|Many butter lovers transferred their allegiance to vegetable-oil margarines in recent decades as health professionals campaigned about the heart risks posed by eating too much animal fat. In the past few years, however, even margarine has taken a public relations beating for loading our diets with another type of potentially unhealthy fat -- one known as trans fatty acid.
Fairly uncommon in nature, trans fats form during the partial hydrogenation of vegetable oils (SN: 3/6/93, p. 150) that manufacturers use to turn the liquids into solids at room temperature.
So with the sticks of both spreads under the nutritional gun, where's a health-conscious bread-and-butter lover to turn?
For now, most nutritionists suspect that those soft "tub" margarines offer the safest bet. They're low in both saturated and trans fats -- which explains why they're so easy to spread, even straight out of the fridge. But diners who still long for the old-fashioned baking and sandwich attributes of stick margarines can take heart. Genetic engineering is designing a host of new options for vegetable-oil spreads that will remain hard without the need for trans fats.
The trick is to tinker with the genetic machinery of plants that make oil-rich seeds. In temperature zones, such plants -- soybeans, cotton, corn, and canola, for example -- tend to make liquid oils. But the difference between a fat that is liquid at room temperature and one that is solid can be as little as the presence or placement of a double bond between two carbon atoms in the backbone of a fat molecule.
The plant world's simplest fats are saturated ones, which have no double bonds in the backbone. Structurally, these fats resemble centipedes whose bodies have been constructed of 8 to 20 segments. Each segment consists of a carbon atom from which a pair of hydrogen atoms emerge -- one on each side, analogous to a pair of legs.
Fats tend to be named for their source: safflower oil, peanut oil, chicken fat, olive oil, or butter. In fact, each of these represents a somewhat variable mixture of saturated, monounsaturated (one double bond), and polyunsaturated fats. Though these constituent fats each have a name, such as palmitic acid, oleic acid, and linoleic acid, chemists tend to refer to them using a numerical designation that reflects the number of carbon segments and the placement of any double bonds.
For instance, stearic acid, a saturated fat, goes by the name of 18:0, signifying that it has 18 linked carbons and no double bonds. Oleic acid, a monounsaturate, is the most common fatty acid in nature. It is alternately referred to as either:
These double bonds develop whenever two adjacent hydrogen atoms -- centipede legs -- have been removed. With these appendages missing, the structure becomes asymmetrical and bends at this point.
Because the skinny saturated fats are straight, they can pack tightly together. "And that makes [the saturates] physically solid," notes John Shanklin, a lipid biochemist at Brookhaven National Laboratory. Try to flex them, he notes, and they will instead snap. "But once you desaturate them, by introducing a double bond, they sort of kink." The inability of these bent molecules to pack together tightly transforms them from a solid into a liquid, he says. The body's membranes need a mix of these saturated and unsaturated fats to blend structure and fluidity for the performance of each cell's many functions.
A spate of new programs to produce unhydrogenated margarines are actively exploring the possibility of engineering temperate oil-seed plants to produce a solid fat instead of their predominantly unsaturated oil by moving or eliminating these double bonds.
Most of these research efforts have focused on modifying fats in the cultivated rapeseed plant (Brassica napus), which in North America is best known as canola. This cool-season crop thrives in dry prairies, like those that span the upper Midwest and central to western Canada. A relative of broccoli, farmers grow this plant for its seeds, which are 40 percent fat.
Canola's oil contains the lowest concentration of saturated fat -- 6 percent -- of any commercially marketed vegetable oil. It also serves as an especially rich source of oleic acid (58 percent). Several studies have suggested that diets high in monounsaturates, especially oleic acid, reduce an individual's risk of heart disease. In contrast, just 30 percent of butter is monounsaturated and 66 percent of its fats are saturated -- the type most likely to raise a diner's serum cholesterol concentrations.
Calgene Inc. of Davis, Calif., is actively redesigning canola plants to make less of those oily unsaturates. One potential cultivar currently undergoing second-generation greenhouse trials shows great promise of producing what Toni A. Voelker terms "naturally" hardened vegetable oils.
A lipid biochemist with Calgene, the biotech firm that pioneered genetically engineered foods with the Flavr Savr tomato (SN: 11/28/92, p. 376), his idea of what constitutes natural may not win universal acceptance. But semantics aside, Calgene's new canola work certainly borrows heavily from nature.
Plants and animals employ a series of enzymes to fashion their fats. One family of these enzymes, which goes by the unwieldy name of thioesterase, allows fats to lengthen or to gain desaturating double bonds -- but only as long as these enzymes are bound to the developing fat. The moment the thioesterase bond is clipped, this phase of a fat's tailoring ends.
Calgene's genetic engineers have identified the oleic thioesterase that allows most of a canola oil's constituent fatty acids to grow to 18-carbon-long chains and then to receive a double bond. They also found a related thioesterase that lets go of the fat a little sooner -- before a double bond can be inserted. What forms in its grip is the 18:0 stearic acid, a fat usually found in meats and cocoa butter.
Though saturated, stearic acid is unusual in that it is the only such fat whose consumption doesn't raise serum cholesterol (SN: 5/21/88, p. 332). As such, some studies have suggested that it should be viewed, health-wise, as somewhat comparable to unsaturated fats like oleic acid. This unusual saturated fat is the one that Calgene now proposes loading into canola.
Ling Yuan, a lipid biochemist with the firm, has compared the structure of the two 18-carbon thioesterases and found that by changing only two amino acids, he can turn one into the other. His team has gone on to insert the gene for the stearic-acid-making enzyme into canola.
