Oleate accumulation, induced by silencing of microsomal omega-6 desaturase, declines with leaf expansion in transgenic tobacco

Transcription

1 Journal of Plant Physiology 164 (2007) Oleate accumulation, induced by silencing of microsomal omega-6 desaturase, declines with leaf expansion in transgenic tobacco Mingfeng Yang, Yinong Xu Key Laboratory of Photosynthesis and Environmental Molecular Physiology, Institute of Botany, Chinese Academy of Sciences, Nanxincun 20, Xiangshan, Beijing , China Received 8 July 2005; accepted 2 November 2005 KEYWORDS Expansion; Lipid; Microsomal omega-6 desaturase; RNA interference; Tobacco Summary All higher plants contain at least one microsomal omega-6 desaturase (FAD2), which inserts a double bond between the carbons 12 and 13 of monounsaturated oleic acid to generate polyunsaturated linoleic acid and controls most of the polyunsaturated lipid synthesis in plant cells. RNA interference can be used to silence endogenous genes by effective degradation of target transcripts. To investigate developmentrelated silencing of the FAD2, fatty acid composition was analyzed in the context of leaf expansion in FAD2-silenced tobacco lines obtained by RNA interference technology. We observed that the increased oleate level in unexpanded leaves due to FAD2-silencing receded significantly in fully expanded leaves. The mechanism involved in this interesting phenomenon was investigated by analyses of individual lipid proportion, fatty acid composition of individual lipids, and FAD2 transcript level in the transgenic leaves at different expansion stages. Data revealed that the expansion-related FAD2-silencing effect was not due to rebound of FAD2 transcript, but rather probably due to chloroplast development with leaf expansion. & 2005 Elsevier GmbH. All rights reserved. Abbreviations: 16:0, palmitic acid; 18:0, stearic acid; 18:1, oleic acid; 18:2, linoleic acid; 18:3, linolenic acid; DAF, days after flower; DGDG, digalactosyldiacylglycerol; EV, empty vector; FAD2, microsomal omega-6 desaturase; FAD6, plastidic omega-6 desaturase; MGDG, monogalactosyldiacylglycerol; PC, phosphatidylcholine; PE, phosphatidylethanolamine; PG, phosphatidylglycerol; SQDG, sulfoquinovosyldiacylglycerol; WT, wild type Corresponding author. Tel.: ; fax: address: ymf@ibcas.ac.cn (Y. Xu). Introduction Vegetable oils are increasingly important economically because they are renewable resources of highly reduced carbon and widely used in diets and industrial applications. One of the major factors influencing the quality of vegetable oils is the content of polyunsaturated fatty acids, and edible /$ - see front matter & 2005 Elsevier GmbH. All rights reserved. doi: /j.jplph

2 24 oils rich in oleic acid were suitable for human and animal consumption for their improved oil stability, flavor and nutrition (Heppard et al., 1996). Two distinct pathways, namely prokaryotic pathway and eukaryotic pathway, cooperate in plant cells for the biosynthesis of glycerolipids and the associated production of polyunsaturated fatty acids (Roughan and Slack, 1982; Browse and Somerville, 1991). In plants, two pathways coordinate in glycerolipid synthesis and the balance of fluxes through these pathways may be altered to ameliorate the effects of mutations that block steps in one of the pathways (Browse and Somerville, 1991; Somerville and Browse, 1996). At least for the endoplasmic reticulum and the plastid, lipid traffic between the membranes is bi-directional and most of the mutations affect the compositions of both chloroplast and extrachloroplast membranes even though the enzymes are located in one compartment or the other (Somerville and Browse, 1996). It is necessary to further our insight into desaturases involved in lipid synthesis for cultivation crops with oils containing suitable fatty acids. However, isolation and characterization of most of the fatty acid desaturases have proved to be a challenge due to their membrane-bound