP by randomly prime labeling method (Prime-a-Gene-Labeling System, Promega). After 24 h hybridization, filters were washed subsequently in 2 SSC (1 SSC is 0.15 M NaCl, 0.015 M sodium citrate, pH 7) with 0.2%

1104

C. Yu et al. / Plant Physiology and Biochemistry 47 (2009) 1102–1112

SDS and 0.2 SSC with 0.2% SDS at 42  C. Autoradiography was performed at 80  C. 2.5. SDS-PAGE and immunological analysis A coding region of LeFAD3 in the pMD18-T vector about 309 bp was subcloned into the pET-30a(þ) vector between the BamHI and SacI sites. A recombinant of prokaryotic expression vector pETLeFAD3 was constructed and transformed to E. Coli BL21 and then expressed by inducing with IPTG. Depositions were solubilized in the presence of 2 SDS loading buffer and separated by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) [13] using 10% separate gels and 4% concentrated gels and containing 10% SDS. The strong induced fusion protein bands were collected into phosphate buffer (PBS) solution and were used to immunize white mice to obtain antiserum. The secondary antibody was peroxidase-conjugated goat anti-mouse IgG bought from biotechnological company. The antibody was used at a dilution of 1:500 and the secondary antibody was used at 1:5000. For immunoblotting, polypeptides were electrophoretically transferred to a polyvinylidene fluoride (PVDF) membranes (Millipore, France) and proteins in wild type and transgenic lines were detected with antibodies raised against LeFAD3. Protein content was determined by the dye-binding assay. 2.6. Plasmid construction and Agrobacterium-mediated transformation of tomato plants The full-length tomato LeFAD3 cDNA was transferred into tomato genome downstream of the 35S-CaMV promoter to form sense constructs (pBI-LeFAD3). The 35S-CaMV LeFAD3 constructs were first introduced into Agrobacterium tumefaciens LBA4404 by the freezing transformation method and verified by PCR and sequencing. Leaf disk transformation using wild type tomato plants was performed as described by Horsch et al. [10]. Discs infected with A. tumefaciens were incubated on medium for inducing shoots. After a few weeks, the regenerated shoots were transferred to medium for inducing roots. Both media contained cefotaxime sodium (250 mg mL1) and kanamycin (50 mg mL1). Transgenic plants were screened using kanamycin selection generated by the incubation of transformed tomato leaf disks [9]. As a consequence, four sense lines were obtained. Each transgenic line seemed to represent an independent integration event since a specific DNA fragment in each line was observed by genomic DNA gel blot analysis (data not shown). In order to further assess the expression of LeFAD3 in screened plants of transgenic tomato, the 3-month-old seedlings were subjected to molecular and physiological analysis. 2.7. Growth condition To ensure the humidity remained constant, the seeds of wild type and three transgenic lines (T1-6, T1-16, T1-20) were grown on Murashige-Skoog (MS) agar media in closed Petri dishes and were treated at 16  C/8  C and 25  C/20  C (day/night temperature regime) under a 14 h photoperiod (300–400 mmol m2 s1 PFD) in an illuminated incubation chamber (GXZ-260C). After the plants were cultivated for 30 days, the fresh weight of their aerial parts was detected.

two-dimensional thin layer chromatography (TLC) [36]. For quantitative analysis, individual lipids were separated by TLC, scraped from the plates, and used to prepare fatty acid methyl esters. Fatty acid composition of individual lipids was determined by gas chromatography [3]. 2.9. Measurement of relative electric conductivity One half gram of fresh samples were put into 10 ml distilled water and vacuumized for 30 min, surged for 3 h to measure the initial electric conductivity (S1). Then the cuvette was filled with samples and distilled water, and the mixture was cooked 30 min to determine final electric conductivity (S2). Relative electric conductivity (REC) was evaluated as: REC (%) ¼ S1  100/S2. 2.10. Measurement of O2 evolution and respiration rate O2 evolution rates were determined using a modified Clarketype O2 electrode unit (Hansatech, Kings Lynn, UK) as described by Walker [35]. After treatments, leaf discs (2 cm2) were vacuumized with 0.1 M NaHCO3. Then the leaf disks were dissected into pieces of about 1 mm2. The reaction mixture was 0.1 M NaHCO3 to maintain a high concentration of CO2. The light intensity was 800 mmol m2 s1 PFD. Respiration rates of leaves and roots were determined using a modified Clarke-type O2 electrode unit (Hansatech, Kings Lynn, UK) as described by Walker [35]. Samples (0.1 g of fresh weight) were kept in the dark for 30 min before respiration measurements were taken. 2.11. Measurement of chlorophyll a fluorescence Chlorophyll fluorescence was measured with a portable fluorometer (FMS2, Hansatech, England) according to the protocol described by van Kooten and Snel [33]. The minimal fluorescence (Fo) with PSII all reaction centers open was determined by a modulated light which was low enough to induce any significant variable fluorescence (Fv). The maximal fluorescence (Fm) with all reaction centers closed was determined by a 0.8 s saturating light of 7000 mmol m2 s1 on dark-adapted leaf. The maximal photochemical efficiency (Fv/Fm) of PSII expressed as the ratio of variable fluorescence to maximum yield of fluorescence. Fv/Fm ¼ (FmFo)/Fm. 2.12. Electron microscopy The transgenic line and WT plants were used for microscopic analysis. Leaf and root tip (2 mm length) samples were collected from five plants from each genetic source after chilling stress. Whole samples were pinned onto Silgard-coated plastic and overlaid with a fixing solution (3.5% glutaraldehyde). Thereafter, samples were washed with 0.1 M PBS buffer, then briefly postfixed in 1% osmium tetroxide, dehydrated in an ascending ethanol series from 10 to 70% ethanol, preceded the endosmosis, embedment and polymerization of material into Epon812 resin. Thin sections were cut from the embedded samples using an LKB-V ultramicrotome. Sections were stained with uranium acetate and lead citrate, and examined under a transmission electron microscope (JEOL-1200EX). 3. Results

