Lipids

In most commercial seed oils, more than 95% of the mass can be attributed to only a few fatty acids —
specifically, lauric, mirystic, palmitic, stearic, oleic, linoleic, and linolenic acids (Murphy, 1996, 1999;
MacKenzie, 1999). Plants produce a wide diversity of fatty acids, the majority of which accumulate in the
seed as triacylglycerols. Vegetable oils generally are preferred to oils and fats from other sources
because of their higher content of mono- and polyunsaturated fatty acids (Figure 1.1). Unsaturated fatty
acids are healthier than saturated fatty acids, and the monounsaturated form, oleic acid, is also more
stable in frying and cooking applications than are the polyunsaturated forms, linoleic and linolenic
(MacKenzie, 1999; Napier et al., 1999).

It is now possible to engineer “designer” oilseeds tailored to specific end uses. Plant lipids have different
end uses, from industrial applications such as lubricants and detergents to food and nutrition (Murphy,
1996, 1999). They also have important roles as nutraceuticals and pharmaceuticals. The global market
for plant-derived oils is immense; consequently, the major biotechnology companies have driven much
of the research.

Saturated Fatty Acids

Laurate, which is used in confectionery, is normally obtained from either coconut or palm oil. Although
both plants yield relatively high levels of the fatty acid, they are limited in their agricultural utility.
Research work carried out at Calgene Company has demonstrated the feasibility of engineering canola
to produce lauric acid by introducing the gene encoding a lauroylspecific acyl-carrier protein (ACP),
thioesterase, from the California bay plant. This enzyme causes premature chain termination of the
growing fatty acid, resulting in the accumulation of lauric acid, a 12-carbon saturated fatty acid, rather
than normal C18 oils (Figure 1.1). Canola normally contains less than 0.1% lauric acid; however, with the
latter approach, Calgene has produced a number of canola lines that accumulate around 40% laurate
(Voelker et al., 1992). The novel fatty acids are recovered from transgenic canola by standard processing
methods. These oils, trivially named canola laurate, are now marketed as a partially hydrogenated
vegetable oil for use in confectionery under the trade name Laurical.

This work has also demonstrated the feasibility of producing large amounts of transgene-modified plant
oils to supplement or replace fluctuating natural sources (Murphy, 1996, 1999; MacKenzie, 1999).
Saturated fatty acids are relatively uncommon in most plant storage lipids because of the presence of
highly active desaturases in developing seeds. The composition of canola oil has been modified by
expressing an antisense stearoyl-ACP desaturase cDNA, under the control of a napin or ACP promoter, in
seeds of Brassica rapa and B. napus. The activity and amount of stearoyl-ACP desaturase were greatly
reduced, resulting in a marked increase in stearic acid from less than 2% to as much as 40%; oleic acid
levels also decreased (Knutzon et al., 1992).

In other experiments, the cloning of the genes encoding each of the soybean fatty acid desaturases
enabled the cosuppression of the seed-specific desaturase, causing seed oleic acid levels to rise from
25% in nontransformed lines to 85% in transgenic soybean (Mazur et al., 1999). The cosuppressed,
seedspecific, high-oleic-acid, transgenic soybean plants showed excellent agronomic properties.
Unsaturated Fatty Acids

It is well documented that the risk of coronary heart disease is positively correlated with elevated levels
of high-density lipoprotein (LDL) cholesterol (Kishore and Shewmaker, 1999; MacKenzie, 1999). The C12,
C14, and C16 saturated fatty acids are known to induce hypercholesterolemic effects and to increase the
proportions of LDLs in the bloodstream, although they differ in their impact (MacKenzie, 1999). The
intake of saturated fatty acids has been shown to be closely related to the clotting activity of platelets
and their response to thrombin-induced aggregation (MacKenzie, 1999). Consequently, there has been,
and continues to be, a parallel research effort to improve the quality of vegetable oils by reducing the
total amount of saturated fatty acids (Kishore and Shewmaker, 1999; MacKenzie, 1999).

Unsaturated fatty acids have recently attracted interest as targets for genetic manipulation (Figure 1.1)
(Napier et al., 1999). Polyunsaturated fatty acids (PUFAs; long-chain fatty acids containing two or more
double bonds introduced by specific desaturase enzymes) have entered the biomedical and
nutraceutical areas as a result of the elucidation of their biological role in certain clinical conditions
(Napier et al., 1999). Besides pharmaceutical applications, public perceptions of healthy foods and
lifestyles have also brought them to the attention of the consumer. Specifically, interest is growing in 18-
to 22-carbon PUFAs containing three or more double bonds, as these compounds have emerging
therapeutic roles (MacKenzie, 1999; Napier et al., 1999). Current commercial production of PUFAs,
largely from seed oils, marine fish, and certain mammals, is considered inadequate for the future PUFA
market. However, the seeds of commonly cultivated agricultural plants contain a rather limited range of
fatty acids that is truly representative of the diversity of unusual fatty acids accumulated predominantly
as storage lipids by plants and fungi (Murphy, 1999). Some of these less common unsaturated fatty acids
are highly valuable, although the organisms from which they originate are generally unsuitable for large-
scale production and cultivation (Murphy, 1996, 1999). For example, linoleic and linolenic acid PUFAs
cannot be synthesized de novo by animals; they are therefore classified as essential and must be
ingested (MacKenzie, 1999; Napier et al., 1999). Furthermore, for metabolic purposes, the two families
cannot be interconverted. Arachidonic acid is synthesized by the elongation of g-linolenic acid to a
longer chain PUFA, a 20-carbon molecule, followed by desaturation at the D5 position (Figure 1.1). Both
PUFAs are precursors of a group of short-lived regulatory molecules called the eicosanoids. The
eicosanoids (prostaglandins, leukotrienes, thromboxanes) have many functions and are particularly
important in reproductive function and the regulation of blood pressure (Horrobin, 1990). The ability,
therefore, to produce crop plants accumulating fatty acids such as arachidonic acid and other PUFAs is
an obvious biotechnological target (Figure 1.1) (Murphy, 1996, 1999; MacKenzie, 1999; Napier.

