Production of Novel Oils in Plants Denis J Murphy

175

Production of novel oils in plants Denis J Murphy

We have now isolated the great majority of genes encoding unexpected ways. We have seen the isolation of some enzymes of storage oil biosynthesis in plants. In the past two potentially key genes that contribute both to the quantity years, particular progress has been made with acyltransferases, and quality of seed storage oils. There has also been an ketoacyl-acyl carrier protein synthetases and with desaturases increasing appreciation of the importance of fatty acids, and their relatives. In some cases, these enzymes have been re- not only as storage or structural components, but also act- engineered to create novel products. Nevertheless, the single or ing as, or giving rise to, important signalling molecules that multiple insertion of such transgenes into oil crops has not always regulate many aspects of plant development [1,2]. This led to the desired phenotype. We are only now beginning to illustrates the importance of ensuring that novel fatty acids appreciate some of the complexities of storage and membrane in transgenic oil crops are correctly targeted to the storage lipid formation, such as acyl group remodelling and the turnover oil and are hence unable to adversely affect membrane or of unusual fatty acids. This understanding will be vital for future signalling functions. The purpose of this article is to attempts at the rational engineering of transgenic oil crops. In review some of the recent progress in understanding the parallel with this, the domestication of plants already synthesising mechanism and regulation of storage oil formation in useful fatty acids should be considered as a real alternative to the plants, and how this may impact on its biotechnological transgenic approach to producing novel oil crops. manipulation.

Addresses Industrial and edible oils Brassica and Oilseeds Research Department, John Innes Centre, Before reviewing some of the recent technical develop- Norwich Research Park, Norwich NR4 7UH, UK; ments, it may be useful to consider what we are trying to e-mail: [email protected] achieve in modifying seed oils and why we are doing it. At Current Opinion in Biotechnology 1999, 10:175–180 present, over 80% of the 75 million tonnes of globally trad- ed seed oils are used for edible purposes, most notably in http://biomednet.com/elecref/0958166901000175 the production of cooking oils, margarines and processed © Elsevier Science Ltd ISSN 0958-1669 foods [3]. Global production of plant oils for industrial use Abbreviations (i.e. oleochemicals) is only about 15 million tonnes per year ACP acyl carrier protein with a value of about $400–800 per tonne. This contrasts DGAT diacylglycerol acyltransferase with the pre-tax price of refined petroleum products which DHA docosahexenoic acid are produced in the hundreds of millions of tonnes at a EPA eicosapentenoic acid price of only $100–300 per tonne. These simple economic TAG triacylglycerol factors mean that oleochemicals from transgenic oil crops can only realistically succeed at present either as high- Introduction value niche products for very specific applications (e.g. The manipulation of seed oil content via transgene inser- 8-linolenic acid as a therapeutic agent) or by competing on tion was one of the early successful applications of modern a larger scale with petrochemicals by virtue of higher puri- biotechnology in agriculture. Indeed, the first transgenic ty, better performance and/or environmental benefits. Of crop with a modified seed composition to be approved for course, oleochemicals will eventually become more com- unrestrictive commercial cultivation in the USA was a lau- petitive with petrochemicals as global reserves of ric oil rapeseed grown in 1995. There are two major reasons fossil-derived hydrocarbons (oil, coal and gas) begin to run for this. Firstly, rapeseed, Brassica napus, is a species that is out in the coming century [4•]. relatively amenable to transformation and regeneration, whereas many other major crops have proved more recalci- The manipulation of oils for new types of edible, rather than trant. Secondly, the metabolic pathways involved in industrial, use appears much more restricted as the vast storage oil biosynthesis appeared at first to be well defined majority of plant lipids are already regarded as desirable and potentially straightforward to manipulate via single dietary components, with better nutritional qualities than gene insertions. animal fats. There are, however, two obvious targets for modified edible oils in seeds. Firstly, to manipulate the ratio of Nevertheless, much of the early optimism for producing saturated to polyunsaturated fatty acids in order to avoid the designer oilseeds has, over recent years, been tempered by need for chemical hydrogenation, which produces the high setbacks in obtaining high yields of specific novel fatty levels of trans-fatty acids that many believe to be undesirable acids in transgenic oilseed crops. During the past two in the diet [5]. Secondly, there is increasing interest in pro- years, there has been an increasing recognition of the com- ducing very long-chain polyunsaturates, such as plexity of the metabolic pathways involved in seed oil docosahexenoic acid (DHA) and eicosapentenoic acid (EPA), biosynthesis and several new enzymes have been discov- which are nutritionally beneficial as precursors for certain ered that contribute to these processes in quite prostaglandins and as cholesterol-lowering agents [6]. These 176 Plant biotechnology