2.8. Lipid and fatty acid analyses 3.1. Characterization of the tomato cDNA clone The overall fatty acid composition of leaves and other tissues was extracted and separated into each lipid class as described by Miquel and Browse [20]. Individual lipids were extracted as described by Siegenthaler and Eichenberger [29], and separated by

A cDNA was isolated from tomato. The full-length sequence of the cDNA consists of 1184 bp nucleotides and a 1134 bp open reading frame at position 51–1184 bp, encoding a 377-residue

C. Yu et al. / Plant Physiology and Biochemistry 47 (2009) 1102–1112

1105

Fig. 1. Deduced amino acid sequence alignment of FAD3 from six plant species. Four conserved acyltransferase motifs are indicated above the aligment (I, II, III and IV). Identical and similar amino acid residues are shaded black, and dashes indicate gaps introduced to optimize alignment. The accession numbers in GenBank of FAD3 are as follows: LeFAD3(EU251190), Lycopersicon esculentum Zhongshu 4; NtFAD3 (P48626), Nicotiana tabacum; EgFAD3 (BAF30810), Elaeis guineensis; GmFAD3 (ABG66302), Glycine max; BnFAD3(P48624), Brassica napus; AtFAD3(NP850139), Arabidopsis thaliana; DsFAD3 (ABK91879), Descurainia sophia; AtFAD7(D14007), A. thaliana; LeFAD7(AY157317), L. esculentum Zhongshu 4; AtFAD8(L27158), A. thaliana; BpFAD8(AAN17502), Betula pendula. The alignment was done using DNAman version 5.2.

1106

C. Yu et al. / Plant Physiology and Biochemistry 47 (2009) 1102–1112

polypeptide. The deduced amino acid sequence of the cDNA showed that it encoded a polypeptide of approximately 42 kDa. This cDNA was designated as LeFAD3 and was submitted to the GenBank databases under accession number EU251190 (address is as follows: http://www.ncbi.nlm.nih.gov/entrez/viewer.fcgi?db¼ nuccore&id¼160334209) and contained four conserved regions (Fig. 1). Some of the conserved motifs (shaded) included four histidine boxes (underlined in Fig. 1). Three of the four histidine boxes were highly conserved with all the membrane bond fatty acid [27]. According to the analysis using DNAMAN software, the putative amino acid sequence had significant homology to other omega3 fatty acid desaturases. According to the analysis, the identity of putative amino acid sequence of LeFAD3(L. esculentum Zhongshu 4) with NtFAD3(Nicotiana tabacum), EgFAD3 (Elaeis guineensis), GmFAD3(Glycine max), BnFAD3(B. napus), AtFAD3(Arabidopsis thaliana), DsFAD3(Descurainia sophia), AtFAD7(A. thaliana), LeFAD7 (L. esculentum Zhongshu 4), AtFAD8 (A. thaliana), BpFAD8 (Betula pendula) was 62.5%, 60.58%, 61.30%, 62.74%, 57.99%, 58.42%, 56.46%, 56.646%, 60.58% and 61.78%, respectively. Southern blot analysis of tomato genomic DNA digested with three different restriction enzymes and using the isolated tomato cDNA as a probe, resulted in a strong hybridizing band in each case. Two weaker bands were observed when genomic DNA was digested with BamHI (Fig. 2), which also cut into the cDNA sequence. The analysis showed that other closely related isoenzyme genes encoding omega-3 FADs might occur in the whole genome. 3.2. Expression of LeFAD3 in tomato It was showed that LeFAD3 constitutively expressed in roots, stems, leaves, petals and fruits of wild type plants at 25  C whereas

Fig. 3. Expression of LeFAD3 in different organs of the 3-month-old tomato seedlings at different temperatures. Total RNA was extracted from roots, stems, petals, fruits and leaves of wild type plants at 25  C. The probe was labeled with [a-32P]-dCTP by randomly prime labeling method (Prime-a-Gene-Labeling System, Promega). The ethidium bromide staining of the RNA gel was shown as control for loading (rRNA).