Transgenic plants expressing genes that encode fatty acid desaturases have been recently generated.
Thus, an animal (Caenorhabditis elegans) microsomal fatty acid desaturase was expressed in
Arabidopsis, resulting in increased levels of a-linolenic acid (Spychalla et al., 1997). Moreover, these
same transgenic plants desaturated exogenously applied 20-carbon PUFAs, such as arachidonic,
indicating that the enzyme was a D15/D17 desaturase (e.g., w-3) with specificity for both 18- and 20-
carbon fatty acids. On the other hand, a Mortierella alpina desaturase with an amino-terminal
cytochrome b5-domain was shown to exhibit D5–fatty acid desaturase activity, converting exogenously
supplied di-homo-g-linolenic acid (D8,11,14-C20:3); also, transgenic canola was obtained resulting in the
expression of a number of unusual, D5- desaturated 18-carbon PUFAs (Michaelson et al., 1998; Napier
et al., 1999). g-Linolenic is a registered pharmaceutical with wide applications for antiviral and cancer
therapies; unfortunately, it only accumulates in a limited number of species, including evening primrose,
borage, and amaranth. In 1997, Sayanova et al. were the first to isolate a cDNA from borage plant
encoding the D6–fatty acid desaturase responsible for synthesis of this fatty acid and to introduce it into
tobacco (used as a model crop). It resulted in the accumulation of g-linolenic at levels equivalent to or
greater than those found in borage (Figure 1.1). The final goal, the production of novel highvalue
pharmaceutical oils and eicosanoid precursors, may be difficult to achieve but will probably require
transgenic oilseed producing both the D5 and D6 desaturases and other as yet undefined enzymes
(Napier et al., 1999).

Reduction of Saturates

The principal saturated fatty acids in vegetable oils are palmitate and stearate. Because these two fatty
acids are generally not equivalent with respect to dietary effects or food functionality, there is
widespread interest in being able to control the relative amounts of these fatty acids. There has been
progress in at least three complementary strategies for meeting this objective.

One goal is to regulate the relative amounts of stearate and palmitate. A theoretical strategy for
reducing palmitate and stearate levels in some oils is to enhance the extent of unsaturation by
introducing genes for novel desaturases. There are now many examples in which heterologous genes
from plant, animal, yeast, and bacterial origins encode a functional protein when expressed in
transgenic plants (12, 38). Examples of the use of this technology in the field of plant lipid engineering
are the transfer into transgenic tobacco of the genes encoding the yeast and animal stearoyl-CoA
desaturases (26, 76). The significance of these experiments was that in most plants, the plastid stearoyl-
ACP desaturase is the only enzyme capable of converting stearate to oleate. Thus, the ratio of 18:0 to
18:1 + 18:2 + 18:3 in TAG is thought to be regulated by competition between the plastid enzymes
stearoyl-ACP thioesterase and stearoyl-ACP desaturase. Once stearate is hydrolyzed from ACP and
exported from the plastid, it is not normally a substrate for desaturation. By contrast, in yeast and
mammals, stearate is desaturated in the ER by stearoyl-CoA desaturase. Thus, by expressing the yeast or
animal stearoyl-CoA desaturase in plants, it is possible to provide a second opportunity to desaturate
stearate that has reached the ER. A second potentially useful effect of expressing stearoyl-CoA
desaturases in plants was that the level of palmitoleate (16:1n-9) increased and the level of palmitate
decreased. This is due to the fact, noted elsewhere, that palmitate is a poor substrate for plant stearoyl-
ACP desaturase but is a good substrate for stearoyl-CoA desaturase. This may be useful for decreasing
the level of palmitate in edible oils.

Decreasing the Oxidation Potential of Plant Fatty Acids

Kubow (56) has noted that PUFAs and plant sterols, when exposed to oxidative stress either thermally or
through aeration, may form potentially toxic oxidized products that can have powerful atherogenic
effects (62, 75). Because frying subjects vegetable oils to lengthy heat treatments, and light-induced
oxidation prevents their long-term storage, a major objective of oil crop engineering is to decrease the
PUFA content in oil crops such as soybean and rapeseed to improve oxidative stability and enhance shelf
life of the oil.

