| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Department of Plant Science, University of Manitoba, Winnipeg, MB, R3T 2N2, Canada
b Department of Botany and Plant Sciences, University of California, Riverside, CA 92521
* Corresponding author (Rachael_Scarth{at}umanitoba.ca)
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONREFERENCES |
|---|
Abbreviations: ACP, acyl carrier protein • ALA, 
-linolenic acid, C18:3 • C12:0, lauric acid • GLA, 
-linolenic acid • LA, linoleic acid, C18:2 • SMCFA, short and medium chain fatty acids • PUFA, polyunsaturated fatty acid • TE, acyl-acyl carrier protein thioesterase
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONREFERENCES |
|---|
Brassica is one of 51 genera in the tribe Brassiceae in the Cruciferae family (syn. Brassicaceae) (Gómez-Campo, 1980). Of the 37 species in the Brassica genus, the four most widely cultivated species for oilseed and vegetable production are B. rapa L., B. juncea (L.) Czernj. & Cosson, B. napus L., and B. carinata A. Braun. (Raymer, 2002; Rakow, 2004; Sovero, 1993). The world's Brassica commerce consists mainly of seed produced from the two species B. napus and B. rapa in Canada and Australia (Rakow, 2004; Raymer, 2002). The three amphidiploid species, including B. napus (genome AACC, 2n = 4x = 38), B. juncea (AABB, 2n = 4x = 36), and B. carinata (BBCC, 2n = 4x = 34), originated from interspecific hybridization and spontaneous chromosome doubling of the three diploid ancestors: B. rapa (AA, 2n = 20, syn. B. campestris L.), B. nigra (L.) Koch (BB, 2n = 16), and B. oleracea L. (CC, 2n = 18) (U, 1935; Lühs and Friedt, 1994; Rakow, 2004). Artificial resyntheses of the amphidiploids by interspecific crosses of the ancestor diploids have confirmed the relationship (Frandsen, 1943 and 1947; U, 1935). Interspecific crossing has been used for transferring genetic variation for seed oil modification among the closely related Brassica species.
The nutritional and industrial value of Brassica oil, like other vegetable oils, is determined by its fatty acid profile, made up of several fatty acid species with distinct carbon chain length and level of desaturation. The concentration of a fatty acid in plant seed oil is reported in most of the literature as weight percentage (wt% or %) of the fatty acid in the total fatty acid composition; alternatively, the concentration is expressed as mole percentage (mol%). The difference between the two methods of expressions can be up to 3 to 4 percentile points for the same fatty acids in the major commercial Brassica oils, depending on the molecular weight of the fatty acid and the fatty acid profile of the seed oil (Eskin et al., 1996). The difference can be more significant for high levels of shorter and longer chain fatty acids in genetically modified Brassica oils; for example, the maximum C12:0 level was 56.1 mol% in a doubled haploid population expressing a Bay thioesterase transgene, which was equivalent to 48.2 wt% (Tang and Scarth, 2004). Accurate conversion of mol% to wt% for a particular fatty acid requires the availability of the complete fatty acid composition of the oil and this is usually not reported in the original papers cited in this review which adopted the mol% expression. Therefore, the authors have chosen to clearly indicate which method of expressions, wt% (represented as %) or mol%, was used in the original source cited.
Rapeseed oil produced by traditional Brassica oilseed cultivars (B. napus, B. rapa, and B. juncea) (Shahidi, 1990; Sovero, 1993), typically has a fatty acid composition of 5% palmitic (C16:0), 1% stearic (C18:0), 15% oleic (C18:1), 14% linoleic (C18:2), 9% linolenic (C18:3), and 45% erucic acid (C22:1) (Ackman, 1990). C22:1 is a nutritionally undesirable fatty acid and has been reduced to very low levels in Brassica oil for edible uses. Low C22:1 Brassica oil has a nutritionally desirable fatty acid profile, with low saturated fatty acids and significant levels of C18:3, an omega-3 fatty acid (Eskin et al., 1996). C22:1 does have significant value in industrial applications and, for these uses, it is desirable to increase the 45% level in traditional rapeseed oil as high as possible to improve the economical competitiveness of the high erucic acid rapeseed (HEAR) oil and its derivatives.
In addition to the common fatty acids in Brassica oil, members of the plant kingdom produce more than 200 unusual fatty acids, particularly in nonagronomic plants (van de Loo et al., 1993; Thelen and Ohlrogge, 2002; Jaworski and Cahoon, 2003). Many of these fatty acids have nutritional benefits or industrial uses. However, most of these plants have limited potential of domestication. Because of the relatively low cost and renewable nature of oilseed production, oilseed crops including Brassica oilseeds can be modified to produce the novel fatty acids as an alternative source to petroleum-derived industrial feedstock (Cahoon, 2003; Thelen and Ohlrogge, 2002). This review paper presents the past achievements, current status, and potential developments in the modification of Brassica oil with both conventional and transgenic breeding approaches.
Conventional Breeding Approaches
Seed oil quality and utility, determined mainly by its fatty acid composition, has been a major consideration in Brassica breeding programs worldwide. Conventional breeding has contributed to the development of four major types of Brassica oils with altered fatty acid compositions aimed for different markets (Burton et al., 2004; Przybylski, 2005). These developments were the result of plant breeding using natural and artificially induced mutations within Brassica species.