fatty acids are particularly enriched in fish oils but are now in genes encoding the vast majority of these enzymes. Some increasingly short supply due to the depletion of the world’s of the most significant developments have taken place in fish stocks. A recent advance here is the isolation of a gene characterising desaturases and related diiron-oxo proteins ∆ • encoding a 5 desaturase from the yeast Mortierella aplina [7]. in plants, as recently reviewed in detail [13 ]. This enzyme is responsible for the conversion of di-homo-γ- linolenic acid to arachidonic acid which is the central It now appears that plants contain two major families of precursor for the production of eicosonoids, such as diiron-oxo enzymes. Firstly, there are the soluble plastid- prostaglandins, leukotrienes and thromboxanes. localised acyl-acyl carrier protein (ACP) desaturases that typically work on C14, C16 and C18 saturated acyl-ACP While the manipulation of oil crops for human consumption substrates. These plant desaturases have similar tertiary has many attractions, this is an acutely sensitive topic of structures and ligand-binding sites to microbial methane public concern at present, particularly in the case of new monooxygenases and ribonucleotide reductases and fall food products. For example, petroselinic acid, which can into class II of the diiron-oxo enzymes [14]. One of these serve as a useful industrial raw material for polymer and plant enzymes, ∆-9 stearoyl-ACP desaturase, was the first detergent manufacture, has also been proposed as a harden- desaturase from any organism for which a high resolution ing agent for margarines [8]. Dietary studies in rats, crystal structure (down to 2.4 Å) was obtained [15]. This however, indicate that petroselinic acid ingestion is associat- knowledge has considerably assisted efforts to engineer ed with liver abnormalities and inhibition of arachidonic novel positional and chain-length specificities into desat- acid biosynthesis [9,10]. This case illustrates the unforeseen urases, for example, via site-directed mutagenesis. For difficulties that may arise from the introduction of novel example, using information from the crystal structure, a (particularly transgenic-derived) oils into the diet and indi- ∆-9 stearoyl-ACP desaturase was converted into an cates that such transgenic oil crops may be better targeted enzyme with a substrate preference for palmitoyl-ACP by initially to produce industrial, rather than edible, products. the replacement of two residues (Leu118→Phe and Pro179→Ile) [16••]. In a parallel study, the single site- Non-oil products directed mutagenesis of residue Leu118→Trp resulted in In addition to producing seed oils with novel fatty acid com- the conversion of a stearoyl-ACP desaturase to an enzyme positions, there are numerous other actual or potential with an 80-fold increase in specificity for palmitoyl-ACP applications of transgenic oil crops. For example, as recent- [17]. This represents one of the first successful attempts at ly reviewed [11], following the insertion of a relatively small the rational modification of an enzyme of lipid biosynthe- number of genes from certain bacteria, such as Alcaligenes sis. In the future, this approach holds great promise for the spp, carbon can be diverted from oil synthesis towards the re-engineering of desaturases and other enzymes for the accumulation of polyhydroxyalkanoates. These polyesters production of novel fatty acids in transgenic oil crops. are biodegradable thermoplastics. Their use is currently limited by their high price (up to tenfold higher than The second type of plant desaturase falls into class III of conventional plastics) due to the high cost of their manufac- the diiron-oxo enzymes; all such proteins are membrane- ture via bacterial fermentation. Significant reductions in the bound and utilise either complex lipid or acyl-CoA price of such biodegradable polyesters could be expected if substrates. It has recently been demonstrated that several they were produced instead via the large-scale cultivation of hydroxylases [18,19], epoxidases and acetylinases [20••] transgenic oil crops. This would probably result in a consid- are also members of this family of enzymes, and that all erable increase in their share of the enormous global market such proteins contain a similar ligand-binding site involv- for plastics. It is interesting that the major US agribusiness ing three separate histidine clusters. All of the class III company Monsanto has now acquired from Zeneca/ICI the diiron-oxo proteins probably have similar tertiary struc- rights to polyhydroxyalkanoate production in plants. tures involving four transmembrane domains, although otherwise their amino acid sequences can be quite diver- In a separate development, rapeseed oil has been used as a gent. The isolation of the above desaturase-like enzymes ‘carrier’ in order to facilitate the purification and large-scale raises the exciting possibility of the rational design of both production of pharmaceutical peptides and other high-value membrane-bound and soluble desaturases in order to carry proteins. This is via a recombinant oleosin-fusion protein out a wide range of chemical reactions, including the stere- technology developed by the Canadian biotech company ospecific insertion of conjugated double and triple bonds SemBioSys [P1]. The level of interest in this technology is and expoxy or hydroxy groups in almost any position in an illustrated by the recent investment by Dow Elanco of $17 alkyl chain. This may allow us in the future to produce a million for its commercialisation via the cultivation of trans- whole host of novel fatty acid derivatives, many of which genic rapeseed plants in western Canada [12]. are difficult, or even impossible, to synthesise by conventional chemical methods. Efforts are now underway to Engineering fatty acid desaturases obtain high-resolution structural information about these An overview of the major pathways involved in storage membrane-bound desaturases and desaturase-related pro- lipid metabolism is shown in Figure 1. Over the past few teins [21•], which will be important for their future years there has been considerable progress in isolating re-engineering to produce commercial products. Production of novel oils in plants Murphy 177