the transcripts were relatively more abundant in roots than in stems, petals and fruits (Fig. 3). And the transcripts were relatively lower in leaves than in other organs. Leaves and roots were used as materials in the following RNA gel blot. The response of LeFAD3 expression in leaves and roots to different temperatures was detected by northern blotting (Fig. 4). The highest transcript level of LeFAD3 was observed at 4  C and a nearly undetectable band of mRNA was noted at 40  C for 6 h (Fig. 4A, B). Transcript level both in the leaves and roots at low temperatures slowly increased, reaching the first peak at 6 h, and decreased thereafter, followed by a slight increase bringing the level to a second peak between 24 and 48 h (Fig. 4C,E). On the contrary, transcript level both in the leaves and roots at high temperatures decreased, reaching the lowest at 6 h, and increased slightly from 24 to 48 h (Fig. 4D,F). Results indicated that LeFAD3 was expressed extensively from 4 to 40  C in leaves and roots, and low temperature induced the expression of LeFAD3 obviously although the level of LeFAD3 mRNA changed slowly (Fig. 4). 3.3. Molecular characterization of the transgenic plant Transgenic plants infecting with A. tumefaciens carrying LeFAD3 were detected by PCR after the first screening with kanamycin (50 mg mL1) (data not shown). Twenty-five individual kanamycin resistant lines were obtained from tissue culture. These initial kanamycin resistant plants were named T0. And the progeny obtained from T0 were named T1. The T1 lines turned out to be 3:1 ratio segregation (Table 1). Four lines named T1-1, T1-6, T1-16 and T1-20 were selected for RNA gel blot and western blot analysis. From the four lines we selected T1-6, T1-16 and T1-20 for physiological measurement. There were no obvious morphological differences between the transgenic and WT plants. T1 kanamycin-tolerant plants were checked by PCR. The upriver primer of PBI121 and 30 primer of LeFAD3 were used in the amplification and an intense 1184 bp band corresponding in size to the LeFAD3 product was obtained from kanamycin-tolerant plants. While nothing was obtained from WT plants (data not shown). RNA gel blot showed that all kanamycin-tolerant plants had strong positive signals, and a weak signal was found in WT plants. Western blot analysis with the antiserum against LeFAD3 revealed the presence of the strong positive protein signals corresponding to LeFAD3 in transgenic plant leaves, while a weak signal in WT tomato (Fig. 5). 3.4. Changes of fatty acid composition of WT and transgenic tomato plants

Fig. 2. Southern blot analysis of LeFAD3. Digestion with XbaI (E), XhoI (D), BamHI (C) from WT plants of the 3-month-old seedlings. The probe was labeled with [a-32P]dCTP by randomly prime labeling method (Prime-a-Gene-Labeling System, Promega).

The transgenic plants overexpressing LeFAD3 produced higher amounts of 18:3 compared with WT (Table 2). This increased level

C. Yu et al. / Plant Physiology and Biochemistry 47 (2009) 1102–1112

1107

Fig. 4. Expression of LeFAD3 in roots (A) and leaves (B) of the 3-month-old tomato seedlings at different temperatures for 6 h. Expression of LeFAD3 in roots was treated at 4  C (C) and 40  C (D) for 48 h, and expression of LeFAD3 in leaves was treated at 4  C (E) and 40  C (F) for 48 h. Total RNA was extracted from wild type plants. About 20 mg of total RNA was analyzed by RNA gel blot using full-length sequence of the LeFAD3 as probe.

of 18:3 was accompanied by a concomitant decrease in the level of 18:2. Therefore, the phenotype of the transgenic plants were more evident when the 18:3/18:2 ratio was compared. The increase in the level of 18:3 was more pronounced in roots than leaves (Table 2). Thus, an 18:3/18:2 ratio in the transgenic plants was almost 2–3 times higher than that in WT for roots and less than 2 times higher in leaves. It was observed that the ratio of 18:3/18:2 in T1-6 was the highest. To obtain more information concerning the biochemical effects of the tansgentic tomato, we separated individual lipid classes from leaf extracts of WT and the transgenic plants. The analysis of leaf lipids revealed that the major chloroplast lipids, MGDG, DGDG and SQDG, had a litter change but PC and PE increased significantly in the proportion of each lipid. However, the level of 18:3 markedly increased in each lipid in the transgenic tomato (Table 3). The significant changes in the ratio of 18:3/18:2 were observed in transgenic lines. The changes of 18:3 level in all six lipid classes of T1-6 were more obvious than those of T1-16 and T1-20. The 18:3 levels of MGDG, DGDG, SQDG, PG, PC and PE in T1-6 were 79.78%, 60.98%, 45.84%, 36.15%, 45.53% and 51.20%, but those

in WT were 73.48%, 55.04%, 37.92%, 28.30%, 26.99% and 30.67%, respectively. As shown in Tables 2 and 3, the amounts of 18:3 increased in transgenic plants. 3.5. Growth analysis It was showed that there were no differences in the growth of the transgenic lines (T1-6, T1-16, T1-20) and WT plants grown at

Table 1 Segregations and expression of exogenous LeFAD3 in T1 of the transgenic lines (T1-6, T1-16 and T1-20) X20.05 ¼ 3.84. Line (number)

Observed PCR results

Theoretically-predicted PCR results

Positive (þ)

Negative ()

Positive (þ)

Negative ()

T16(96) T116(109) T120(100)

68 75 74

28 34 26

72 82 75

24 27 25

x2

0.89 2.23 0.05

Fig. 5. RNA gel and western blot analyses of LeFAD3 in tomato plants. Total RNA and protein were extracted from the leaf tissue of the 3-month-old seedlings of WT and transgenic plants, respectively. The probe of RNA gel blot was labeled with [a-32P]dCTP. About 20 mg of total RNA was analyzed by RNA gel blot. The antibody against LeFAD3 was produced by immunizing white mice and used at a dilution of 1:500. The dilution of secondary antibody was used at 1:5000. About 35 mg of total protein was analyzed by the dye-binding assay. And 35 mg of total protein separated by SDS-PAGE was taken as the control protein.

1108

C. Yu et al. / Plant Physiology and Biochemistry 47 (2009) 1102–1112

Table 2 Fatty acid composition of total lipids from roots and leaves of the 3-month-old seedlings of WT and the transgenic lines (T1-6, T1-16 and T1-20). Present at trace levels (<0.1% of total fatty acid). Data are expressed as mean values  SD (n ¼ 4) and are presented as mole percent. Standard deviations between triplicates was <3% of the indicated values.