nature (Browse and Somerville, 1991). Fortunately, many genes of lipid synthesis have been cloned in a wide variety of plant species among which the omega-6 desaturase gene is of particular interest, for it is an enzyme that places the second double bond in fatty acids and catalyzes the first step of polyunsaturated fatty acid biosynthesis. In plant cells, one omega-6 desaturase is FAD6, located in plastid, and involved in polyunsaturated fatty acid synthesis through prokaryotic pathway; the other is FAD2, located in endoplasmic reticulum, and involved in polyunsaturated fatty acid synthesis through eukaryotic pathway (Browse and Somerville, 1991). All higher plants contain at least one FAD2 that insert a double bond between the carbons 12 and 13 of monounsaturated oleic acid to generate polyunsaturated linoleic acid, and the FAD2-catalyzed pathway is likely the primary route for production of polyunsaturated lipids (Shanklin and Cahoon, 1998). However, many other details about FAD2 remain elusive, since FAD2 is an endoplasmic reticulum membrane-bound desaturase (Dyer and Mullen, 2001; McCartney et al., 2004). RNA interference technology has been successfully used to silence FAD2 in several plant species, thereby modifying the unsaturated level of seed oil (Wesley et al., 2001; Liu et al., 2002; Stoutjesdijk et al., 2002). There is yet little research on developmentrelated silencing of the FAD2 gene and associated fatty acid composition in FAD2-silenced plant cells. In addition, though both prokaryotic pathway and eukaryotic pathway leading to the synthesis of unsaturated fatty acids are known, the cooperation and regulation of these pathways are poorly understood. Investigation of the developmentrelated silencing of FAD2 may facilitate the application of RNA interference technology in manipulation of the polyunsaturated fatty acid composition of plant membranes, and may further our insight into the cooperation of both the lipid synthesis pathways. In this report, we used the FAD2-silenced tobacco plants obtained by RNA interference technology to investigate the relation of leaf expansion to the silencing effect of FAD2 gene. To our knowledge, this is the first report on the development-related effect of FAD2 gene silencing in transgenic plant. Materials and methods Gene-silencing construct A FAD2-silencing construct was generated using the pkannibal vector, and was introduced into Agrobacterium tumefaciens for transformation of tobacco as described (Wesley et al., 2001; Yang et al., 2006). A 452 bp partial coding region, corresponding to nucleotides of the tobacco FAD2 gene (GenBank: AY660024), was used in the FAD2-silencing construct. The construct without insertion of the FAD2 segment was designated as empty vector (EV). Plant materials M. Yang, Y. Xu Identification of primary transformants and confirmation of the integration of the transgene into tobacco (Nicotiana tabacum, cv. Wisconsin 38) genome were conducted as described in our parallel work (Yang et al., 2006). The transgenic line S18, herein referred as S1 for convenience, was used as materials in this paper. Transgenic tobacco seedlings were grown in a growth chamber under conditions of 16 h 80 mmol m 2 s 1 light/8 h dark at 24 1C. The 2nd or 3rd leaf of 25-day-old seedlings was used to extract total lipids or to extract RNA for RT-PCR. Then these seedlings were transferred to soil and kept in the chamber under the same conditions. When the tobacco plants were 2 months old, the 3rd, 5th, 8th, 11th and 13th leaves numbered starting with the apical bud (position 1) were harvested and stored at 70 1C for analysis of fatty acid composition and RNA gel blot hybridization. The 13th leaf represented fully expanded