Figure 1

Sucrose Plastid Endoplasmic reticulum

Other C14–C18 24:1 monounsaturates C14–C18 KAS monounsaturates G3P G6P G6P 16:1 18:1∆ 6 22:1 DGAT ∆6 KAS

LPA 20:1 HYD 18:1-OH 14:1 18:1∆9 KAS ∆9 Pyr Pyr acyl-CoAs EPX LPAT 18:1∆ 18:1 epoxy acyl-CoA ACS ACBP? 9 ACP– ACT pool Signalling & ϖ DES 3DES PA membrane Acetyl-CoA 18:2 acetylinic ∆ lipids ACC 6DES 18:2 γ18:3 Malonyl-CoA C –C DAG 8 18 ϖ TA saturates 3DES KAS DGAT α FAS 18:3 TE Storage C8–C18 TAG saturates oil body

OLN (i) Fatty acid (ii) Fatty acid (iii) Assembly of biosynthesis modification complex lipids

Unusual (iv) Removal of unusual fatty acids faty acids

Acetyl-CoA β-oxidation

Sucrose Glyoxylate cycle Current Opinion in Biotechnology

Storage lipid metabolism in plant tissues. Fatty acid precursors, such β-ketoacyl-ACP synthetase-dependent elongases (KAS) [26•]. All of as pyruvate (Pyr) [39] and malate [40], are imported into plastids for these modified fatty acids, together with the plastid-derived saturates conversion to acetyl-CoA. The nature of the imported precursor may of monounsaturates, comprise the acyl-CoA pool of the ER. This pool be a major determinant of whether carbon is channelled to fatty is utilised by acyltransferases (glycerol-3-phosphate acyltransferase acids, and hence oil, or to starch, for example, as in cereals [41]. The [GPAT], lysophosphatidate acyltransferase [LPAT], and diacylglycerol fatty acid synthetase (FAS) complex then converts acetyl-CoA and acyltransferase [DGAT]) for the synthesis of storage triacylglycerols • • malonyl-CoA units into C8–C18 saturated acyl-ACPs, whose final (TAGS) [27 ,28,29 ], although some fatty acids may also be chain length is regulated by β-ketoacyl-ACP synthetases (KAS) channelled to signalling or membrane lipid formation. In some [24,25•,26•] and thioesterases (TE) [23]. Depending on the plant transgenic plants, the accumulation of unusual fatty acids (possibly β species, C14–C18 saturates may be desaturated by a variety of on membrane lipids) induces acyl breakdown via the -oxidation and soluble acyl-ACP desaturases (ACP-DES) [12]. The acyl-ACPs are glyoxysomal pathways [35,36•]. Finally, storage oil bodies normally then converted to acyl-CoAs [42,43] and exported to the bounded by an oleosin annulus bud off from the ER, although even endoplasmic reticulum (ER), possibly with involvement of an acyl- here the triacyloglycerol may still be available for further metabolism, CoA binding protein (ACBP) [44,45]. On the ER membrane, oleate for example, via transacylases (TA) [30••,31•,32•]. ACS, acyl-CoA is a central metabolite that can be subject to a variety of synthetase; DAG, diacylglycerol; G3P, glycerol 3-phosphate; G6P, modifications by various desaturases (DES) [12,21•,22], acetylinases glucose 6-phosphate; LPA, lysophosphatidic acid; OLN, oleosin; (ACT) [19], epoxidases (EPX) [19], hydroxylases (HYD) [17,18] and PA, phosphatidic acid.