Low Erucic Acid
The selection and development of Brassica oils low in C22:1 was initiated in 1950s, following the identification of potential human health concerns in animal feeding studies with rapeseed oil high in C22:1. Animal studies showed diets high in C22:1 rapeseed oil was associated with myocardial damage characterized by fatty deposits around the heart and kidneys and muscle lesions in the heart (Eskin et al., 1996; Taylor et al., 1994). Mutants with low levels of C22:1 in the seed oil were first identified from a German spring-type B. napus forage cultivar Liho in 1959 (Stefansson et al., 1961). The low C22:1 plants were backcrossed with adapted cultivars, resulting in the development of the first low C22:1 B. napus cultivar Oro in 1968 and the first low C22:1 B. rapa cultivar Span in 1971 (Stefansson and Downey, 1995). By 1974, 95% of the rapeseed growing in Canada was of a low C22:1 fatty acid composition (Eskin et al., 1996).
The term "Canola" is a registered trademark of the Canola Council of Canada (Canola Council of Canada, 2001). Conversion to canola varieties in Canada began with the licensing of the first B. napus canola cv. Tower in 1974 (Canola Council of Canada, 2001), and the first B. rapa canola cv. Candle in 1977. In North America and Europe, there has been a complete conversion of the commercial production from traditional rapeseed and low erucic acid varieties to canola quality varieties (Canola Council of Canada, 2001). Rapeseed cv. with high erucic acid levels continue to be grown to produce oil for human consumption in some countries, such as India and China, although there is some conversion to low C22:1 cultivars (Sivaraman et al., 2004; Gupta et al., 2004).
High Erucic Acid
Although undesirable nutritionally, high C22:1 oils and C22:1 derivatives have more than 200 potential industrial applications, e.g., as an additive in lubricants and solvents, as a softener in textiles, and the amide derivative is used in the manufacture of polymers, high temperature fluidity lubricants, surfactants, plasticizers, surface coatings, and pharmaceuticals (Katavic et al., 2000; McVetty and Scarth, 2002; Taylor et al., 1994; Sonntag, 1991, 1995; Leonard, 1994). More than 1000 patents for applications of C22:1 have been issued (Mietkiewska et al., 2004). For many of these industrial uses, the economics are limited by the level of C22:1 content in the seed oil of approximately 45% (Sovero, 1993). To compete with petroleum-based products, it is desirable to increase the C22:1 level to as high a level as possible to reduce the cost of purification.
The high C22:1 Brassica germplasm used for the development of current HEAR cultivars released in Canada originated from a Swedish B. napus summer rape strain. The first HEAR cultivar with a low glucosinolate content, cv. Hero, was derived from a cross between a high C22:1 female parent with an adapted high C22:1 rapeseed cv. Reston with a high glucosinolate content in the seed meal, followed by selection for improved agronomic performance, high C22:1 and low glucosinolate levels (Scarth et al., 1991, 1992). Using the same selection procedure, the HEAR cv. Mercury with 54% C22:1 was developed and registered in 1992 (Scarth et al., 1995a). Recently released HEAR cv. Castor and MilleniUM01 have improved agronomic performance with the C22:1 level up to 55% (McVetty et al. 1998, 1999).
The goal of increasing the C22:1 level has been approached by resynthesizing B. napus, crossing selected lines of the two ancestral diploids, B. rapa and B. oleracea, which can incorporate C22:1 into the sn-2 position, followed by chromosome doubling (Taylor et al., 1995). Resynthesized plants can accumulate levels of C22:1 up to 60% (Lühs and Friedt 1995). In addition, PEG-induced asymmetric somatic hybridization between B. napus cv. Maplus (47.9% C22:1) and Crambe abyssinica Hochst. ex R.E. Fires cv. Galactica (55.0% C22:1), showed increased C22:1 in recovered B. napus lines (51% C22:1) (Wang et al., 2003). Progeny with up to 61.5% C22:1 were recorded from a backcross of a somatic hybrid between a zero-C22:1 B. napus line and Lesquerella fendleri (A. Gray) S. Watson to a high C22:1 line (Schröder-Pontoppidan et al., 1999). However, there has been no report to date of Brassica germplasm with >66% C22:1, which is believed to be the theoretical limit of the C22:1 level in the Brassica genetic background (Frentzen and Wolter, 1998; Schröder-Pontoppidan et al., 1999).
Genetic analysis indicated that the C22:1 level in B. napus and other two amphidiploid Brassica oilseed species is controlled mainly by two genes (genetic loci) with additive genetic interaction (Chen and Beversdorf, 1990; Downey and Harvey 1963; Getinet et al., 1997; Harvey and Downey, 1964; Kirk and Hurlstone 1983; Lühs et al., 1999). This conclusion is supported by the identification of two major loci controlling C22:1 level by genetic mapping (del Río et al., 2003; Gupta et al., 2004; Jourdren et al., 1996b; Mahmood et al., 2003). Resynthesized rapeseed showed an average additive contribution of 16 to 17% of C22:1 per allele (Lühs et al., 1999). In B. juncea, the two loci on the two different genomes contribute unequally to the total C22:1 level (Bhat et al., 2002).