3 Leaf expansion reverts effect of FAD2-silencing 25 leaf. After self-pollination, the seeds were harvested about 7, 14, 21 and 28 days after flower (DAF) and stored at 70 1C for analysis of fatty acid composition. RT-PCR analysis Total RNA was prepared from the FAD2-silenced and wild-type tobacco leaves using Trizol reagent (Invitrogen, Carlsbad, CA) following the manufacturer s instructions, and was treated with RNasefree DNase I (Promega, Madison, WI, USA). The quality and concentration of RNA preparations were accurately determined with a UV visible spectrophotometer (model Cary 50BIO; Varian, Walnut Creek, CA). Reverse transcription was carried out using reverse transcriptase (Invitrogen, Carlsbad, CA) according to the manufacturer s instructions. The primers for tobacco 452 bp FAD2 fragment were sense primer 5 0 -CGTCGCCACCATTC- CAACAC-3 0 and antisense primer 5 0 -CCCCTAAGC- CAATCCCACTC-3 0. The procedure of PCR is 5 min at 94 1C, 26 cycles for 30 s at 94 1C, 45 s at 54 1C, 1 min at 72 1C, and extension 10 min at 72 1C. The abundance of RT-PCR products was normalized to the level of a constitutively expressed gene, tobacco actin Tac9 (GenBank: X69885). The PCR for Tac9 fragment was carried out using the same samples and procedure as above. Primers for tobacco actin Tac9 were sense primer 5 0 -CCCTCC CACATGCTATTCT-3 0 and antisense primer 5 0 -AGAG CCTCCAATCCAGACA-3 0. RNA gel blot analysis Fifteen micrograms of total RNA was separated on a denatured formaldehyde gel and transferred to Hybond N + nylon membrane according to Sambrook et al. (1989). The probe used for FAD2 mrna was a [ 32 P]dCTP-labeled FAD2 452-bp fragment obtained by PCR corresponding to nucleotides of the tobacco FAD2 gene. To confirm that lanes were equally loaded, the nylon membrane was stripped by washing in 10 mm Tris-Cl (ph 7.4), 0.2% (w/v) SDS at 75 1C for 1 h, then re-hybridized with an a- 32 P dctp-labeled Tac9 probe. Prehybridization and hybridization for both the FAD2 and Tac9 were carried out using Hybridization cocktails I and III (Sangon, Shanghai, CN) following the manufacturer s instructions. Fatty acid analysis Lipids were extracted from different tobacco tissues according to the method of Bligh and Dyer (1959). The individual lipid separation and the fatty acid analysis were carried out according to the method of Xu et al. (2003). Unless otherwise stated, extrachloroplastic lipids included phosphatidylcholine (PC) and phosphatidylethanolamine (PE); chloroplastic lipids included sulfoquinovosyldiacylglycerol (SQDG), phosphatidylglycerol (PG), digalactosyldiacylglycerol (DGDG) and monogalactosyldiacylglycerol (MGDG). Results Identification of primary transformants Tobacco was transformed with a gene-silencing construct consisting of a fragment of FAD2 cdna in inverted repeat configuration driven by a CaMV35S promoter essentially as described (Wesley et al., 2001; Yang et al., 2006). As expected, many transgenic lines showing a significant increase of oleic acid (18:1) level with concomitant reduction of linoleic (18:2) and linolenic (18:3) were recovered by analyses of fatty acid composition of total leaf lipids from primary transformant and wild-type (WT) tobacco seedlings. Three lines (S1, S2, and S3) with high levels of 18:1 were chosen to be representatives to examine the FAD2 mrna abundance by RT-PCR (Fig. 1). The results showed that, in these high-oleate lines, FAD2 transcript dropped to a very low level in contrast to WT and EV transformed lines. It suggested that the markedly elevated 18:1 content in transgenic plant leaves resulted from the low level of FAD2 transcript in transgenic lines induced by the silencing construct. Variation of 18:1 level with leaf expansion The level of 18:1 was about 15% in total leaf lipids of nine transgenic lines in contrast to about Figure 1. RT-PCR analysis for FAD2 transcript level. Ethidium bromide-stained gel of products obtained after RT-PCR analysis of mrna from wild type (WT), empty vector transformed line (EV) and different primary transformants (S1, S2, and S3) are shown (A). Control reactions were carried out using primers for tobacco actin Tac9 (GenBank, X69885) mrna from the same samples (B).