Despite their often quite significant sequence differences, specificities, biological function and possible biotechno- virtually all plant membrane-bound desaturases fall into a logical applications remain to be determined [22]. single recognisable grouping [13•]. These enzymes use complex lipid substrates, such as phosphatidylcholine or Another interesting development has been the isolation of monogalactosyl diacylglycerol, and are localised mainly on a class of plant desaturases containing an amino-terminal the endoplasmic reticulum and plastid envelope mem- cytochrome b5 domain [23]. At present, desaturase– branes. A new subclass of membrane-bound desaturases cytochrome b5 fusions are only found in ‘front end’ desat- with similarity to animal and yeast acyl-CoA-dependent urases, that is, enzymes that introduce a double bond into and cyanobacterial acyl lipid-dependent desaturases has an acyl chain between an existing double bond and the car- recently been identified in plants, although their substrate boxy terminal. As this type of desaturase is involved in the 178 Plant biotechnology

synthesis of medically important fatty acids, such as insect cells, produced a protein with DGAT activity in vitro γ-linolenic acid, EPA and DHA, this discovery has implica- [30••]. Very recently, a homologue of this gene has been iso- tions for ongoing efforts to engineer transgenic oil crops lated from Arabidopsis and the derived protein has been with high levels of such products. shown to have DGAT activity when expressed in insect cells (MJ Hills, personal communication). The isolation of this Other key enzymes of fatty acid modification key gene may allow for more radical manipulation of seed oil yield in transgenic crops. It also holds out the prospect of One of the earliest successes in producing transgenic plants engineering high levels of storage oil accumulation in other with modified storage oil was the addition of a California sink tissues, such as tubers and fruits. As the biomass of the Bay thioesterase gene to rapeseed, resulting in the accumu- latter is normally much higher than that of most seeds, this lation of ~40% lauric acid in its seed triacylglycerol (TAG). could both increase yields and cut the costs of vegetable oils The accumulation of higher levels (50–60%) of this C12 to the extent that they may eventually compete economical- fatty acid required the additional transfer of a coconut sn-2 ly with petroleum as bulk industrial raw materials. acyltransferase gene [3]. The important contribution of thioestereases to oil quality has also been shown by the accumulation of ~20% stearic acid in transgenic rapeseed Until recently, the TAG produced in storage tissues was containing a thioesterase gene from the tropical tree man- regarded as an end product which remained metabolically gosteen [24]. Such an oil could have uses in the production inert until its mobilisation following seed germination. This of margarines and other spreads. There are also several view has now been questioned by several studies that reports, however, that demonstrate the importance of demonstrate the accessibility of storage TAG to further β-ketoacyl-ACP synthases in regulating the accumulation of metabolism, for example, by desaturases [31•]. The concept both short- [25•,26•] and long-chain [27•] fatty acids in stor- of TAG remodelling has received further support from stud- age oils. These studies indicate that it may well be possible, ies in developing safflower seeds showing transacylase in principle, to use β-ketoacyl-ACP synthase genes as part of activities capable of exchanging acyl groups between mono- a strategy to engineer oils with fatty acid chain lengths from , di- and tri-acylglycerols [32•]. It is already known that the C8 to at least C24. In order to achieve the high levels of the three acyltransferases of the TAG biosynthetic pathway can, desired fatty acid that are often required by industry, how- by virtue of their substrate specificities, play important roles ever, it may be necessary to transfer at least two additional in regulating the fatty acid composition of storage oils [3]. genes (thioesterase and sn-2 acyltransferase), and possibly The additional discovery of transacylases in safflower and several more, into the oil crop of interest. In addition to castor bean [33] raises the question of whether such increasing the development time and financial costs, the enzymes are distributed more widely and whether they too presence of multiple transgenes can sometimes lead to play a role in the regulation of oil quality in plants. instability of gene expression (e.g. co-suppression effects). Fatty acid segregation and recycling Clearly the primary enzymes of fatty acid biosynthesis and An important challenge facing biotechnologists is to devel- modification are essential to lipid accumulation. There is op transgenic oil crops, such as rapeseed, with high levels now a growing recognition, however, that enzymes further of useful fatty acids, many of which are not normally pro- downsteam in the metabolic pathways also play key roles duced by such species [3]. To date, most transgenic lines in regulating both the channelling of fatty acids to storage have been reported to accumulate relatively low (typically (rather than membrane) lipids and in determining their 1–40%) levels of the new fatty acids, such as ricinoleic overall yield in the seed or fruit [28]. For example, the [18,19], stearic [24], or γ-linolenic [34]. One explanation for expression of a yeast sn-2 acyltransferase gene in trans- this is that rapeseed appears to be less efficient at segre- genic Arabidopsis and rapeseed has been reported to result gating exotic fatty acids away from accumulation in in substantial (8–48%) increases in seed oil content [29•]. membrane lipids than are the species that originally pro- This result was unexpected as the sn-2 acyltransferase was duced such fatty acids [35]. not regarded to be a rate-limiting step in triacylglycerol formation. It is possible that this is partially due to the use of Accumulation of some fatty acids can lead to membrane the highly active CaMV 35S promoter to drive expression instability and may trigger protective mechanisms leading of the transgene but, if confirmed, this finding also illus- to the removal of these fatty acids. For example, the pres- trates how little we know about the regulation of carbon ence in transgenic rapeseed of exotic fatty acids, such as flux to oil in plant storage tissues. lauric [36•] and petroselinic [37], can induce the pathways for β-oxidation and the glyoxylate cycle leading to the The only enzyme that is unique to storage TAG formation is selective breakdown of the novel fatty acids. In some diacylglycerol acyltransferase (DGAT) — all of the other cases, this breakdown is compensated for by an upregula- enzymes can and do also contribute to membrane lipid tion of fatty acid biosynthesis [36•] but in transgenic biosynthesis. The isolation of a DGAT gene has for long rapeseed lines producing petroselinic acid, we observed a been a ‘Holy Grail’ of researchers in both animal and plant dramatic and specific breakdown of this fatty acid during lipid metabolism. It was, therefore, interesting to learn of the seed development [37]. Clearly, it will be necessary in cloning of a mouse cDNA which, when expressed in H5 future to elucidate the mechanisms involved in channeling Production of novel oils in plants Murphy 179