The alleles or genes at the two loci may encode ß-ketoacyl-CoA synthase (KCS, also termed FAE) in the fatty acid elongation complex, which is required for C22:1 synthesis with C18:1-CoA as the substrate (Barret et al., 1998; Fourmann et al., 1998; Han et al., 1998; Lassner et al., 1996; Roscoe et al., 1998). Two KCS genes were found to cosegregate with the two major loci in independent linkage groups controlling the C22:1 level in B. napus (Jourdren et al., 1996b; Gupta et al., 2004). In B. juncea, the two loci controlling the C22:1 level also cosegregated with two KCS genes. The two genes differ at four or three single nucleotides which distinguish the low C22:1 mutant from the wild-type high C22:1 lines (Gupta et al., 2004). Additional evidence supporting the association of KCS genes with the C22:1 mutations in Brassica is that HEAR varieties possess readily detectable C18:1-CoA elongase, whereas low C22:1 cultivars do not (Pollard and Stumpf, 1980), and the Jojoba [Simmondsia chinensis, Simmondsiaceae, (Link) C. Schnieder] FAE homolog could complement the low C22:1 mutant in B. napus (Lassner et al., 1996).
Four genetic loci have been identified as determining the C22:1 level in B. carinata (del Río et al., 2003). Two genetic loci with additive gene action were found to control the C22:1 level in zero-C22:1 lines developed from an interspecific cross. Chemical mutagenesis in another two genetic loci reduced the C22:1 level in a high-C22:1 line from 40% to about 10%. These alleles are another possible source for altered C22:1 levels in addition to the KCS gene.
Low Linolenic Acid
C18:3 and C18:2, together with their longer-chain and more unsaturated derivatives, are the two series of essential polyunsaturated fatty acids (PUFA) n-3 and n-6 fatty acids, respectively, required for human development and health (James et al., 2000; Ghafoorunissa et al., 2002; Hornstra, 2000; de Lorgeril and Salen, 2004). However, the increased number of double bonds in the chemical structures of the PUFA makes them susceptible to oxidation (Browse et al., 1998; Lauridsen et al., 1999). Oxidation rates of C18:2 and C18:3 are approximately 10 and 25 times higher, respectively, than that of C18:1 (Friedt and Lühs, 1999). Therefore, oils high in C18:2 and C18:3 deteriorate more rapidly on exposure to air, especially at high temperatures, resulting in shortened shelf life of the oil (Tanhuanpää and Schulman, 2002) and making the oil less healthy for human consumption (Röbbelen and Nitsch, 1975). Normally, less stable oils are hydrogenated to enhance the stability, but the hydrogenation process causes the formation of trans fatty acids and positional isomer fatty acids (Kochhar, 2000). Trans fatty acids contain at least one double bond in the trans configuration and there is concern about their physiological functions, especially possible roles in atherosclerosis, increase of blood cholesterol, and coronary heart diseases, although this issue is still being debated (Hayakawa et al., 2000; Lichtenstein, 2000; Nicolosi and Rogers, 1997). In addition, hydrogenation, or addition of natural antioxidative components (e.g., tocopherols) as alternatives to improve the stability, increases the cost of processing commodity oils (Kochhar, 2000; Ohlrogge, 1994).
The first low C18:3 canola, cv. Stellar (3% C18:3), released for commercial production in 1987, was developed by crossing a low C18:3 germplasm M11 with a high C18:3 Canadian cv. Regent (Scarth et al., 1988). The low C18:3 line ‘M11’ was produced by seed chemical mutagenesis treatment of a high C18:3 Canadian spring-type B. napus canola cv. Oro (Röbbelen and Nitsch, 1975). Further selection for low C18:3 and improved agronomic performance resulted in the development of cv. Apollo (Scarth et al., 1995b), with lower C18:3 and a distinct fatty acid profile (McVetty and Scarth, 2002). The total elimination of C18:3 from the seed oil by conventional breeding is probably impossible, since C18:3 plays an essential role for normal plant growth and reproduction (Hugly et al., 1989; McConn and Browse, 1996; Tanhuanpää and Schulman, 2002).
The C18:3 level in B. napus is controlled by two genetic loci with additive effect (Scarth, 1995; Jourdren et al., 1996c). These loci have been colocated to two fad3 genes in B. napus by genetic mapping (Jourdren et al., 1996a; Barret et al., 1999; Somers et al., 1998; Thormann et al., 1996; Tanhuanpää and Schulman, 2002). Minor genes and maternal and cytoplasmic effects have also been associated with the variation in the C18:3 level in B. napus (Bartkowiak-Broda and Krzymanski, 1983; Diepenbrock and Wilson, 1987; Pleines and Friedt, 1989; Tanhuanpää and Schulman, 2002). As a result, the C18:3 content generally shows continuous variation in crosses between high and low C18:3 lines (Jourdren et al., 1996c). Three or more loci were found to influence the C18:3 level in B. rapa and B. juncea (Tanhuanpää and Schulman, 2002; Mahmood et al., 2003).
High Oleic Acid
Comparative studies showed that oils with high C18:1 and low C18:3 levels possess a higher oxidative stability without the requirement of partial hydrogenation and produce less undesirable products during deep frying (Fitzpatrick and Scarth, 1998; Warner and Mounts, 1993; Töpfer et al., 1995). High oleic acid oils have equivalent heat stability to saturated fats and are suitable replacements for them in commercial food-service applications that require long-life stability (Stoutjesdijk et al., 2000). The profile of the optimum oil based on these comparisons has a 67 to 75% C18:1 level (Burton et al., 2004). However, for oleochemical applications, an increase of the C18:1 fatty acid in the oil to over 90% would be of a considerable value because of reduced costs for homogenous or near-homogeneous starting materials (Töpfer et al., 1995; Abbadi et al., 2004; Katavic et al., 2000).