4 26 M. Yang, Y. Xu 2% in WT (Fig. 2). These lines with a marked increase of 18:1 were transplanted into a growth chamber and maintained under controlled conditions. The fatty acid composition of total lipids from the fully expanded leaves was analyzed again when these lines were 2 months old, about 70 cm in height. Interestingly, the 18:1 content dropped to a level similar to WT, and some lines had no difference with WT. The fact that 18:1 content was high in the seedling leaves and low in the fully expanded leaves of transgenic lines implied that the change was due to sampling at different leaf expansion stages. To test this, the 3rd, 5th, 8th, 11th and 13th leaves of S1 and WT plants were sampled, representing the different stages of leaf expansion ranging from unexpanded to fully expanded. Total lipids were extracted from these leaves to examine the detailed expansion-related FAD2-silencing effect. In contrast to WT, the 18:1 content in total leaf lipids declined swiftly with the expansion of leaves on transgenic line S1, and this decline was accompanied by a significant increase of 18:2 plus 18:3 (Fig. 3). It suggested that in the expansion of leaves, 18:1 was converted into 18:2 and 18:3 smoothly, leading to the decrease of 18:1 accumulation in the FAD2-silenced lines. 18:1 (%) Lines Figure 2. Content of oleic acid (18:1) in total lipids of tobacco leaves. Total lipids were extracted from leaves of wild-type seedlings (open circles), fully expanded leaves from 2-month-old wild-type plants (closed circles), leaves from transgenic seedlings (open triangles) and fully expanded leaves from 2-month-old transgenic plants (closed triangles). Content of 18:1 was calculated as the percentage of it represented of the total measured fatty acids. Percent of total fatty acids :2+18: Leaf position 18:1 Figure 3. Contents of 18C unsaturated fatty acids [18:1 and (18:2+18:3)] in total lipids of tobacco leaves. Total lipids were extracted from leaves at different positions on wild-type (closed circles) and S1 plants (open circles). Leaves are numbered starting with the apical bud (position 1). Contents of 18:1 and 18:2 plus 18:3 were calculated as the percentage they represented of the total measured fatty acids. Each value is the mean of three individual replicates (7SD). Variation of 18:1 level in individual leaf lipids To examine the variation of 18:1 level of individual lipids in the context of leaf expansion, we analyzed the fatty acid composition of individual lipids extracted from the unexpanded and fully expanded leaves (Fig. 4). In the unexpanded leaves (the 3rd ones) of S1 plants, both PC and PE, the main individual lipids of extrachloroplast membranes, showed a marked increase in the level of 18:l and a concomitant decrease in the amount of 18 carbon polyunsaturated fatty acids, especially a marked decrease in 18:2. In the unexpanded leaves of line S1, the proportion of 18:1 in PC and PE increased to 55.5% and 28.3%, respectively, which was more than 5-fold over WT levels. Those lipids located predominantly in chloroplast (SQDG, PG, DGDG and MGDG) also showed significant

5 Leaf expansion reverts effect of FAD2-silencing 27 Percent of total fatty acids PC PE SQDG PG DGDG MGDG Individual lipids Figure 4. Contents of 18C unsaturated fatty acids (18:1, 18:2 and 18:3) in individual lipids of tobacco leaves. Unexpanded (1, 3) and fully expanded (2, 4) leaves of wild-type (1, 2) and S1 (3, 4) plants are shown. Black bars, 18:1; white bars, 18:2; striped bars, 18:3. Contents of 18:1, 18:2 and 18:3 were calculated as the percentage that each fatty acid represented of the total measured fatty acids in each individual lipid. Each value is the mean of three individual replicates, and the variation of individual analysis is insignificant (o2%). Except for DGDG and MGDG in wild type, the differences between all values are significant ðpo0:05þ. Percent of total polar lipids WT unexpanded S1 unexpanded WT expanded S1 expanded PC PE SQDG PG DGDG MGDG Individual lipids Figure 5. Proportions of individual lipids in total polar lipids of tobacco leaves. Unexpanded and fully expanded leaves of wild type (WT) and transgenic plants (S1) are shown. Individual lipid proportions were calculated as the percentage that each individual lipid represented of the total measured individual lipids. Each value is the mean of three individual replicates (7SD). increases in their 18:l levels, but much lower than that of PC and PE. The 18:1 content in most individual lipids changed significantly with leaf expansion for both S1 and WT plants (Fig. 4). In line S1, the 18:1 levels of extrachloroplastic lipids in unexpanded leaves were much higher than those in WT, and then 18:1 decreased violently almost to WT levels when the leaves fully expanded. By contrast, the 18:1 content in all individual lipids from leaves of WT decreased slightly or insignificantly with leaf expansion. It clearly showed that the 18:1 accumulation in all the individual lipids of transgenic leaves, especially in PC and PE, decreased with leaf expansion. Variation of extrachloroplastic lipid proportion in leaves To