unusual fatty acids away from membrane lipids and ensur- 2. McConn M, Creelman RA, Bell E, Mullet JE, Browse J: Jasmonate is essential for insect defense in Arabidopsis. Proc Natl Acad Sci ing that such protective catabolic pathways are not USA 1997, 94:5473-5477. induced. This will be an important objective if we are to 3. Murphy DJ: Engineering oil production in rapeseed and other oil realise the goal of producing transgenic plants with high crops. Trends Biotechnol 1996, 14:206-213. yields of novel valuable fatty acids. 4. Kerr RA: The next oil crisis looms large — and perhaps close. • Science 1998, 281:1128-1131. This review compares estimates from six independent geological studies, all Conclusions of which predict that world petroleum production will peak between 2000–2020. Even recently discovered oil fields in the Caspian Basin may Although nearly all of the genes encoding enzymes of stor- only extend this by 2–3 years. Although some optimists disagree, the gen- age lipid biosynthesis have now been cloned, there have eral consensus is for a steady fall in petroleum supply in the coming decades and, therefore, large price increases. This may enable many plant-derived been many surprising results when these genes are oleochemicals to compete economically with petrochemicals in the not too expressed in transgenic plants. This highlights our relative distant future. ignorance of the interactions between the components of 5. Fritsche J, Steinhart H: Analysis, occurrence, and physiological properties of trans fatty acids (TFA) with particular emphasis on this and other metabolic pathways in vivo. We also know conjugated linoleic acid isomers (CLA) — a review. Fett/Lipid 1998, very little about the mechanisms regulating the partitioning 100:190-210. of carbon to storage products in sink tissues such as oilseeds. 6. Newton IS: Long-chain polyunsaturated fatty acids — the new A very promising recent approach is to attempt to identify frontier in nutrition. Lipid Technol 1998, 10:77-81. and map quantitative trait loci (QTL) that contribute to 7. Michaelson LV, Lazarus CM, Griffiths G, Napier JA, Stobart AK: Isolation of a D5-fatty acid desaturase gene from Mortierella characteristics such as oil yield or fatty acid composition. alpina. J Plant Biochem 1998, 30:19055-19059. This can be combined with map-based cloning of the major 8. Ohlrogge JB: Design of new plant products: engineering of fatty genes involved and hence the elucidation of their function acid metabolism. Plant Physiol 1994, 104:821-826. • [38 ,39]. Such a ‘top down’ genetics approach may allow for 9. Weber N, Vosmann K, Bruhl L, Mukherjee KD: Metabolism of dietary the isolation of higher level regulatory components (e.g. petroselinic acid: a dead-end metabolite of desaturation/chain transcription factors), that have already been shown to be elongation reactions. Nutr Res 1997, 17:89-97. important in the control of other metabolic pathways, such 10. Richter KD, Mukherjee KD, Weber N: Fat infiltration in liver of rats induced by different dietary plant oils: high oleic-, medium oleic- • as anthacyanin biosynthesis [38 ]. It is important that this is and high petroselinic-acid-oils. Z Ernahrungswiss 1996, combined with the ‘bottom up’ approaches, via biochem- 35:241-248. istry and analysis of individual genes and enzymes, in order 11. Steinbuchel A, Fuchtenbusch B: Bacterial and other biological to understand fully and hence be able to modify the com- systems for polyester production. Trends Biotechnol 1998, 16:419-427. plex processes of oil accumulation in plants. 12. Anon: Chemical Week, December 24/31 1997. 