Regular canola cultivars have about 61% C18:1 in the seed oil (Scarth and McVetty, 1999). Identification of plants with 69% C18:1 in the seed oil, followed by self-pollination and recurrent selection, led to the development of B. napus breeding lines with 85 to 90% C18:1 (Vilkki and Tanhuanpää, 1995). Mutagenesis treatment of seeds or microspores has resulted in B. napus lines producing seed oils with 80 to 86% C18:1 (Rücker and Röbbelen, 1995; Schierholt and Becker, 1999; Wong et al., 1991). Brassica napus cultivars with 70 to 75% C18:1 and reduced C18:3 level have been released including cv. Clear Valley 75 and MONOLA (Scarth and McVetty, 1999).
High C18:1 mutation has been associated to the fad2 gene in B. napus and B. juncea (Laga et al., 2004). In B. juncea, two QTLs together could account for 32.2% variation in the C18:1 level (Sharma et al., 2002). The variation in the C18:1 level has been colocated with fad2 in B. rapa (Tanhuanpää et al., 1998).
Low Saturated Fatty Acids
C16:0 is the major contributor to the total saturated fatty acid level of vegetable oils including Brassica oil. C16:0, as well as the shorter chain fatty acids C8:0 to C14:0, are widely reported to raise plasma total cholesterol (TC) and low density lipoprotein cholesterol (LDL-C) levels in animals and humans, presumably by decreasing LDL receptor activity and/or increasing LDL-C production (Nicolosi and Rogers, 1997; Pandian et al., 2004). Of the two types of serum cholesterols classified on the basis of its carrier, the high-density lipoproteins (HDL) is beneficial as it is associated with the removal of cholesterol from the blood stream, while LDL is undesirable as it is responsible for the movement of cholesterol within the bloodstream (Pandian et al., 2004).
Canola oil is the only commercial vegetable oil meeting the criteria of the low saturated oils (<7%) as defined in the labeling regulations in the USA and Canada. This strategic position of canola oil is being challenged by considerable progress in breeding soybean with low saturate levels through conventional breeding (Burton et al., 2004; Wilcox et al., 1994). It is desirable to further reduce the saturated fatty acid level to achieve zero saturated fat levels (Scarth and McVetty, 1999). A minor decrease in the total saturated fatty acids to less than 6% was found in breeding populations derived from interspecific crosses of B. napus with B. rapa and B. oleracea (Raney et al., 1999). The lower total saturated fatty acid content in these lines represents a general reduction of all of the saturated fatty acids (Raney et al., 1999).
Transgenic Breeding Approaches
Brassica napus has been the species predominantly used as the genetic transformation targets in the Brassica genus. The fatty acid composition in the seed oil was significantly modified following the introduction of transgenes, which were usually inserted by means of the Agrobacterium-mediated transformation system. The transgenes include those coding for the enzymes which play important roles in the formation of the fatty acids and in the incorporation of the fatty acids into triacylglycerols (TAG), such as ACCase, KAS, acyl-ACP thioesterase, desaturase, elongase, and acyltransferase. Various sources have been utilized including related Brassica species, other higher or lower plants, yeast, bacteria, and mammals.
Very Low Levels of Saturated Fatty Acids
Lack of genetic variation for reduced saturated fatty acid levels within Brassica has limited the progress by conventional breeding. An alternate approach to decrease the saturate level is the regulation of the expression of a number of genes, including KAS, desaturases, and thioesterases (Dehesh, 2004). Expression of a KAS II gene in B. napus resulted in a slight reduction of the C16:0 level (Bleibaum et al., 1993). Altered C16:0 level was observed in Arabidopsis expressing KAS IV isolated from Cuphea wrightii A. Gray (Leonard et al., 1998; Slabaugh et al., 1998). Recently, the total level of saturated fatty acids in Brassica coexpressing C. pullcherima KAS I and KAS IV sequences as well as a safflower 
9-desaturase, was reported to be reduced to below 3.4%, compared with around 6.0% in nontransformed B. napus plants (Dehesh, 2004).
High Levels of Short and Medium Chain Fatty Acids
Plants oils rich in short and medium chain fatty acids (SMCFA) are useful in a number of food and nonfood industries (Health Canada, 1999; Martini et al., 1995; Ohlrogge, 1994; Töpfer et al., 1995). The commercial sources of SMCFA are coconut and palm kernel oils (Ohlrogge, 1994; Töpfer et al., 1995). Brassica seed oil has traces of SMCFA with hardly detectable levels of C8:0 and C10:0 and less than 0.02 mol% C12:0 (Dehesh et al., 1996; Voelker et al., 1996).
With the objective of developing Brassica oilseed as an alternative source of SMCFA, FatB thioesterase genes were isolated from a number of noncrop plant species accumulating high levels of SMCFA C16:0 and C18:1 in their seed oil. These FatB genes were transformed into B. napus genetic background in combination with a seed specific promoter, usually the promoter of the gene coding for B. rapa seed storage protein napin, to direct the expression of the SMCFA in the developing seeds. Other genes, such as KAS and LPPAT, from various plant species, were also introduced into B. napus in combination with thioesterase transgenes for high yield of SMCFA (Dehesh et al., 1998; Knutzon et al., 1999b; Leonard et al., 1998).