Linoleic acid (18:2n-6), because of its low-density lipoprotein (LDL) lowering effect, has been the main
PUFA used in test diets (55). Although metabolic and epidemiological studies of humans and laboratory
animals generally support the concept that higher intake of PUFAs are beneficial in terms of lipoprotein
metabolism and cardiovascular health (30), the consumption of high amounts of PUFA when
insufficiently protected by antioxidants has led to enhanced susceptibility of membrane lipids to
peroxidation (95). Many studies have shown that dietary supplementation with either n-6 or n-3 PUFA is
associated with increased susceptibility of tissue lipids to peroxidation both in vivo and in vitro.

In view of all the negative effects of lipid oxidation, vegetable oils with reduced levels of PUFAs and a
higher content in MUFAs have become highly desirable. Nutritionally, oleic acid (18:1) appears to have
the same LDL lowering effect as linoleic acid and is not as susceptible to in vivo oxidation as linoleic acid
(55). A higher ratio of oleic to linoleic acid is inversely correlated with the peroxidation rate of plasma
LDL. Because of its low cost, soybean oil is widely used for frying. Thus, researchers have sought to
increase the ratio of MUFAs to PUFAs in soybean and canola oil by modifying the activity of a
microsomal membrane-bound oleate desaturase designated FAD2.

Eliminating Trans Fatty Acids

An ongoing trend has been to substitute solid fats derived from plant oils for animal fats. Industrial
processes have been developed to partially hydrogenate vegetable oils to form semisolid or solid fats.
Although hydrogenated vegetable oil has been the best nutritional alternative to saturated animal fats,
this process results in the formation of undesirable trans isomers of unsaturated fatty acids. The oils
with a high proportion of transisomers tend to have unfavorable effects on plasma concentrations of
total cholesterol, triglyceride, LDL, and highdensity lipoproteins (46, 55, 64, 104, 105). The average
dietary uptake of transfatty acids in western countries is estimated at 7–8 g per person per day, of which
about 80% are contributed by hydrogenated oils (22). A promising approach to reducing trans fatty acids
is to eliminate the need for hydrogenation. For some applications, this can be accomplished by using
very high–oleic sunflower oil. For example, several modified sunflower varieties contains 87–90% oleic
acid and only about 6.5% saturates.

A related objective of researchers working on modifying plant lipids has been to obtain transgenic line–
producing oils with a high stearate and low palmitate content; these oils do not require extensive
hydrogenation. Nutritionally, palmitate increases plasma LDL cholesterol. Stearate, on the other hand,
appears to have a neutral effect on total plasma cholesterol and LDL cholesterol (27, 55, 85). High
stearate content has been achieved in canola through antisense reduction of the expression of stearoyl-
CoA desaturase (23, 54). One major drawback of increasing the stearate content of oilseeds is that it
may have deleterious effects on seed development and germination (67). As a result, it may not be
possible to completely eliminate the use of hydrogenation in the production of solid fats from plant oils
derived from plants grown in temperate regions.
Production in Plants of Essential Long-Chain PUFA

Members of the n-3 and n-6 PUFA (α-linolenic acid and linoleic acid) are important constituents of cell
membranes (29) and serve as precursors for long-chain PUFAs with carbon lengths of C20/C22, which
subsequently form a variety of biologically active compounds known as eicosanoids (e.g. prostaglandins,
leukotrienes, and thromboxanes). These eicosanoids play a critical role in coordinating physiological
interactions among cells and influence a wide variety of functions, including those of the central nervous
system and the contractions of smooth muscles. They also inhibit the mobilization of fatty acids from
adipose tissue, have an antiinflammatory effect, and influence blood pressure, the aggregation of blood
platelets, and cardiac function. A potential need for dietary long-chain PUFAs (20:4n-6 and 22:6n-3) in
some groups, particularly newborn infants, is a major area of current research (41). Supplementation
with long-chain PUFAs in formula foods produces beneficial effects on cardiovascular disease, renal
function, and the development of normal function of the retina and brain (2, 37, 60a, 103). The
consumption of these long-chain PUFAs (C20/C22) also has beneficial effects in patients with chronic
inflammatory diseases (88), a history of colonic polyps (34), ulcerative colitis (4, 84, 94), and genetic
disorders of peroxisomal biogenesis (5, 83), such as Zellweger syndrome. One alternative to high intakes
of long-chain n-3 fatty acids to control chronic inflammatory diseases such as rheumatoid arthritis is
thought to be dietary supplementation with γ -linolenic acid (C18:3n-6) derived from the oils of evening
primrose and borage plants (42, 66).

Genes encoding 16 linoleate desaturases have been cloned from a variety of organisms (25, 68, 71, 80,
86). Expression of borage and Synechocystis genes in transgenic tobacco resulted in the accumulation in
leaves of substantial amounts of γ -linolenic and octadecatetraenoic acids (80, 86), a first step toward
the production of γ -linolenic acid, eicosapentaenoic acid, and arachidonic acid in crop plants.