Tải bản đầy đủ - 0 (trang)
FD&C Yellow No. 6

FD&C Yellow No. 6

Tải bản đầy đủ - 0trang

thetic equivalents (Table 8.4). The FDA has approved a

number of such substances that are exempt from certification. However, the amounts used are limited (Freund,

1985). Unlike the certified colors, most of the natural colors have been subject to very limited toxicity testing. The

toxicological properties of some of the important natural

colorants are briefly described in the following.

Nonsynthetic (Natural) Colorants

Annatto Extracts

The annatto extract pigments are extracted from the

pericarp of the seeds of the annatto tree (Bixa orellana L.).

Annatto extracts are prepared by leaching the seeds with

one or more approved food-grade solvents such as edible

vegetable oils and fats and alkaline and alcoholic solutions. The major coloring compound of the oil-soluble extract is the carotenoid bixin (Color Index No. 75120)

(Figure 8.3).

The mutagenic action of annatto extracts was tested

at 0.5 g/100 ml in cultures of E. coli. No adverse effects

were found (Lueck and Rickerl, 1960). The administration

of aqueous extracts of bixa roots depresses the spontaneity

of motor activity in mice; the intraperitoneal effective dose

ED50 is 21 mg/kg body weight. The extract also affects the

volume of gastric secretion but not its pH when injected

intraduodenally at a 400-mg/kg level. Annatto extracts are

also antispasmodic (1 mg/ml, isolated guinea pig ileum)

and hypotensive (intravenous at 50 mg/kg body weight in

rats) (Durham and Allard, 1960). Several long-term tests

in mice and rats performed on well-defined types of extracts containing 0.1%–2.6% bixin, however, have not

shown any carcinogenic potential associated with their



Fruit juices and grape skin extracts provide a wide

spectrum of anthocyanins, which typically color foods

blue, rose red, or violet. They are the glycosides of anthocyanidins consisting of 2-phenyl benzopyrillium (flavylium) structure (Figure 8.3). Widely used in a variety of

food products, anthocyanins seldom pose any toxicity concerns.

Dehydrated Beets (Beet Powder, Betalains,


Betalains are found in the members of the Centrospermae family of plants such as red beets, chards, cactus

fruits, pokeberries, bougainvillea, and amaranthus flowers.

The color additive beet powder is defined as “a dark red

powder prepared by dehydrating sound, mature, good

Copyright 2002 by Marcel Dekker. All Rights Reserved.

quality, edible beets” (Marmion, 1984). Betanin (Figure

8.3) is the principal pigment in beet colorant, accounting

for 75%–95% of the total betacyanins.

Pink lemonade can get its color from the addition of

beet extract. Even though this colorant is natural, the beverage is labeled as artificially colored because lemonade is

not naturally pink.


The most abundant of naturally occurring plant pigments, chlorophylls are the green and olive-green pigments (Figure 8.3) in green plants. Chlorophyll extracts

are not permitted for food use in the United States. However, they may be added to foods in the form of green

vegetables. In such cases, they are classified as food ingredients. Chlorophylls marketed as chlorophyll-copper complexes are permitted food colorants in Canada and Europe

(Newsome, 1990).


Caramel belongs to the group of melanoidin pigments responsible for the attractive red-brown color of

cooked foods. Approximately 75%–85% of the caramel

produced in the United States is used in soft drinks, especially root beers and colas.

In toxicological studies, no abnormalities were detected after observation of animals for 14 days after administration of 12 different caramel color products (Foote

et al., 1958; Chacharonis, 1960, 1963). Sharratt (1971)

showed that a single dose of caramel, up to 10 g/kg body

weight in mice and 15 g/kg in rabbits, did not cause convulsions or other signs of distress. Several short-term toxicological studies in rats at levels of 10–15 g caramel/kg

body weight or 10%–20% caramel/kg body weight did not

show any abnormalities or significant differences in results

of histochemical and hematological studies (Prier, 1960;

Haldi and Calandra, 1962; Key and Calandra, 1962). Similarly, many long-term toxicological and reproduction studies (WHO, 1975) also showed no adverse effects in up to

three generations.