evaluate the contribution of 18:1 variation in PC and PE to the 18:1 level of total leaf lipids, the possible unstable ratio of extrachloroplastic lipids to chloroplastic lipids during the leaf development must be taken into account. The percentage of six individual lipids extracted from the unexpanded and fully expanded leaves, was therefore analyzed to roughly estimate the variation of extrachloroplastic lipid levels with leaf development. As shown in Fig. 5, the percentage of PC, PE, MGDG and DGDG varied drastically with leaf expansion, but no difference was found between WT and S1 plants, either in the unexpanded or the fully expanded leaves. The ratio of extrachloroplastic lipids to chloroplastic lipids was in the Figure 6. RNA gel blot analysis of FAD2 transcript level of tobacco leaves. Fifteen micrograms total RNA per lane was resolved on an agarose gel and hybridized with 32 P-labeled FAD2 probe (A). The blots were stripped and reprobed with tobacco actin Tac9 gene as loading control (B). Lanes from left to right are fully expanded leaves of wild type, fully expanded leaves of line S1, unexpanded leaves of wild type and unexpanded leaves of line S1. unexpanded leaves of line S1, and this ratio changed into eight to 92 in the fully expanded leaves, showing a 6.5-fold decrease of the extrachloroplastic lipid levels. FAD2 transcript abundance in leaves In order to investigate if the significant decrease of 18:1 level in PC and PE was due to the increase of FAD2 transcript in the expansion of transgenic leaves, RNA extracted from the unexpanded or the fully expanded leaves of S1 and WT plants was used in RNA gel blot analysis to probe the FAD2 transcript level (Fig. 6). The results showed that

6 28 M. Yang, Y. Xu Table 1. Fatty acid composition in selfed seeds with different DAF borne on S1 DAF Fatty acid (%) 16:0 18:0 18: WT S WT S WT S WT S Content of each fatty acid was calculated as the percentage of it that represented the total measured fatty acids. Each value is the mean of three individual replicates (7SD). the FAD2 transcript level was very low and similar in both the unexpanded and fully expanded leaves of line S1. This indicated that the decrease in 18:1 level of leaf lipids was not caused by the rebound of FAD2 transcript in the fully expanded leaves of transgenic lines. Fatty acid composition of total seed lipids The fatty acid composition of the bulk seeds sampled at different DAF from the self-fertilized S1 line was analyzed to roughly estimate the silencing effect on seeds with different ages (Table 1). The data showed that the 18:1 level in total seed lipids declined quickly with attendant increase of 18:2 at early development stage when the seeds were in pale green color. However, at later development stages of seeds, i.e. after 14 DAF, no change of 18:l level was found in the seeds. Discussion We applied RNA interference technology to silence tobacco FAD2 and obtained FAD2-silenced seedlings with high 18:1 level in leaves. The RT-PCR and RNA gel blot analyses proved that the greatly elevated 18:1 content in unexpanded leaves of the transgenic plants was due to the reduction of FAD2 transcript. Unexpectedly, the elevated 18:1 content receded to a level similar to WT in fully expanded leaves of the transgenic plants and this was not induced by rebound of FAD2 transcript. In tobacco plants, most chloroplasts are in proplastid phase or in early developmental phase in the unexpanded leaves, and they enter into the later stage of differentiation or maturation in the larger leaves (Ehara and Misawa, 1975). Based on these observations, we assumed that the development of chloroplasts might be involved in the decrease of 18:1 accumulation in the FAD2-silenced tobacco throughout leaf expansion. The fact that extrachloroplastic lipids were more than chloroplastic lipids in the unexpanded leaves of line S1, combined with the observation that the highest 18:1 level was found in the extrachloroplastic lipids, suggested that the drastic increase of 18:1 in extrachloroplastic lipids contributed the most to the high-oleate phenotype of line S1. However, in the fully expanded leaves of line S1, the level of extrachloroplastic lipids dropped to about 8% in total individual lipids analyzed. This drop was caused by the significant increase of chloroplastic lipids, owing to the development of chloroplasts and the increase of its amount in plant cells. Thus, the relative decrease of extrachloroplastic lipids should have some impact on the high-oleate phenotype of the fully expanded leaves of the transgenic tobacco. However, previous work suggested that the two pathways of lipid synthesis contribute almost equally to the production of chloroplast membrane lipids in wild-type Arabidopsis (Browse et al., 1986), and the FAD2 enzyme is quantitatively more important in leaf cells in which FAD6 also operates (Browse and Somerville, 1991; Miquel and Browse, 1992). Since some chloroplastic lipids were synthesized by eukaryotic pathway, the alteration of lipids ratio might contribute only partially to the decrease of 18:1 levels in the fully expanded leaves of the transgenic tobacco. It appears that significant interplay occurs between the eukaryotic and prokaryotic pathways through a reversible exchange of fatty acids