Table 3 Fatty acid composition of leaf lipids from the 3-month-old seedlings of WT and the transgenic lines (T1-6, T1-16 and T1-20). Present at trace levels (<0.1% of total fatty acid). Data are expressed as mean values  SD (n ¼ 4) and are presented as mole percentage. Standard deviations between triplicates was <3% of the indicated values.

Fatty acid

Fatty acid

Root

Leaf

Genotype Percent of Fatty acid composition* (mol %) total polar 16:0 16:1 18:0 18:1 lipids

WT

T16

T116

T120

WT

T16

T116

T120

16:0

31.76% 0.03

30.90% 0.06

30.06% 0.04

30.36% 0.08

20.59% 0.09

16.97% 0.04

17.08% 0.07

21.25% 0.02

16:1

–

–

–

–

–

1.75% 0.06

1.53% 0.02

0.42% 0.08

18:0

3.66% 0.06

2.10% 0.08

2.40% 0.05

3.86% 0.07

4.38% 0.04

2.96% 0.05

4.96% 0.04

4.80% 0.09

18:1

3.50% 0.08

3.38% 0.09

4.38% 0.06

2.43% 0.06

5.10% 0.07

3.10% 0.03

4.10% 0.05

4.77% 0.04

18:2

41.35% 0.06

27.77% 0.04

30.9% 0.02

28.23% 0.05

11.70% 0.08

7.91% 0.02

10.24% 0.07

8.10% 0.07

T1-6

18:3

19.74% 0.03

35.85% 0.06

32.25% 0.06

35.12% 0.04

58.22% 0.07

69.07% 0.03

62.09% 0.09

64.76% 0.08

T1-16

18:3/18:2

0.48

1.29

1.04

1.24

4.97

8.73

6.07

7.99

MGDG WT T1-6 T1-16 T1-20 DGDG

T1-20 SQDG

 C/20  C

25 for 30 days (day/night temperature regime)(Fig. 6A), and the fresh weights of T1-6, T1-16, T1-20 and WT plants were (199.2  25)mg, (193.2  27)mg, (198.7  28)mg and (198.3  24)mg (n ¼ 6), respectively. However, after germination and cultivation for 30 days at 16  C/8  C (day/night temperature regime), the fresh weight of WT plants was (30.4  9.4)mg, while the aerial parts of the T1-6, T1-16, and T1-20 plants weighed (102.2  8.9)mg, (68.2  6.5)mg and (78.4  5.6)mg (Fig. 6B). The results indicated that the transgenic plants with higher 18:3 content had the higher chilling endurance.

WT

WT T1-6 T1-16 T1-20

PG

WT T1-6 T1-16 T1-20

3.6. Changes of relative electric conductivity and respiration rate It was showed that the relative electric conductivity of roots and leaves increased both in WT and transgenic plants under chilling stress (Fig. 7A,B). But it increased more slowly in transgenic plants than that in WT plants. After chilling stress for 12 h, the relative electric conductivity of roots in WT, T1-6, T1-16 and T1-20 was 25.0%, 18.3%, 19.9% and 19.3%. The relative electric conductivity of leaves in WT, T1-6, T1-16 and T1-20 was 28.0%, 22.1%, 25.3% and 24. 8%. The respiration rate increased at first and then declined under chilling stress in both WT and transgenic plants (Fig. 7C,D). After chilling stress for 12 h, the respiration rates of roots in WT, T1-6, T1-16 and T1-20 were 7.06, 10.92, 10.08 and 10.18 mmol g1 h1 FW. The respiration rates of leaves in WT, T1-6, T1-16 and T1-20 were 10.08, 13.78, 11.64 and 11.9 mmol g1 h1 FW. 3.7. Overexpression of LeFAD3 alleviates PSII photoinhibition under chilling stress In order to estimate the role of unsaturated 18:3 fatty acids in protecting the photosynthesis apparatus from chilling stress, the O2 evolution rate was determined and photoinhibition of PSII was evaluated by measuring the maximal photochemical efficiency of PSII (Fv/Fm). Under chilling stress, the O2 evolution rates and Fv/Fm in WT and transgenic tomato plants obviously decreased. This decrease was more obvious in WT than that in transgenic plants (Fig. 7E,F). After chilling stress for 6 h, the O2 evolution rates of WT, T1-6, T1-16 and T1-20 lines decreased to about 34%, 50%, 48% and 49%, respectively. After chilling stress for 12 h, the O2 evolution rates of WT, T1-6, T1-16 and T1-20 lines decreased to about 22.9%,

PC

WT T1-6 T1-16 T1-20

PE

WT T1-6 T1-16 T1-20

48.43% 0.03 48.24% 0.06 48.24% 0.02 48.78% 0.02

13.76% 0.04 10.17% 0.07 10.90% 0.04 15.44% 0.03

–

26.33% 0.05 24.03% 0.04 25.15% 0.05 25.39% 0.04

23.10% 0.07 20.67% 0.03 20.47% 0.07 19.63% 0.02

–

10.13% 0.06 9.71% 0.06 9.33% 0.08 11.52% 0.04

32.58% 0.04 29.93% 0.03 28.49% 0.05 28.66% 0.02

–

8.58% 0.03 8.27% 0.02 8.56% 0.03 8.84% 0.08

32.90% 0.04 20.79% 0.06 20.33% 0.04 38.58% 0.07

–

3.41% 0.09 5.53% 0.04 5.15% 0.07 5.47% 0.06 3.11% 0.06 4.22% 0.04 3.56% 0.03 4.10% 0.05

