16.2: Food Additives

Many substances in food weren’t there originally. Some are added intentionally (e.g., food additives), some are residues (e.g., pesticides sprayed on food crops), and some are contaminants (e.g., industrial pollutants in ground water taken up by plants).

People often see packaged food as unhealthful and full of additives. Food-safety experts don’t worry much about additives; food additives are very closely scrutinized before being approved. It could be argued that the most unhealthful thing about packaged food is that it’s so handy and appealing—so attractively packaged, easily stored, and convenient to eat, that it’s easy to overeat. How often would you eat potato chips if you had to make them yourself from a fresh potato? Or ice cream, if you had to start by milking a cow?

Some additives are listed in Table 16-1: Some are used in the kitchen, e.g., acetic acid (vinegar), sodium bicarbonate (baking soda). Some make food more nutritious, e.g., vitamin D added to milk, and safer, e.g., mold inhibitors. Some are “gums,” e.g., agar, carrageenan, guar, found naturally in plants and often used as stabilizers and thickeners in salad dressings and ice cream.

The most heavily criticized additives are coal tar dyes (named FD&C red no. 40, etc., because they’re used to color Food, Drugs, & Cosmetics). They, too, have been scrutinized for safety. We know much less about the safety of the many natural colorants, even though coloring our food isn’t new. Ancient Egyptians routinely made food colorants by grinding up colorful plants and insects.

Prior to 1958, just about anything could be added to food. It was up to the government to prove it was harmful before disallowing its use. Safety programs of the FDA began with Dr. Harvey Wiley, a chemist with the Dept. of Agriculture from 1883 to 1930. He used a volunteer “poison squad” of 12 men who tested food additives by eating a lot of the additive-containing food to see if it caused them any harm! The 1958 Food Additives Amendment turned things around by requiring that manufacturers do extensive safety tests before asking the FDA for approval.*

Risk assessments are made for cancer, nerve damage, birth defects, reproductive problems, etc., not only for food additives and processed foods, but for pesticide residues on fresh foods as well. Safety for children receives special emphasis.

For each proposed additive, males and females of at least two species of animals are studied for at least two generations. They are fed a range of doses over their lifetime, to find the highest dose that has no adverse effect (no-effect dose). Usually, the highest expected dose must not be more than one-hundredth of the no-effect dose, e.g., a no-effect lifetime dose of 100 mg/day means that the additive can’t be added in amounts over 1 mg/day. Getting a new additive approved is a major investment for a manufacturer. Very few additives have been approved since 1958.

GRAS List: At the time of the 1958 amendment, hundreds of additives (including salt, sugar, and some common spices) had already been in use for years without apparent harm, and were widely accepted as safe by scientists at that time. These were put on the GRAS (Generally Regarded As Safe) list, and the safety of each has since been reassessed. Their safety, as well as that of newer additives, continues to be reassessed in light of new scientific information—or when a public outcry to do so arises.

Since 1997, because of the long and expensive process of getting a new food additive approved, new rules have been introduced by which a manufacturer can, instead, have its additive recognized as GRAS by showing that its intended use complies with safety standards, etc., as determined by the expert scientific community. This streamlines the FDA’s ability to evaluate more and higher priority food additives.

^{*In contrast to food additives, dietary supplements don’t require risk assessments, nor FDA approval. It’s up to the government to prove harm before disallowing their use, as they did with tryptophan supplements (Chap. 15)—and with food additives prior to 1958.}

Assessing Risk

People worry most about carcinogens (cancer-causing agents). Identification often begins by testing substances in animal tissue cells or bacteria to see if they are capable of causing mutations (e.g., Ames test; see Chap. 12). Mutations don’t necessarily cause cancer, but if a substance can cause a mutation, it’s regarded as a potential carcinogen.

Typically, substances that pass screening tests are then tested in animals using doses that are much larger than what we’d normally eat. Large doses are used because, even if a substance is known to cause cancer, most such substances might cause only a few cancers when fed in small amounts. When cancer rates are low, large numbers of animals must be tested to find a statistically significant difference in the number of cancers between experimental and control groups. Using thousands of rats to test each substance over many years is expensive and impractical. So, high doses are given to fewer animals.

Table 16-1: Some Food Additives

Feeding a large dose (e.g., of an additive proposed for use on apples) is often ridiculed (it’s equivalent to eating a bushel of apples a day). The question is, if a substance causes cancer at high doses in animals, will it cause cancer at low doses in humans?

Scientific concerns about using a high-dose test to predict a low- dose effect include the fact that DNA-­repair systems and immune defenses may take care of minor damage caused by small amounts of carcinogens. Also, extremely high doses given to animals can cause adverse effects that don’t occur with normal doses. These can affect whether test animals develop cancer.