7 Leaf expansion reverts effect of FAD2-silencing 29 between plastidic and endoplasmic reticulum membranes (Miquel and Browse, 1992, 1994; Browse et al., 1993; Somerville and Browse 1996), enabling each pathway to partially compensate for mutations in the other pathway to maintain adequate levels of polyunsaturated fatty acids. It was suggested that, in Arabidopsis fad2 mutant, 18:1 containing lipids that re-enter the chloroplast could be converted to 18:2 by FAD6 (Miquel and Browse, 1992). In the FAD2-silenced tobacco, the decrease of 18:1 level in fully expanded leaves was not due to the rebound of FAD2 transcript, as suggested by RNA gel blot analysis. Based on these observations, we propose that the lipid exchange may also take place in the FAD2-silenced tobacco and the decrease of 18:1 associated with increase of 18:2 and 18:3 in fully expanded leaves of line S1 would be a consequence of such an exchange. That is to say, the 18:1 containing lipids may be transported into chloroplasts and desaturated by FAD6. Thus, the development of chloroplasts and the increase of its amount in plant cells throughout the leaf expansion may result in more 18:1 containing lipids entering the chloroplast and being desaturated by FAD6. The resulted polyunsaturated fatty acids either remain in the chloroplasts or are exported and subsequently incorporated in the extrachloroplastic lipids. Because of the change of extrachloroplastic lipids/chloroplastic lipids ratio and the interplay between eukaryotic and prokaryotic fatty acid biosynthesis pathways mentioned above, we may expect that, at the beginning of chloroplast biogenesis when the prokaryotic pathway could not produce enough lipids for its own use, it is essential for chloroplasts to use lipids produced by the eukaryotic pathway; when the leaves are fully expanded, more polyunsaturated fatty acids are desaturated by FAD6 and exported to the extrachloroplast, since the rapid synthesis of chloroplast membrane lipids has ceased. It may help to explain why the 18:1 level of chloroplastic lipids increased significantly in the unexpanded leaves and why the 18:1 level of extrachloroplastic lipids decreased drastically in the fully expanded leaves (Fig. 4). The data from seeds (Table 1) indicated that the level of 18:1 decreased markedly in the pale green color seeds where some plastids existed, whereas no change of 18:l level was found in the 14 DAF seeds where plastids disappeared. FAD2 is responsible for more than 90% of the polyunsaturated fatty acid synthesis in nonphotosynthetic tissues (Miquel and Browse, 1994). In a seed, a nonphotosynthetic organ without mature chloroplasts, the chloroplastic desaturase may have not so much interference on the fatty acid composition as in leaf. This may be the reason why the FAD2-silenced tobacco plants kept the high-oleate character in their mature seeds. An unlikely, alternative explanation for the development-related silencing effect is that there exists an isoenzyme of FAD2 in tobacco, and the sequence of this putative isoenzyme should be sufficiently divergent to escape cross silencing by the FAD2-silencing construct used in this study. This enzyme might not be very active even if it exits, since the level of 18:1 increased very slowly throughout the leaf expansion procedure. Browse et al. (1993) reported that in Arabidopsis the 18:3 content of PC is relatively low in young leaves, and that the proportion of 18:3 in this species increases with plant age. The 18:3 level was very low in the small leaves of EV transformed tobacco but increases during leaf expansion (Kodama et al., 1995). The 18:3 level of total leaf lipids increased with leaf age in the FAD2-silenced lines and WT plants (data not shown), which demonstrate the increasing activity of omega-3 desaturase with leaf age. As a consequence, 18:2 was efficiently desaturated precluding from 18:2 accumulations, and such a procedure may facilitate the 18:1 desaturation. To sum up, though other possibilities exist, the decrease of 18:1 accumulation in fully expanded leaves of FAD2-silenced tobacco was at least partially, if not totally, owing to the development of chloroplasts accompanied by the relative decrease of extrachloroplastic lipids and the lipid exchange between the two pathways. Acknowledgments We are grateful to CSIRO Plant Industry in Australia for giving us the pkannibal vector. This work was supported by the National Sciences Foundation of China (Grant Nos and ) and the Chinese Academy of Sciences (Grant No. KSCX2-SW-328). References Bligh EG, Dyer WJ. A rapid method of total lipid extractions and purification. Can J Biochem Physiol 1959;37: Browse J, Somerville C. Glycerolipid synthesis: biochemistry and regulation. Annu Rev Plant Physiol Plant Mo1 Biol 1991;42: Browse J, Warwick N, Somerville CR, Slack CR. Fluxes through the prokaryotic and eukaryotic pathways of lipid synthesis in the 16:3 plant Arabidopsis thaliana. Biochem J 1986;235:25 31.