High Lauric Acid
The ability to engineer a nonlaurate-accumulating plant for C12:0 production was first demonstrated by transforming Arabidopsis with a FatB gene cloned from California Bay tree [Umbellularia californica (Hook. & Arn.) Nutt.] (Voelker et al., 1992), a species able to accumulate up to 70% C10:0 and C12:0 in the seed oil (Pollard et al., 1991). The same FatB gene was transformed into B. napus cv. 212/86, a low C22:1 breeding line (Health Canada, 1999; Voelker et al., 1996). The transgenic plants produced seed oil with up to 56 mol% C12:0 (Voelker et al., 1996).
The lower C12:0 level in transgenic B. napus seed oil compared with the seed oil of the transgene source Bay tree is partly due to the differences between the Brassica and the native Bay acyltransferases. A second transgene coding for coconut (Cocos nucifera L.) LPAAT, which acylates laurate into the sn-2 position was introduced into the FabB transgenic plants to increase the C12:0 level. The range in C12:0 levels in 100 dihaploid plants cotransformed with both the LPAAT and the thioesterase transgenes was from 0 to 67 mol% (Knutzon et al., 1999b). When the thioesterase transgenes were expressed in different B. napus genetic backgrounds, a range of C12:0 levels was also observed (Tang and Scarth, 2004).
The first genetically engineered oilseed crop with modified seed oil, the high C12:0 Brassica, was grown commercially in Georgia, USA, as a winter crop in the 1995–1996 growing season by Calgene under the brand name Laurical (GEO-PIE, 2004; Murphy, 1999). The seed oil of the Bay-FatB transformed B. napus plants was similar to the composition of coconut and palm kernel oils in the level of SMCFA (Voelker et al., 1996), which are used in food products such as in chocolates, candy coatings, confections, nondairy creamers, low-fat margarines, soaps, detergents, and cosmetics (GEO-PIE, 2004).
High Caprylic Acid and Capric Acid
The levels of caprylic acid (C8:0) and capric acid (C10:0) can exceed 80% of the total fatty acids in the seed oil of some Cuphea species (Dehesh et al., 1996; Graham et al., 1981; Hilditch and Williams, 1964; van de Loo et al., 1993). Separate transformation of B. napus with two FatB genes from C. lanceolata W. T. Aiton resulted in the accumulation of these two novel fatty acids in Brassica seed oil although at low levels (Martini et al., 1995). Higher SMCFA levels, up to 40% C8:0 and C10:0, were obtained following transformation with the Ch FatB2 thioesterase gene isolated from C. hookeriana Walp. (Dehesh et al., 1996).
High Palmitic Acid
Significant increases in the C16:0 level of seed oil were observed in transgenic plants expressing FatB thioesterase and KAS from MCFA-accumulating plant species. Transgenic plants expressing the Cuphea Ch FatB1 gene have the C16:0 level of up to 34 mol% (Jones et al., 1995). Similar levels of C16:0 were found in the seed oils of transgenic B. napus plants expressing FatB genes from elm (Ulmus americana L.) and nutmeg (Myristica fragrans Houtt.) (Voelker et al., 1997).
High Stearic Acid
Vegetable oils with high saturated fatty acid levels have applications in the manufacture of solid fat food products, such as margarine and shortening, saving the cost of hydrogenation and avoiding the production of unwanted trans fatty acid (Kritchevsky et al., 1995; Pérez-Vich et al., 2004). It is estimated that vegetable oil with approximately 30% total saturates could make a suitable trans-free margarine through the process of interesterification (List et al., 2000). C18:0 has an advantage over other forms of saturated fatty acids because it either reduces or has no effect on serum lipoprotein cholesterol (Emken, 1994; Grundy, 1994; Kris-Etherton and Yu, 1997; Pearson, 1994; Pérez-Vich et al., 2004; Spencer et al., 2003). Canola cultivars have only 1.1 to 2.5% C18:0 in the seed oil (Hawkins and Kridl, 1998). No natural or induced high C18:0 Brassica germplasm has been reported. The C18:0 level of seed oil is likely controlled by at least four enzymes (Cahoon, 2003; Hawkins and Kridl, 1998; Somerville et al., 2000), including the reaction catalyzed by KAS II, which elongates C16:0 to stearoyl-ACP; 9-desaturase, which uses C18:0 as substrate for desaturation; acyl-ACP thioesterases, which releases acyl-ACP formed in the plastid to the cytosol; and acyltransferases.
Several strategies were applied to manipulate the activity of these thioesterases and desaturases enzymes in developing Brassica seeds. Increasing the activity of FatA, the enzyme responsible for the formation of C18:, by hydrolyzing the newly synthesized C18:1-acyl ACP in the plastid by the expression of a soybean FatA gene in B. napus cv. Westar produced seed oil with up to 10.1% C18:0 (Hitz et al., 1995). Expression of a FatA gene from mangosteen (Garcinia mangostana L.), which can accumulate up to 56% C18:0 in the seed oil, increased the C18:0 level of B. napus cv. Quantum to more than 22% (Hawkins and Kridl, 1998). Expression of a mutated mangosteen FatA from site-specific mutagenesis led to a 55 to 68% increase in the C18:0 level compared with the wild-type FatA version (Facciotti et al., 1999).