18:2

18:3

0.16% 0.04 2.34% 0.07 1.67% 0.05 0.66% 0.06

3.95% 0.05 3.93% 0.05 1.45% 0.02 1.96% 0.04

8.65% 0.06 3.79% 0.08 8.54% 0.09 4.48% 0.03

73.48% 0.04 79.78% 0.09 77.44% 0.06 77.46% 0.09

7.34% 0.04 6.26% 0.04 7.67% 0.02 10.46% 0.04

6.05% 0.04 6.32% 0.05 7.64% 0.05 5.41% 0.03

8.48% 0.04 5.77% 0.07 7.34% 0.08 7.07% 0.06

55.04% 0.06 60.98% 0.05 56.89% 0.04 57.44% 0.07

6.97% 0.07 8.31% 0.04 9.98% 0.03 6.93% 0.09

6.29% 0.09 6.66% 0.05 9.25% 0.06 12.45% 0.04

16.24% 0.08 9.26% 0.04 11.28% 0.03 10.04% 0.03

37.92% 0.05 45.84% 0.08 41.00% 0.02 41.93% 0.02

13.95% 0.08 18.45% 6.08% 0.02 0.04 20.24% 6.84% 0.06 0.07 – 5.71% 0.03

3.71% 0.05 2.30% 0.03 2.75% 0.02 4.06% 0.03

21.14% 0.08 13.88% 0.01 19.88% 0.06 16.26% 0.08

28.30% 0.05 36.15% 0.08 32.31% 0.04 35.40% 0.04

26.77% 0.04 26.99% 0.05 28.60% 0.04 21.01% 0.05

–

11.50% 0.04 7.73% 0.05 8.65% 0.03 2.00% 0.06

– 3.45% 0.04 3.39% 0.02 9.26% 0.07

34.74% 0.03 16.31% 0.03 18.23% 0.07 16.44% 0.07

26.99% 0.04 45.53% 0.04 41.13% 0.08 43.63% 0.09

25.97% 0.04 22.69% 0.03 22.45% 0.01 21.40% 0.04

–

2.45% 0.05 6.70% 0.07 7.32% 0.06 9.90% 0.08

20.45% 0.03 5.11% 0.05 5.70% 0.04 3.50% 0.09

20.45% 0.07 14.30% 0.02 16.30% 0.03 15.37% 0.05

30.67% 0.06 51.20% 0.08 48.23% 0.07 49.83% 0.04

– – –

– – –

– – –

– – 7.66% 0.03

– – –

38.3%, 32% and 32.2%, respectively. After chilling stress for 12 h, Fv/ Fm of WT, T1-6, T1-16 and T1-20 lines decreased 47.6%, 14.8%, 33.5% and 29.7%, respectively. These results indicated that the increase of 18:3 of tomato thylakoid membranes had a role in protecting the photosynthetic apparatus from chilling stress.

3.8. Change of leaf and root tip cell ultrastructure under chilling stress Before chilling stress, the structures of the root tip cells, leaf cells and chloroplasts appeared similar in both WT and transgenic plants. They had clearly membrane structures, a few starch grains and

C. Yu et al. / Plant Physiology and Biochemistry 47 (2009) 1102–1112

1109

Fig. 6. Growth analysis of tomato plants at different temperatures. (A) 25  C/20  C. (B) 16  C/8  C.

regulated granum stacks in cells (Fig. 8 A–D,I–L). In a word, the structure of cells was intact. After chilling stress for 12 h, the chloroplast of WT was obviously destroyed (Fig. 8E,F). The mitochondria, vacuole, endoplasmic reticulum and nuclear membrane of root

tips were not integrate (Fig. 8G,H). And some of cells were dissolved. The starch grains in chloroplast of the transgenic plants decreased and there had a few regular chloroplasts in order to sustain normal function (Fig. 8M,N). Most mitochondria vacuole and nuclear

Fig. 7. Effect of chilling stress on relative electric conductivity, the respiration rate, Fv/Fm and oxygen evolution in the 3-month-old seedlings of WT and transgenic plants. Plants were exposed to chilling stress (4  C) for 0, 3, 6, 9, 12 h before measurement of relative electric conductivity of leaves (A) and roots (B), respiration rates of leaves (C) and roots (D). Plants were exposed to chilling stress (4  C) for 6 h and 12 h before measurement of Fv/Fm (E) and oxygen evolution (F). Each point represents the means  SD of 5 measurements on each of five plants.

1110

C. Yu et al. / Plant Physiology and Biochemistry 47 (2009) 1102–1112

Fig. 8. Effect of chilling stress on leaf cell and root tip cell ultrastructure in the 3-month-old seedlings of WT and transgenic plants. Transmission electron micrographs of leaf cell ultrastructure (A, B, E, F, I, J, M, N) and root tips cell ultrastructure(C, D, G, H, K, L, O, P) in WT and T1-6. A-H were WT. I-P were T1-6. A-D and I-L were exposed to chilling stress (4  C) for 0 h E-H and M-P were exposed to chilling stress (4  C) for 12 h. A, E, I, M: 5000; B, C, D, F, G, H, J, K, L, N, O, P : 25000. Nucleus (N), vacuoles (V), mitochondria (MC), endoplasmic reticulum (ER), plasma membrane (PM), Lipid droplet (LD) and cell wall (CW).