8 30 M. Yang, Y. Xu Browse J, McConn M, James D, Miquel M. Mutants of Arabidopsis deficient in the synthesis of a-linolenate. J Biol Chem 1993;268: Dyer JM, Mullen RT. Immunocytological localization of two plant fatty acid desaturases in the endoplasmic reticulum. FEBS Lett 2001;494:44 7. Ehara Y, Misawa T. Occurrence of abnormal chloroplasts in tobacco leaves infected systemically with the ordinary strain of cucumber mosaic virus. Phytopathol Z 1975;84: Heppard EP, Kinney AJ, Stecca KL. Developmental and growth temperature regulation of two different microsomal o-6 desaturase genes in soybeans. Plant Physiol 1996;110: Kodama H, Horiguchi G, Nishiuchi T, Nishimura M, Iba K. Fatty acid desaturation during chilling acclimation is one of the factors involved in conferring lowtemperature tolerance to young tobacco leaves. Plant Physiol 1995;107: Liu Q, Singh SP, Green AG. High-stearic and high-oleic cottonseed oils produced by hairpin RNA-mediated post-transcriptional gene silencing. Plant Physiol 2002;129: McCartney AW, Dyer JM, Dhanoa PK, Kim PK, Andrews DW, McNew JA, et al. Membrane-bound fatty acid desaturases are inserted co-translationally into the ER and contain different ER retrieval motifs at their carboxyl termini. Plant J 2004;37: Miquel MF, Browse JA. Arabidopsis mutants deficient in polyunsaturated fatty acid synthesis. J Biol Chem 1992;267: Miquel MF, Browse JA. High-oleate oilseeds fail to develop at low temperature. Plant Physiol 1994;106: Roughan PG, Slack CR. Cellular organization of glycerolipid metabolism. Annu Rev Plant Physiol 1982;33: Sambrook J, Fritsch EF, Maniatis T. Molecular cloning: a laboratory manual, 2nd ed. Plainview, New York: Cold Spring Harbor Laboratory Press; p Shanklin J, Cahoon EB. Desaturation and related modifications of fatty acids. Annu Rev Plant Physiol Plant Mol Biol 1998;49: Somerville C, Browse J. Dissecting desaturation: plants prove advantageous. Trends Cell Biol 1996;6: Stoutjesdijk P, Singh SP, Liu Q, Hurlstone C, Waterhouse P, Green A. hprna-mediated targeting of the Arabidopsis FAD2 gene gives highly efficient and stable silencing. Plant Physiol 2002;129: Wesley SV, Helliwell CA, Smith NA, Wang M, Rouse DT, Liu Q, et al. Construct design for efficient, effective and high-throughput gene silencing in plants. Plant J 2001;27: Xu YN, Wang ZN, Jiang GZ, Li LB, Kuang TY. Effect of various temperature on phosphatidylglycerol biosynthesis in thylakoid membrane. Physiol Plant 2003;118: Yang MF, Zheng GL, Zhang FY, Xu YN. FAD2 silencing has pleiotropic effect on polar lipids of leaves and varied effect in different organs of transgenic tobacco. Plant Sci 2006;170:170 7.