A second strategy is the overexpression of 9-desaturase, the enzyme which directs the carbon flux to C18:1-ACP production. A reduced 9-desaturase activity should result in higher C18:0 accumulation (Knutzon et al., 1992). Antisense suppression using B. rapa 9-desaturase gene increased the C18:0 level to greater than 32% in transgenic B. rapa and to 40% in B. napus (Knutzon et al., 1992), although sense suppression with a soybean 9-desaturase was not as effective in B. napus (Hitz et al., 1995).
A third strategy is simultaneous manipulation of the activities of the two enzymes. Overexpression of FatA thioesterase and downregulation of 9-desaturase increased the C18:0 level up to 45%, higher than separately expressing the FatA thioesterase transgene (11% C18:0) and the 9-desaturase transgene (13% C18:0) (Töpfer et al., 1995). Down-regulation of both FatA and FatB in B. napus and B. juncea with a dual silencing construct containing inverted repeats of the target genes was initiated (Pandian et al., 2004). An additional approach for developing high C18:0 Brassica oil could be the downregulation of the activities of both 9 and 12 desaturases, as shown in cottonseed engineered with a hairpin RNA silencing constructs for the two desaturases which increased cottonseed C18:0 level from 2 to 3% to 40% (Liu et al., 2002).
Very High Oleic Acid
Considerable progress has been made in developing very high C18:1 oilseed by engineering of the 12-desaturase and FatB thioesterase. Transgenic B. napus could accumulate as high as 89% C18:1 with the PUFA fraction being reduced in the seed oils by sense or antisense 12-desaturase constructs (Hitz et al., 1995; Kinney, 1994; Stoutjesdijk et al., 2000). Transgenic B. juncea engineered with a 12-desaturase antisense gene of B. rapa had the C18:1 level increased from 53 to 73% (Sivaraman et al., 2004).
Super High Erucic Acid
Super high erucic acid rapeseed (SHEAR) oil with a greater than 80% C22:1 level is desired to reduce the cost of producing this fatty acid and its derivatives as a renewable, environment friendly industrial feedstock (Leonard, 1994; Mietkiewska et al., 2004; Taylor et al., 2001). Existing high erucic acid rapeseed (HEAR) cultivars have less than 1% C22:1 incorporated into the central position (sn-2) of the glycerol backbone because of the poor affinity of the rapeseed acyltransferase LPAAT for very long chain fatty acids, including C22:1 (Brough et al., 1995; Frentzen and Wolter, 1998; Lühs et al., 1999). This restricts the level of C22:1 in the existing HEAR seed oil to the theoretical limit of 66 mol%, while the maximum expression is closer to 60% in commercially produced spring habit HEAR cv. (McVetty and Scarth, 2002).
Increasing the C22:1 level of HEAR cv. has been attempted by manipulating acyltransferase and fatty acid elongase expression (Katavic et al., 2001; Mietkiewska et al., 2004; Taylor et al., 2001; Zou et al., 1997). Expression of a meadowfoam (Limnanthes douglasii R. Br.) LPAAT in high C22:1 B. napus plants resulted in trierucin accumulation (Brough et al., 1996; Lassner et al., 1995; Weier et al., 1997), but it did not increase the total C22:1 level in HEAR oil. No correlation was found in a separate study between the level of protein and fatty acid distribution, and trierucin accumulation did not correlate with the 22:1 levels at sn-2 (Wilmer et al., 2000). Expression of a yeast sn-2 acyltransferase increased the C22:1 of the high C22:1 rapeseed cv. Hero by 2.4 to 7%, and increased incorporation of C22:1 into the sn-2 position was confirmed (Katavic et al., 2000; Zou et al., 1997).
The failure to significantly increase the C22:1 level by engineering LPAAT could be due to a limitation in the acyl-CoA pool in the cytosol, which is required to support high levels of trierucin synthesis (Lühs et al., 1999). This hypothesis was supported by the increase of the C22:1 level to 48 to 53% in transgenic Hero plants expressing the yeast FAE1 compared with the wild-type control lines average of 43% C22:1 (Katavic et al., 2000). Similar results were obtained with the expression of Arabidopsis FAE1 in Hero (Katavic et al., 2001). The potential of using FAE genes from high-C22:1 accumulating species to increase C22:1 accumulation in Brassica seed oil was demonstrated by a 90% increase in the C22:1 level with the expression of a nasturtium (Tropaeolum majus L.) FAE gene in Arabidopsis of (Mietkiewska et al., 2004). On the basis of these studies, the proportion of 22:1 in rapeseed oil is limited by both C22:1 synthesis and its subsequent incorporation into TAG (Katavic et al., 2000). SHEAR oil could eventually be produced by combining these and other genetic modifications.
Other Novel Fatty Acids
"Novel" or "unusual" fatty acids are defined broadly as fatty acids that have chemical structures different from those commonly found fatty acids in major oilseed crops (Jaworski and Cahoon, 2003). More than 200 different types of fatty acids have been identified in the plant kingdom, with the majority accumulated as major components in the seed oil of nonagronomic species (Jaworski and Cahoon, 2003; Thelen and Ohlrogge, 2002; van de Loo et al., 1993). Many of these unusual fatty acids have industrial values because of their carbon chain length, number or position of double bond, or special functional groups such as hydroxy and epoxy fatty acids. Because the original source plants typically have limited agronomic potential (Cahoon, 2003), engineering existing oilseed crops with the genes isolated from these species has been a major research focus for the commercial production of these novel fatty acids (Cahoon, 2003).