C. Yu et al. / Plant Physiology and Biochemistry 47 (2009) 1102–1112

membrane of root tips were slightly damaged (Fig. 8O,P). Compared with the untreated control, the cell membrane system in root tips of WT suffered significantly damage after chilling stress for 12 h. Lipid droplet appeared in some cells of WT. In contrast, the membrane in transgenic plant still could kept intact structure. There were some communications with the other cells. 4. Discussion Low temperature is an abiotic stress known to induce changes in membrane lipid composition. Most studies that had proposed the chilling tolerance of the TAs level had focused on events in plastid omega-3 fatty acid desaturase [18], rather than on event in the extraplastid omega-3 fatty acid desaturase [5]. ER-type omega-3 fatty acid desaturase catalyzes the conversion of 18:2 to 18:3 in phospholipids which are the main constituents of extrachloroplast membranes. And TAs account for as much as 70% of the lipids in non-photosynthetic tissues such as roots [8]. So transgenic tomato plants with overxepression of LeFAD3 were used to investigate whether the high level TAs could improve chilling tolerance. A cDNA clone encoding endoplasmic reticulum membranebound omega-3 fatty acid desaturase was isolated from tomato. The analysis of the predicted amino acid sequence of LeFAD3 with endoplasmic reticulum FAD3 from other higher plants showed that LeFAD3 shared high sequence identity with FAD3 in other plants and clearly identified the LeFAD3 cDNA as encoding a member of endoplasmic reticulum FAD3. The deduced amino acid sequence alignment of FAD3 further revealed that the amino acid sequences had four conserved motifs (Fig. 1) and the three of the latter were necessary for desaturase enzyme function and conserved with all the membrane bond fatty acid [27]. Since expression patterns of FAD3 varied greatly among plant species [12,31], it was necessary to examine the expression regulation of LeFAD3 in tomato. In this experiment, LeFAD3 constitutively expressed in roots, stems, leaves, petals and fruits of tomato, but the expression level was the highest in roots compared with the lowest in leaves (Fig. 3). The highest level of LeFAD3 transcripts was observed in leaves and roots on exposure to 4  C for 6 h (Fig. 4A,B). However, the expression of LeFAD3 at 40  C was markedly inhibited and hardly detected at 6 h. The strong positive signals of RNA gel and western blot in transgenic plants compared with WT plants showed that sense LeFAD3 had been introduced into tomato genome and the gene had expressed both at RNA and protein level (Fig. 5). The signals of RNA and protein in T1-6 were stronger than those in T1-16 and T1-20. Previous studies had shown a positive correlation between the 18:3 levels and the mRNA level for ER-type omega-3 fatty acid desaturase. The 18:3 levels increased significantly in transgenic tobacco plants in which the tobacco ER-type omega-3 fatty acid desaturase gene (NtFAD3) was overexpressed [7]. Similar results were obtained with transgenic rice plants overexpressing NtFAD3, which accumulated high level of TAs in the extraplastid lipids of leaf and root tissues [28]. The up- and down-regulation of transcript level in LeFAD3 changed the 18:3 content in tomato (Table 2, 3). Overexpressing LeFAD3 produced higher amounts of 18:3 compared with WT and accompanied by a concomitant decrease in the level of 18:2 (Table 2). The increase in the level of 18:3 of total fatty acids was more pronounced in roots than leaves and was more in T1-6 than T1-16 and T1-20 (Table 2). The level of 18:3 in each lipid were obviously increased in other lipids including MGDG, DGDG, SQDG and PG especially in PC and PE, which indicated the fact that 18:3 could be synthesized on ER-type omega-3 fatty acid desaturases (eukaryotic pathway) and lipids could transfer between the compartments [11,32]. This observation suggested that overexpression LeFAD3 could markedly affect the 18:3 contents of all the lipids of leaves and

1111

roots and the ER-type omega-3 desaturase was major in contributing to the non-photosynthetic root tissues [26]. Increased contents of 18:3 were found to accompany cold acclimation in many plants, and a positive relationship was observed between a higher degree of fatty acid desaturation and cold tolerance [39]. In order to investigate the relationship between TAs with overexpression of LeFAD3 and the capacity of chilling tolerance, first, growth analysis of WT and the transgenic lines under chilling stress was studied. The fresh weight of the aerial parts of the T1-6, T1-16 and T1-20 was higher than that of WT at low temperature. This indicated that the levels of TAs were important for growth of tomato at low temperature, and T1-6 with highest 18:3 had lowest chilling injury (Fig. 6). Second, because the increase in the unsaturation degree of fatty acids was supposed to be a major factor that influences membrane fluidity [24], the membrane was thought to be the primary site of injury during chilling stress and the most commonly used method for assaying chilling tolerance tests the level of membrane damage by quantifying electrolyte leakage from chilling injured tissues. Results in our experiment showed that relative electric conductivity of leaves and roots in transgenic plants increased lower than that in WT, which showed that the damage of membrane was more serious in WT than that in transgenic lines (Fig. 7A,B). Furthermore, the membrane system ultrastructure of chloroplast in leaf cell and all the subcellular organelles in root tips of transgenic plants under chilling stress kept more intact than those of WT compared to an untreated control (Fig. 8). The higher unsaturated fatty acid content of total lipids in transgenic plants (Tables 2 and 3) alleviated membrane damage during chilling stress. Third, the thylakoid membranes were the site of oxygenic electron transport. It was frequently happened to cause photoinhibition under chilling stress. The crucial events of photoinhibition of PSII were the turnover of protein D1 in the reaction center and the replacement of the degraded D1 protein with newly synthesized D1 protein for the recovery [38], otherwise the plant suffered serious damage under excess photon energy. The maximal photochemical efficiency of PSII and the O2 evolution rate in WT decreased more than those in transgenic plants, which implied that photoinhibition in WT plants was more serious than that in transgenic tomato under chilling stress (Fig. 7E,F). Increased TAs in photosynthetic membrane were required to maintain chloroplast function at low temperature [24] and could help the renewal of the degraded D1 protein at low temperatures, resulting in the alleviation of photoinhibition [2,38]. At last, the chilling stress decreased respiration rate in WT, but had a less influence on root and leaf in the transgenic plants (Fig. 7C,D). There might be the reason to explain the accumulation of TAs in polar lipids contributing to the activity of membrane-bound proteins such as complexes involved in mitochondrial respiratory function. Then the increased content of 18:3 in extrachloroplastic lipids played an important role in maintaining the mitochondria respiration [17] and be beneficial to mitochondria electron transfer under chilling stress [6]. In conclusion, we demonstrated that the expression of LeFAD3 was induced by low temperature. Overexpression of LeFAD3 increased level of 18:3 markedly accompanied by a concomitant decrease in the level of 18:2 under chilling stress. The transgenic tomato with high 18:3 level alleviated membrane damage and had a higher chilling tolerance in transgenic plants than WT. Acknowledgements This research was supported by the State Key Basic Research and Development Plan of China (2009CB118500), the Natural Science Foundation of China (30871458), Program for Changjiang Scholars and Innovative Research Team in University (Grant IRT0635) and the Natural Science Foundation of Shandong province (Y2007D50).