13. Murphy DJ, Piffanelli P: Fatty acid desaturases: structure In the future, it will also be useful to consider alternatives to • mechanism and regulation. In Plant Lipid Biosynthesis: Recent producing more and more novel oils in the very restricted Advances of Agricultural Importance. Edited by Harwood JL. number of major oil crops that are cultivated at present. The Cambridge: Cambridge University Press; 1998:95-130. A comprehensive review of soluble and membrane-bound desaturases in management problems of segregating dozens of batches of plants and their roles in the biosynthesis of membrane, signalling and stor- transgenic seeds, which may look identical but have very age lipids. different oil profiles, have been pointed out [3]. An alterna- 14. Kurtz DM: Structural similarity and functional diversity in diiron- oxo proteins. 2 tive approach is to use modern biotechnological methods, J Bioinorganic Chem 1997, :159-167. 15. Lindqvist Y, Huang W, Schneider G, Shanklin J: Crystal structure of a such as genome mapping and molecular marker assisted delta-9 stearoyl-acyl carrier protein desaturase from castor seed selection, to effect the rapid domestication of new oil crops and its relationship to other diiron proteins. EMBO J 1996, that already produce very high levels of the desired fatty 15:4081-4092. acids [38•,39]. In the longer term, it may be agronomically 16. Cahoon EB, Lindqvist Y, Schneider G, Shanklin J: Redesign of •• soluble fatty acid desaturases from plants for altered substrate more manageable, as well as being more desirable ecologi- specificity and double bond position. Proc Natl Acad Sci USA cally, to cultivate both a limited number of transgenic crops 1997, 94:4872-4877. This is a landmark study, in which high-resolution structural information about plus some newly domesticated species. These oil crops will a castor bean ∆-9 stearoyl-ACP desaturase was used for the rational engi- be essential sources of both edible and industrial oils that neering of a novel substrate specificity into both ∆-9 and ∆-6 acyl-ACP desaturases. In one case, five residues were changed to convert a ∆-6 to a will be in particular demand in the coming decades as glob- ∆-9 desaturase, whereas in the other case, only two residue alterations were al populations increase and non-renewable fossil oils needed to convert a ∆-9 to a ∆-6 desaturase. This opens up numerous pos- • sibilities for the manipulation of both fatty acid chain length and double bond become significantly depleted [4 ]. insertion specificities in soluble acyl-ACP desaturases. 17. Cahoon EB, Shah S, Shanklin J, Browse J: A determinant of References and recommended reading substrate specificity predicted from the acyl-acyl carrier protein Papers of particular interest, published within the annual period of review, desaturase of developing cat’s claw seed. Plant Physiol 1998, have been highlighted as: 117:593-598. • of special interest 18. Broun P, Somerville C: Accumulation of ricinoleic, lesquerolic, and •• of outstanding interest densipolic acids in seeds of transgenic Arabidopsis plants that express a fatty acyl hydroxylase cDNA from castor bean. Plant 1. Kirsch C, Takamiya-Wik M, Reinold S, Hahlbrock K, Somssich IE: Physiol 1997, 113:933-942. Rapid, transient, and highly localized induction of plastidial omega-3 fatty acid desaturase mRNA at fungal infection sites in 19. Broun P, Boddupalli S, Somerville C: A bifunctional oleate 12- Petroselinum crispum. Proc Natl Acad Sci USA 1997, hydroxylase: desaturase from Lesquerella fendleri. Plant J 1998, 94:2079-2084. 13:201-210. 180 Plant biotechnology