Unusual Monoenoic Acid
Unusual monounsaturated fatty acids are produced by special desaturases which insert the double bond into an unusual position in the acyl chain, rather than between carbons 8 and 9 as seen in the common fatty acid C18:1. Plant oils rich in petroselinic or palmitoleic acid could be used as alternatives to petroleum in the production of biodegradable lubricants, surfactants, and plastic precursors (Ohlrogge, 1994). Desaturases that introduce double bonds at the D4, D6, or D9 position of C18:0-ACP have been identified from a number of plant species producing seed oils with up to 80% unusual monoenes (Cahoon et al., 1992, 1994, 1998). Expression of several of these desaturases in oilseed plants or in Arabidopsis resulted in the accumulation of the desired monoenes in the seed oils at the levels of 5 to 15% (Salas and Ohlrogge, 2002). Expressing specific thioesterase, ACP, KAS, and acyltransferases could be necessary to produce higher yields of the targeted monoenes (Salas and Ohlrogge, 2002; Thelen and Ohlrogge, 2002).
Gamma-Linolenic Acid
Gamma-linolenic acid (GLA) is a PUFA in the n-6 family of essential fatty acids. GLA is one of nutritionally important polyunsaturated fatty acids in human and animal diets. The majority of the cultivated oilseed crops do not produce GLA (Li et al., 2004). GLA-enriched oils from borage (Borago officinalis L.), evening primrose (Oenothera biennis L.), and black currant (Ribes nigrum L.) are expensive because of high costs of cultivation, seed harvesting, and oil extraction (Barre, 2001; Huang et al. 2004).
The key step in GLA production is the insertion of a double bond between carbons 6 and 7 of C18:3, a reaction mediated by 6-desaturase. The first report on engineering plants to produce PUFA was the transformation of tobacco (Nicotianum spp.) plants with a cyanobacterial 6 desaturase for the production of GLA (Reddy and Thomas, 1996). Considerable progresses have been reported recently in engineering oilseed crops for GLA production including B. napus, B. juncea, and also soybean. Coexpression of a M. alpina 6-desaturase and a 12-desaturase gene in a low C18:3 B. napus resulted in the generation of GLA at a level of greater than 40% (Huang et al., 2004; Knutzon et al., 1999a; Liu et al., 2001). Expression of a Pythium irregulare Buis. 6-desaturase gene in B. juncea also generated 25 to 40% GLA in the seed oil (Hong et al., 2002).
Very-Long-Chain Polyunsaturated Fatty Acids
VLCPUFA have 20 or 22 carbon atoms with four to six interrupted double bonds, including fatty acids with important therapeutic and nutritional benefits in humans, such as arachidonic (ARA), eicosapentaenoic (EPA), and docosahexaenoic acid (DHA) (Abbadi et al., 2004; Huang et al., 2004). Reports on the health benefits have resulted in increased use of these long-chain PUFAs as nutritional supplements in the past few years (Huang et al., 2004). Brassica species synthesize the very long chain fatty acid C22:1, but like other higher plants, do not produce very long chain PUFAs such as ARA, EPA and DHA (Harwood, 1996). GLA and ALA are not widely encountered in higher plants (Alonso and Maroto, 2000). To develop oilseeds for ARA and EPA, at least three genes have to be introduced (Abbadi et al., 2004; Huang et al., 2004). These enzymes catalyze the desaturation of LA, followed by an elongation, and a subsequent desaturation, respectively. The feasibility of engineering ARA pathway was demonstrated in tobacco with PUFA genes from various organisms, including algae, moss, lower animal, and plants (Huang et al. (2004).
Conjugated Fatty Acids
Conjugated fatty acids are polyunsaturated fatty acids with double bonds which are not separated by a methylene unit. Oils rich in conjugated fatty acids (such as calendic acid) have superior properties as drying oils in coating applications. Expression of a gene from pot marigold (Calendula officinalis L.) encoding an enzyme that introduces conjugated double bonds into polyunsaturated fatty acids resulted in the accumulation of calendic acid to 20 to 25% of the total fatty acids in soybean oil (Cahoon et al., 2001).
Epoxy and Hydroxy Fatty Acids
Epoxy fatty acids such as vernolic acid, are produced by monooxygenases and divergent forms of di-iron desaturases (Hatanaka et al., 2004). These fatty acids are valuable raw materials for the production of resins, glues, plastics, polymers, and other surface coatings. Currently, epoxy fatty acids are derived by the chemical epoxygenation of highly unsaturated vegetable oils, such as soybean and linseed oils, or by synthesis from petrochemicals (Singh et al., 2000). Expression of an epoxygenase gene from Stokesia laevis L., driven by a seed-specific phaseolin promoter in Arabidopsis plants, led to accumulation of vernolic acid in the seed oil but at low levels (Hatanaka et al., 2004). Expression of a 12-expoxygenase from Crepis palaestina (Boiss.) Bornm., a species accumulating up to 60% epoxy fatty acid, produced seed oil with a small amount of epoxy fatty acid in Arabidopsis. Coexpression with a 12-desaturase resulted in a twofold increase in the epoxy fatty acid level (Singh et al. 2000). However, expression of a rat medium chain hydrolase gene in Brassica did not change the fatty acid profile (Safford et al., 1993).