Even with extensive studies, risk estimates have to be made with incomplete knowledge. What’s harmful to animals may be harmless to humans, and vice versa. An additive might be harmful only to some people under only certain conditions. Some effects may not be measurable by current technology.

Because it’s considered better to err on the high side, estimates typically use worst-case scenarios, e.g., apples, lettuce, spinach, carrots, etc., aren’t washed, peeled, or cooked, in estimating ingestion of additives or pesticide residues. Knowing that estimates are based on “soft” numbers, it’s easy to see why different organizations with or without biases come up with such different “correct” estimates.

Because such uncertainties can never be completely addressed, risk assessment often is subjective, and scientists themselves argue about it. So industry, consumer and environmental groups, letters to congress, etc., can have more influence than scientific evidence in determining whether a particular food additive is approved.

“Macro-Additives”

A food additive, by definition, is something added to food. Should it be called something else if it makes up the bulk of a food? What if it can make up the entire food and has no nutritional value? Is it then neither a food nor an additive?

Olestra (Olean®) is such a “macro-additive.” It’s no longer used much as a food additive, but illustrates the complexity of such additives.

Olestra is a non-caloric fat substitute developed by Proctor and Gamble in 1968 and approved in 1996 for use in some snack foods (e.g., crackers, chips). Unlike any fat substitute on the market, it has all the cooking and sensual qualities of regular fat, without the calories. It can be liquid like salad oil, or solid like margarine. It can be used just as fat is used now—to fry french fries or donuts, to make salad dressing, pie crust, or chocolate cake with fudge frosting, though it hasn’t been approved for these additional uses.

Glycerol is the 3-carbon-sugar backbone of regular fat (triglyceride), with a fatty acid attached to each of the carbons (Figure 5.2). Olestra’s backbone is the double-sugar sucrose (Figure 5.1), which has 12 carbons—more fatty acids can be attached in a variety of positions.

Olestra is completely man-made; it doesn’t occur naturally. Chemically, it’s a fat, but we don’t get any calories from it because we can’t digest it —our digestive enzymes can’t split the fatty acids from its backbone, so it remains undigested and unabsorbed.*

Fat-soluble vitamins (and other fat-soluble substances such as carotene and some pesticides) that are in foods eaten at the same time can dissolve in olestra. So when olestra leaves the digestive tract, so do any fat-soluble substances that are dissolved in it. Fat-soluble vitamins are added to olestra to offset such losses.

This loss can be useful in some cases—fat-soluble substances like cholesterol and fat-soluble toxins can be carried out of the digestive tract this way. In fact, olestra was found to lower blood cholesterol, initially prompting Proctor and Gamble to seek its approval as a drug. As it turned out, it didn’t lower cholesterol enough to be useful as a drug.

An earlier version of olestra had a laxative effect—the same effect people seek when they take mineral oil for constipation. Colon bacteria don’t have the enzymes to digest olestra (or mineral oil), so the oily olestra remained to make stool slippery and easier to pass—fine if that’s what you want, not so fine if it isn’t.**

To overcome this problem, olestra was made more solid—less oily. Some people may still get a laxative effect when they eat a large amount (and we wouldn’t expect them to do this again) but, as mentioned in relation to lactose intolerance in Chapter 6, gastrointestinal (GI) upsets are so common that the cause often is hard to pinpoint.

It’s hard to get a regular food additive approved by the FDA. It was even more so for olestra, because many of the standard testing protocols weren’t applicable. Proctor and Gamble spent about $300 million and about 20 years doing studies on olestra, including extensive studies on rats, mice, pigs, and humans to show safety. The reports totaled more than 100,000 pages—a moving van was needed to take them to the FDA. As an additive, olestra broke new ground. For this reason, the FDA was extra cautious.

Initially, the FDA required that foods containing olestra (e.g., Fat-Free Pringles®) have a warning on the label that olestra may cause abdominal cramping and loose stools. The FDA later removed the label requirement, saying that the warning could cause consumers to blame olestra for digestive problems when olestra wasn’t to blame. In fact, double-blind studies of people fed regular potato chips vs. those fried in olestra show the same number of GI upsets in each group.

Olestra failed in the food marketplace. Other uses have been found, however. Because olestra can carry out fat-soluble substances in the digestive tract, it can be used to treat people who have been exposed to fat-soluble toxins. Also, because olestra is chemically a fat, it can be used industrially as lubricants, etc.

^{*Recall that enzymes are very specific. Digestive enzymes that remove fatty acids from triglycerides don’t work on olestra, just as those that digest starch (i.e., break apart the glucoses) don’t break apart the glucoses in the fiber cellulose (Chap. 5). Both cellulose and olestra have calories as measured in a bomb calorimeter (Chap. 3), but not for us.
*Regular fat doesn’t normally reach the colon because it’s normally digested and absorbed in the small intestine. If it reaches the colon, colon bacteria digest it and make substances that can irritate the colon and cause diarrhea.}