20. Lee M, Lenman M, Banas A, Bafor M, Singh S, Schweizer M, 32. Stobart K, Mancha M, Lenman M, Dahlqvist A, Stymne S: •• Nilsson R, Liljenber C, Dahlqvist A, Gummeson PO et al.: • Triacylglycerols are synthesised and utilised by transacylation Identification of non-heme diiron proteins that catalyze triple reactions in microsomal preparations of developing safflower bond and expoxy group formation. Science 1998, 280:915-918. (Carthamus tinctorius L.) seeds. Planta 1997, 203:58-66. This paper reports the isolation and expression in Arabidopsis of two genes In a complementary study to [31•], the authors describe evidence for a novel encoding a ∆-12 epoxidase and a ∆-12 acetylinase. It is an important exper- transacylase system in higher plants. This can transfer acyl groups between imental confirmation that such enzymes of fatty acid modification fall into the mono-, di-, and tri-acylglycerols and provides a mechanism for explaining same family of membrane-bound class III diiron-oxo enzymes as desaturas- how storage triacylglycerols can be made available for remodelling. It will be es and hydroxylases. It also demonstrates the potential for producing epoxy important to elucidate the contribution of transacylases both to acyl compo- and acetylinic oils in transgenic crops, such as rapeseed. sition and to overall oil yield in plants. 21. Shanklin J, Cahoon EB, Whittle E, Lindqvist Y, Huang W, 33. Mancha M, Stymne S: Remodelling of triacylglycerols in • Schneider G, Schmidt H: Structure-function studies on desaturases microsomal preparations from developing castor bean (Ricinus and related hydrocarbon hydroxylases. In Physiology, Biochemistry communis L.) endosperm. Planta 1997, 203:51-57. and Molecular Biology of Plant Lipids. Edited by Williams JP, Khan MU, Lem NW. Dordrecht, The Netherlands: Kluwer; 1997:6-10. 34. Sayanova O, Smith MA, Lapinskas P, Stobart AK, Dobson G, Christie A brief report on the initial steps to determine the high-resolution structure of an WW, Shewry PR, Napier J: Expression of a borage desaturase insoluble class III diiron-oxo enzyme. Although the initial target is the alkaline cDNA containing an N-terminal cytochrome b5 domain results in hydroxylase of Pseudomonas oleovorans, the structural information is likely to the accumulation of high levels of delta-6-desaturated fatty acids be relevant to the plant membrane-bound desaturases and related enzymes. in transgenic tobacco. Proc Natl Acad Sci USA 1997, 94:4211-4216. 22. Fukuchi-Mizutani M, Tasaka Y, Tanaka Y, Ashikari T, Kusumi T, Murata N: Characterization of delta-9 acyl-lipid desaturase 35. Wiberg E, Banas A, Stymne S: Fatty acid distribution and lipid homologues from Arabidopsis thaliana. Plant Cell Physiol 1998, metabolism in developing seeds of laurate-producing rape 39:247-253. (Brassica napus L.). Planta 1997, 203:341-348. A new class of 36. Eccleston VS, Ohlrogge JB: Expression of lauroyl-acyl carrier 23. Napier JA, Sayanova O, Stobart AK, Shewry PR: • cytochrome b5 fusion proteins. Biochem J 1997, 328:717-718. protein thioesterase in Brassica napus seeds induces pathways for both fatty acid oxidation and biosynthesis and implies a set 24. Hawkins DJ, Kridl JC: Characterization of acyl-ACP thioesterases of point for triacylglycerol accumulation. Plant Cell 1998, mangosteen (Garcinia manostana) seed and high levels of 10:613-621. stearate production in transgenic canola. Plant J 1998, 13:743-752. The presence of a thioesterase transgene in rapeseed leads to the induction of β-oxidation and glyoxylate cycle genes. This leads to the breakdown of a 25. Slabaugh MB, Leonard JM, Knapp SJ: Condensing enzymes from large proportion of newly synthesised lauric acid to acetyl-CoA and sucrose. • Cuphea wrightii associated with medium chain fatty acid Most of this carbon is recovered for oil formation due to an increase in fatty biosynthesis. Plant J 1998, 13:611-620. • acid synthetase activity but the result is a wasteful futile cycle of metabolism. This report and the accompanying paper [26 ] describe the importance of β- The results indicate that rapeseed is not efficient in channeling novel fatty ketoacyl-ACP synthases in the determination of short-medium chain lengths acids towards storage oil and that this leads to the induction of mechanisms in seed oils. This is in addition to the long recognised role of thioesterases in to prevent their accumulation in membranes. This illustrates some of the specifying chain length and confirms that this is a complex trait likely to be unexpected pleiotropic consequences of transgene insertions in plants. regulated by at least three sets of genes (including acyltransferases) in plants. 