Expression Level and Stability of Transgenic Traits
The expression level and stability of the trait created by transgenic modification of fatty acid profiles are important determinants of the feasibility of this approach. Several factors, including genetic background, genomic position and copy number of the transgenes, the coexistence of other transgenes, and growth conditions, have been confirmed to affect the target fatty acids in thioesterase transgenic B. napus (Tang and Scarth, 2004; Tang et al. (2003, 2004). In these studies, B. napus lines transformed with TE genes from Bay, Cuphea, nutmeg, and elm (Voelker et al., 1996; Jones et al., 1995) as the source of the higher saturate levels were crossed to nontransgenic lines with distinct fatty acid profiles. The effect of genetic background on the expression of the target fatty acid was evident in the comparison between the F1 from the crosses with the low C18:3 cv. Apollo and the high C22:1 cv. Mercury. The mean C12:0 level in the F1 with Apollo was 37% compared with 26.2% in the F1 of the cross with Mercury. There was no difference between the canola and low C18:3 crosses so the difference was concluded to be associated with the competitive effect of the long chain elongation pathway which is blocked by the "canola" mutation to produce the low C22:1 trait in both the canola and the low C18:3 lines.
Doubled haploid (DH) lines were developed from the microspores of the F1 plants and the DH plants with the TE in homozygous condition were compared between the crosses. The low C22:1 DH lines produced higher levels of the target fatty acids than the high C22:1 lines, confirming the competitive effect of the elongation pathway to produce C22:1 reduces the accumulation of the medium chain saturated fatty aids.
The interaction between the transgenes was also investigated to determine the effect on the target fatty acid accumulation. Coexpression was demonstrated, with the accumulation of C12:0 and enhanced levels of C16:0. For example, transgenic seeds coexpressing the Bay TE and the Cuphea TE, produced 17.6% C12:0 and 16.5% C16:0. The transgene copy number and genomic position affected the level of accumulation of the target fatty acids in DH lines carrying the Cuphea and the elm TE (Tang et al., 2004). Target fatty acid levels were affected by temperature during seed development. DH lines carrying the elm TE or the Cuphea TE grown under high temperatures 25/20°C showed higher levels of C16:0 than the plants grown under lower temperatures (Tang and Scarth, 2004). The influence of the environment on oil quality is therefore an important consideration in the development of modified oilseed varieties and their commercial production.
Summary
There are both benefits and challenges to the modification of Brassica oil with conventional and transgenic approaches to achieve new oil profiles. The choice of approach is determined by the desired profile and the most efficient method of creating the desirable oil quality. Conventional breeding approaches are limited to the potential of natural and induced variation in the pathways involved in the biosyntheses of the existing fatty acids within the species or in close relatives. To date, several types of Brassica oil with altered fatty acid profile have been developed for edible and industrial applications. These variations have been associated with altered enzymes in the biosynthetic pathway and biochemical analysis. Inheritance and gene mapping studies have allowed the manipulation of these genes in subsequent cultivar development. The transgenic approach has the advantage of introducing novel fatty acids or manipulating the expression of existing fatty acids to levels outside the range that occurs with conventional breeding. Considerable progress has been made in the expression of unusual fatty acids in Brassica oil, as a result of transformation with transgenes from nonagronomic plant species or isolated from other organisms. With this technology, the source of the transgenes is not limited to the genes coding for naturally occurring enzymes in the fatty acid pathway. There is potential for the production of fatty acids which do not occur in nature such as 12-carbon nylon monomers with a modified desaturase enzyme. There are still limitations to the transgenic approach, including the ability to achieve the same level of expression of the fatty acids targeted by the transgenes as occurs in the source plant species. Further progress in modifying Brassica oil will occur with the ability to regulate multiple enzymes including those involved in the upstream and downstream steps as well as the enzyme activities which directly influence the formation of the desired fatty acids. Finally, the same three determinants of successful adoption of a modified oil quality, function, nutrition, and economics, will apply to the oils produced by either the conventional or transgenic approaches.
Received for publication August 11, 2005.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONREFERENCES |
|---|
6 hexadecenoic acid is synthesized by the activity of a soluble 6 palmitoyl-acyl carrier protein desaturase in Thunbergia alata endosperm. J. Biol. Chem. 269:27519–27526.8, 10–double bonds by 12–oleic acid desaturase related enzymes: Biosynthetic origin of calendic acid. J. Biol. Chem. 276:2637–2643.-linolenic acid in Brassica juncea using a 6 desaturase from Pythium irregulare. Plant Physiol. 129:354–362.-linolenic acid. p. 61–71. In Y.S. Huang and V.A. Ziboh (ed.) -Linolenic acid: Recent advances in biotechnology and clinical applications. AOCS Press, Champaign, IL.6–desaturase gene results in -linolenic acid production in transgenic plants. Nat. Biotechnol. 14:639–642.[CrossRef][Web of Science][Medline]12–desaturases. Biochem. Soc. Trans. 28:938–940.This article has been cited by other articles:
|
|
 |
|
  A. Nabloussi, J. M. Fernandez-Martinez, and L. Velasco Inheritance of Low Linolenic Acid Content in Zero-Erucic Acid Ethiopian Mustard Crop Sci., March 17, 2009; 49(2): 549 - 553. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| The SCI Journals | Agronomy Journal | Vadose Zone Journal | |||
| Journal of Natural Resources and Life Sciences Education |
Soil Science Society of America Journal | ||||
| Journal of Plant Registrations | Journal of Environmental Quality |
The Plant Genome | |||