Normally, only about 1 percent of canola's fat is stearic acid. To produce margarine, Voelker says that stearic's proportion of the oil must skyrocket to about 35 percent. For use as shortening -- which is softer than margarine -- he observes that perhaps as little as 30 percent stearic acid will do.
Currently, Calgene has engineered canola that, as second-generation greenhouse plants, reliably produce 30 percent of their fat as stearic acid. "So if we could get that [to hold] over several generations in the field," Voelker says, "it would be a product for shortening." Overall, he says, Calgene hopes to have a high-stearic-acid canola plant ready for marketing by the year 2000.
Though stearic acid does not raise serum cholesterol, its image has tarnished a bit, in response to a controversial British study published last year. It indicated that this fat may promote the fatty buildup of artery-clogging plaque -- independent of any effects on cholesterol (SN: 8/10/96, p. 87).
So individuals who would prefer a margarine free of even this saturate may find comfort in the efforts of researchers at Michigan State University.
Several years ago, while he was studying there, Edgar B. Cahoon isolated a gene from coriander seeds that controls the production of another fat-tailoring enzyme. Known as a desaturase, it cuts off two of the centipedes' legs in 18:0 fats, inserting a double bond after the sixth carbon. In other words, it triggers the production of an 18:delta-6 fat. Called petroselenic acid, this fat is, like oleic acid, a monounsaturate. Unlike the 18:delta-9 oleic, however, it's solid at room temperature.
Since coriander is not an oil-seed crop, Cahoon's team, led by lipid biochemist John B. Ohlrogge, has been working to insert the gene for petroselenic into both soybeans and canola -- plants that ordinarily don't make this fat. The first bioengineered crops were sown about 3 years ago. "We have found that we can produce some petroselenic acid in the transgenic plants," Ohlrogge told Science News Online, "but the levels right now -- about 5 percent -- are not high enough to be commercial." In fact, he suspects the plants will only look economically attractive if they can boost their petroselenic production to about 80 percent of an oil-seed's fat.
"I'm confident that we'll be able to do that," Ohlrogge says. "Maybe not tomorrow, but within the next few years." His data now suggest that prompting canola to greatly increase its petroselenic output will require tinkering with at least one more gene -- "which may be a common story in genetic engineering of plant oils," he says: "The challenge will be finding out what combination of genes are needed."
While petroselenic-rich soy or canola would permit the production of margarine straight from seed, "we really don't know how the Food and Drug Administration will view this," Ohlrogge admits. People who ingest the seeds of carrots, celery, and fennel already get small amounts of this fat in the diet. However, he notes that additional feeding studies with animals will have to been undertaken to ensure that eating larger amounts will be safe.
Even if petroselenic-rich oilseeds don't pan out as a new source of margarines, Ohlrogge suggests, these new cultivars may still prove commercially attractive -- for instance, as a renewable alternative to petroleum as a source of fats used in the production of nylon.
Cahoon has since moved on to Brookhaven National Laboratory, where he is working with Shanklin on the scientific underpinnings of a related strategy. They've begun isolating the individual amino acids that determine what length fat a desaturase enzyme will act upon or where within a given-length fat that enzyme fashions a double bond.
In the May 13 Proceedings of the National Academy of Sciences, they just reported swapping 5 of the 360 amino-acid building blocks in one desaturase to make it mimic the actions of another -- even though that other enzyme has 95 additional amino-acid differences (A giant step toward creating better fats).
Their ultimate goal is not just to swap enzymes from one plant to another, or to mimic the function of existing enzymes, but to create truly novel ones that have the attributes they want --such as the production of fats that are solid at room temperature -- but that may not exist in nature. One potential advantage for the food industry, Shanklin says, is that these arcane creations may evade recognition by key elements of the digestive tract.
In the body, fats tend to get packaged in threes onto a glycerol backbone. This complex -- called a triglyceride -- shuttles fats around the body. But to absorb these fats, Shanklin notes, they must be cut apart from the glycerol. If the enzymes that the body uses don't recognize the triglyceride's constituent fats because double bonds are in novel locations, "then you would not absorb it," he says, essentially giving you "a diet margarine."
Adler, T. 1994. Tomato biotechnology heads for the market 145(May 28):342.
Cahoon, E.B, ... and J. Shanklin. 1997. Redesign of soluble fatty acid desaturases from plants for altered substrate specificity and double-bond position. Proceedings of the National Academy of Sciences 94(May 13):4872.
Raloff, J. 1996. Unusual fats lose heart-friendly image. Science News 150(Aug. 10):87.
_____. 1995. New trans fat studies muddy the waters. Science News 147(Feb. 25):127.
_____. 1994. Margarine is anything but a marginal fat. Science News 145(May 21):324.
Yuan, L., T.A. Voelker, and D.J. Hawkins. 1995. Modification of the substrate specificity of an acyl-acyl carrier protein thioesterase by protein engineering. Proceedings of the National Academy of Sciences 92(November):10639.
Edgar B. Cahoon
Brookhaven National Laboratory
Upton, NY 11973
Canola Council of Canada
400-167 Lombard Avenue
Winnipeg, Manitoba R3B 0T6
Canola Information Service
ASA 116 103rd Street, E.
Saskatoon, Saskatchewan S7N 1Y7
Phone: 306-387-6610; FAX: 306-387-6637
John B. Ohlrogge
Department of Botany
Michigan State University
East Lansing, MI 48824
Brookhaven National Laboratory
Upton, NY 11973
Toni A. Voelker
1920 Fifth Street
Davis, CA 95616
This week's Food for Thought is prepared by Janet Raloff, senior editor of Science News.
Illustration: Wendy Temple.