1112

C. Yu et al. / Plant Physiology and Biochemistry 47 (2009) 1102–1112

References [1] V. Arondel, B. Lemieux, I. Hwang, S. Gibson, H.M. Goodman, C. Somerville, Map-based cloning of a gene controlling omega-3 fatty acid desaturation in Arabidopsis, Science 258 (1992) 1353–1355. [2] T. Ariizumi, S. Kishitani, R. Inatsugi, I. Nishida, N. Murata, K. Toriyama, An increase in unsaturation of fatty acids in phosphatidylglycerol from leaves improves the rates of photosynthesis and growth at low temperatures in transgenic rice seedlings, Plant Cell Physiol. 43 (2002) 751–758. [3] Z.Q. Chen, C.H. Xu, M.J. Chen, L. Xu, K.F. Wang, S.Q. Lin, T.Y. Kuang, Effect of chilling acclimation on thylakoid membrane protein of wheat, Acta Bot. Sin. 36 (1994) 423–429. [4] R. Collados, V. Andreu, R. Picorel, M. Alfonso, A light-sensitive mechanism differently regulates transcription and transcript stability of omega-3 fattyacid desaturases (FAD3, FAD7 and FAD8) in soybean photosynthetic cell suspensions, FEBS Lett. 580 (2006) 4934–4940. [5] J. Davy de Virville, C. Cantrel, A.L. Bousquet, M. Hoffelt, A.M. Tenreiro, V. Vaz Pinto, J.D. Arrabaça, O. Caiveau, F. Moreau, A. Zachowski, Homeoviscous and functional adaptations of mitochondrial membranes to growth temperature in soybean seedlings, Plant Cell Environ. 25 (2002) 1289–1297. [6] J. Davy de Virville, F. Cochet, G. Tasseva, F. Moreau, A. Zachowski, Changes in electron transport pathways in endoplasmic reticulum of rapeseed in response to cold, Planta 228 (2008) 875–882. [7] T. Hamada, H. Kodama, K. Takeshita, H. Utsumi, K. Iba, Characterization of transgenic tobacco with an increased a-linolenic acid level, Plant Physiol. 118 (1998) 591–598. [8] J.L. Harwood, Plant acyl lipids: structure, distribution and analysis, in: P.K. Stumpf, E.E. Conn (Eds.), The Biochemistry of Plants, vol. 4, Academic Press, New York, 1980, pp. 1–55. [9] M. Holsters, D. De Waele, A. Depicker, Transfection and transformation of Agrobacterium tumefaciens, Mol. Gen. Genet. 163 (1978) 181–187. [10] R.B. Horsch, J.E. Fry, N.L. Hoffmann, D. Eichholtz, S.G. Rogers, R.T. Fraley, A simple and general method for transferring gene into plants, Science 227 (1985) 1229–1231. [11] G. Horiguchi, H. Iwakawa, H. Kodama, N. Kawakami, M. Nishimura, K. Iba, Expression of a gene for plastid u-3 fatty acid desaturase and changes in lipid and fatty acid compositions in light-and dark-grown wheat leaves, Physiol. Plant 96 (1996) 275–283. [12] G. Horiguchi, T. Fuse, N. Kawakami, H. Kodama, K. Iba, Temperature-dependent translational regulation of the ER omega-3 fatty acid desaturase gene in wheat root tips, Plant J. 24 (2000) 805–813. [13] U.K. Laemmli, Cleavage of structural proteins during the assembly of head of bacteriophoge T4, Nature 227 (1970) 680–685. [14] B. Lemieux, M. Miquel, C. Somerville, J. Browse, Mutants of Arabidopsis with alterations in seed lipid fatty acid composition, Theor. Appl. Genet. 80 (1990) 234–240. [15] X.G. Li, W. Duan, Q.W. Meng, Q. Zou, S.J. Zhao, The function of chloroplastic NAD(P)H dehydrogenase in tobacco during chilling stress under low irradiance, Plant Cell Physiol. 45 (2004) 103–108. [16] J.K. Lyons, E.A. Chapman, Membrane phase changes in chilling-sensitive Vigna radiata and their significance to growth, Aust. J. Plant Physiol. 3 (1976) 291–299. [17] A.R. Matos, C. Hourton-Cabassa, D. Cicek, N. Reze´, J.D. Arrabaça, A. Zachowski, F. Moreau, Alternative oxidase involvement in cold stress response of Arabidopsis thaliana fad2 and FAD3 þ cell suspensions altered in membrane lipid composition, Plant Cell Physiol. 48 (2007) 856–865. [18] O. Matsuda, H. Sakamoto, T. Hashimoto, K. Iba, A temperature-sensitive mechanism that regulates post-translational stability of a plastidial omega-3 fatty acid desaturase (FAD8) in Arabidopsis leaf tissues, J. Biol. Chem. 280 (2005) 3597–3604.