37. Fairbairn DJ, Bowra S, Murphy DJ: Expression of unusual fatty acids 26. Leonard JM, Knapp SJ, Slabaugh MB: A Cuphea b-ketoacyl-ACP in transgenic rapeseed causes induction of glyoxylate cycle • synthase shifts the synthesis of fatty acids towards shorter genes. John Innes Centre Annual Report, 1998-99: in press. chains in Arabidopsis seeds expressing Cuphea FatB thioesterases. Plant J 1998, 13:621-628. 38. Murphy DJ: Impact of genomics on improving the quality of See annotation to [25•]. • agricultural products. In Genomics: Commercial Opportunities from a Scientific Revolution. Edited by Dixon GK, Copping LG, 27. Millar AA, Kunst L: Very-long-chain fatty acid biosynthesis is Livingstone D. Cambridge: University of Cambridge, Society of • controlled through the expression and specificity of the Chemical Industry; 1997:199-210. condensing enzyme. Plant J 1997, 12:121-131. This review looks at some of the general issues relating to the use of An important study showing that it is the β-ketoacyl-ACP synthase (KAS) genomics for the improvement of quality characters, such as oil yield that is the component of the fatty acid elongase which determines acyl chain in crops. length in both plants and yeast. Hence, the same class of enzyme is a key regulator of both short, medium [25•,26•] and very long chain fatty acid 39. Martin GB: Gene discovery for crop improvement. Curr Opin accumulation in oilseeds. Biotechnol 1998, 9:220-226. 28. Kinney AJ: Manipulating flux through plant metabolic pathways. 40. Eastmond PJ, Rawsthorne S: Developmental changes in substrate Cur Opin Plant Biol 1998, 1:173-178. utilization for fatty acid synthesis by plastids isolated from oilseed rape embryos. In Physiology, Biochemistry and Molecular 29. Zou J, Katavic V, Giblin EM, Barton DL, MacKenzie SL, Keller WA, Biology of Plant Lipids. Edited Williams JP, Khan MU, Lem NW. • Hu X, Taylor DC: Modification of seed oil content and acyl Dordrecht, The Netherlands: Kluwer; 1997:66-68. composition in the Brassicaceae by expression of a yeast sn-2 acyltransferase gene. Plant Cell 1997, 9:909-923. 41. Eastmond PJ, Dennis DT, Rawsthorne S: Evidence that a In this unexpected finding, the authors report that the expression of a yeast malate/inorganic phosphate exchange translocator imports sn-2 acyltransferase gene in transgenic rapeseed and Arabidopsis resulted carbon across the leucoplast envelope for fatty acid synthesis in in significant (<48%) increases in seed oil yield. If confirmed, this indicates developing castor seed endosperm. Plant Physiol 1997, that apparently non rate-limiting enzymes may still exert considerable control 114:851-856. over carbon flux in oilseeds. 42. Eastmond PJ, Kang F, Rawsthorne S: Carbon flux to fatty acids in 30. Cases S, Smith SJ, Zheng Y, Myers HM, Sande ER, Novak S, plastids. In Regulation of Primary Metabolic Pathways in Plants. •• Lear SR, Erickson SK, Farese RV: Cloning and expression of a Edited by Kruger NJ, Hills SA, Ratcliffe RG. Dordrecht, The candidate gene for diacylglycerol acyltransferase. FASEB J 1998, Netherlands: Kluwer; 1999:137-157. 12:A814. 43. Fulda M, Heinz E, Wolter FP: Brassica napus cDNAs encoding fatty This is the first report of the isolation of a diacylglycerol acyltransferase acyl-CoA synthetase. Plant Mol Biol 1997, 33:911-922. (DGAT) from any organism. The enzyme catalyses the last step in triacylglycerol biosynthesis and is the only activity not shared with membrane lipid 44. Chye ML: Arabidopsis cDNA encoding a membrane-associated formation. As such, DGAT is a key target for attempts to manipulate oil yield protein with an acyl-CoA binding domain. Plant Mol Biol 1998, in seeds and fruits, or even to redirect oil accumulation to other tissues, such 38:827-838. as tubers. Doubtless, the sequence data from this murine DGAT will be useful to isolate homologs from plants. 45. Brown AP, Johnson P, Rawsthorne S, Hills MJ: Expression and properties of acyl-CoA binding protein from Brassica napus. Plant 31. Sarmiento C, Garces R, Mancha M: Oleate desaturation and acyl Physiol Biochem 1998, 36:629-635. • turnover in sunflower (Helianthus annuus L.) seed lipids during rapid temperature adaptation. Planta 1998, 205:595-600. The latest in a series of reports that triacylglycerols in sunflower seeds are Patent still available for modification by desaturases, that is, storage oil is not nec- P1. Moloney M: Oil body proteins as carriers of high value proteins. essarily an inert end-product of metabolism. Industrial patent application 11 July 1997, WO 96/21 029.