[19] O. Matsuda, C. Watanabea´, K. Iba, Hormonal regulation of tissue-specific ectopic expression of an Arabidopsis endoplasmic reticulum-type omega-3 fatty acid desaturase (FAD3) gene, Planta 213 (2001) 833–840. [20] M. Miquel, J. Browse, Arabidopsis mutants deficient in polyunsaturated fatty acid synthesis. Biochemical and genetic characterization of a plant oleoylphosphatidylcholine desaturase, J. Biol. Chem. 267 (1992) 1502–1509. [21] Y. Murakami, M. Tsuyama, Y. Kobayashi, H. Kodama, K. Iba, Trienoic fatty acids and plant tolerance of high temperature, Science 287 (2000) 476–479. [22] J. Ohlrogge, J. Browse, Lipid biosynthesis, Plant Cell 7 (1995) 957–970. [23] I.V. Orlova, T.S. Serebriiskaya, V. Popov, N. Merkulova, A.M. Nosov, T.I. Trunova, V.D. Tsydendambaev, D.A. Los, Transformation of tobacco with a gene for the thermophylic acyl-lipid desaturase enhances the chilling tolerance of plants, Plant Cell Physiol. 44 (2003) 447–450. [24] J.M. Routaboul, S.F. Fischer, J. Browse, Trienoic fatry acids are required to maintain chloroplast function at low temperature, Plant Physiol. 124 (2000) 1697–1705. [25] J. Sambrook, E.F. Fritsch, T. Maniatis, Molecular Cloning, a Laboratory Manual, vol. 1, Cold Spring Harbor Laboratory Press, NY,, 1989, pp. 2,3. [26] S. Shah, Z.G. Xin, J. Browse, Overexpression of the FAD3 desaturase gene in a mutant of Arabidopsis, Plant Physiol. 114 (1997) 1533–1539. [27] J. Shanklin, E.B. Cahoon, Desaturation and related modification of fatty acids, Annu. Rev. Plant Physiol. Plant Mol. Biol. 49 (1998) 611–641. [28] T. Shimada, Y. Wakita, M. Otani, K. Iba, Modification of fatty acid composition in rice plants by transformation with a tobacco ER-type u-3 fatty acid desaturase gene (NtFAD3), Plant Biotechnol. 17 (2000) 43–48. [29] P.A. Siegenthaler, W. Eichenberger, Structure, Function and Metabolism of Plant Lipids, Elsevier Science Publishers, Amsterdam, 1984. [30] C. Somerville, J. Browse, Dissecting desaturation: plants prove advantageous, Trends Cell Biol. 6 (1996) 148–153. [31] G. Tasseva, J. Davy de Virville, C. Cantrel, F. Moreau, A. Zachowski, Changes in the endoplasmic reticulum lipid properties in response to low temperature in Brassica napus, Plant Physiol. Biochem. 42 (2004) 811–822. [32] G. Tasseva, L. Richard, A. Zachowski, Regulation of phosphatidylcholine biosynthesis under salt stress involves choline kinases in Arabidopsis thaliana, FEBS Lett. 566 (2004) 115–120. [33] O. Van Kooten, J.P.H. Snel, The use of chlorophyll fluorescence nomenclature in plant stress physiology, Photosynth. Res. 25 (1990) 147–150. [34] P. Vrinten, Z.H.Y. Hu, M.A. Munchinsky, G. Rowland, X. Qiu, Two FAD3 desaturase genes control the level of linolenic acid in flax seed, Plant Physiol. 39 (2005) 79–87. [35] D. Walker, The use of O2 electrode and fluorescence probes in simple measurements of photosynthesis, Robert Hill Institute, University of Sheffield, UK, 1990. [36] Y.N. Xu, P.A. Siegenthaler, Low temperature treatments induce an increase in the relative content of both linolenic and l3-trans-hexadecenoic acids in thylakoid membrane phosphatidylglycerol of squash cotyledons, Plant Cell Physiol. 38 (1997) 611–618. [37] K.T. Yamamoto, H. Mori, H. Imaseki, Novel mRNA sequences induced by indole-3-acetic acid in sections of elongating hypocotyls of mung bean (Vigna radiata), Plant Cell Physiol. 33 (1992) 13–20. [38] L. Zhang, V. Paakkarinen, K.J. van Wijk, E.M. Aro, Biogenesis of the chloroplastencoded D1 protein: regulation of translation, elongation, insertion, and assembly into photosystem II, Plant Cell 12 (2000) 1769–1782. [39] M. Zhang, R. Barg, M. Yin, G.D. Yardena, L.F. Alicia, Y. Salts, S. Shabtai, B.H. Gozal, Modulated fatty acid desaturation via overexpression of two distinct u-3 desaturases differentially alters tolerance to various abiotic stresses in transgenic tobacco cells and plants, Plant J. 44 (2005) 361–371. [40] J. Zou, G.D. Abrams, D.L. Barton, D.C. Taylor, M.K. Pomeroy, S.R. Abrams, Induction of lipid and oleoisin biosynthesis by(þ)- abscisic and its metabolites in microspore-derived embroyos of Brassica napus L.cv Reston, Plant Physiol. 108 (1995) 563–571.