Cochineal Extract (Carmine, Carminic Acid)

Cochineal extract (Color Index No. 75470) is the

concentrated solution obtained after the removal of alcohol from an aqueous, alcoholic extract of cochineal. It is

obtained from the dried bodies of the female insect Coccus

cacti (Dactylopius coccus costa). Cochineal extracts consist mainly of carminic acid (Figure 8.3).

Short-term toxicological studies on cochineal extracts were conducted in mice and rabbits. When 1%–2%

aqueous solutions of the lithium salt of carminic acid were

injected intraperitoneally for a period of 60 days, the only

abnormality observed in mice was a proliferation of spleen

tissue (WHO, 1975). Similar effects were also observed in

rabbits given intravenous injections of 3–10 ml of 2%–4%

aqueous solution of the lithium salt of carminic acid every

5–6 days. In another study, groups of 40 rats received

cochineal carmine in 0.4% aqueous gum tragacanth at 0,

2.5, 5.0, and 10 mg/kg body weight, by intubation. The

process was carried out 5 days per week for a period of 13

weeks. No adverse hematological effects were noted

(WHO, 1975).

Turmeric and Turmeric Oleoresin

Turmeric (Color Index No. 75300) is the fluorescent

yellow extract of the dried, ground rhizome of Curcuma

longa. It is a perennial herb of the Zingiberaceae family

native to southern Asia and is cultivated widely in China,

India, South America, and the East Indies. Turmeric oleoresin is the combination of flavor and color principles obtained from turmeric by solvent extraction.

The major pigment in turmeric, and its oleoresin is

curcumin (1,6-heptadiene-3,5-dione-1,7-bis[4-hydroxy-3methoxyphenyl]) (Figure 8.3). They are often used to replace FD&C Yellow No. 5 in a variety of foods. It is a

GRAS substance in the United States. Little is known

about the toxicity of curcumin.

Nature-Identical Colorants

β-Carotene (Provitamin A)

β-Carotene (Color Index No. 75310) (Figure 8.3) is

an isomer of the naturally occurring carotenoid pigment

carotene. It was one of the first “natural” colorants synthetically produced on a commercial scale. Its use in food

as a permitted colorant eventually led to the creation of the

colorants exempt from certification category (Marmion,

1984). Unlike for other nature-identical carotenoid colorants, the FDA permits the addition of β-carotene to color

foods at any levels consistent with good manufacturing

practice (GMP).

In humans, about 30%–90% of the ingested β-carotene is normally excreted in the feces. A concomitant intake of fat does not improve its absorption. Excessive

doses of β-carotene depress the vitamin A activity of the

absorbed fraction; only a small fraction appears in the serum. When dissolved in oil, as much as 10%–41% and

50%–80% of β-carotene is absorbed in adults and children, respectively (Fraps and Meinke, 1945). Although hypercarotenemia associated with excessive intakes of βcarotene is harmless and produces no adverse symptoms,

Copyright 2002 by Marcel Dekker. All Rights Reserved.

it usually disappears when the vitamin intake is discontinued (Abrahamson and Abrahamson, 1962; Nieman et al.,

1954). Bagdon and coworkers (1960) also noted the absence of hypervitaminosis in human volunteers given βcarotene over an extended period. Similarly, Greenberg

and associates (1959) saw no symptoms of hypervitaminosis in 15 individuals who received daily doses of 60 mg of

β-carotene over a 3-month period. Serum carotene levels

rose from 120 µg/100 ml to a maximum of 308 µg/100 ml

after 1 month, while vitamin A levels remained essentially

unchanged. Long-term studies of up to four generations in

rats fed 0 to 100 ppm of β-carotene daily for 110 weeks

also showed no adverse effects in any of the generations

(Bagdon et al., 1960).

β-Apo-8′-Carotenal (Apocarotenal)

β-Apo-8′-carotenal (Color Index No. 40820) (Figure

8.3) is an aldehydic carotenoid that occurs naturally in oranges, spinach, grass, tangerines, and marigolds. It is

available commercially as a synthetic color.

In toxicological studies, no adverse effects were observed in dogs of both sexes when apocarotenal was fed at

daily levels of 0.1 or 1 g per animal during a 14-week period. In other studies, three- to fivefold higher vitamin A

levels were found in test animals as compared to the respective controls. Histopathological examination also

showed pigmentation of the kidney, adipose tissue, and adrenal cortex (Bagdon et al., 1962).


The carotenoid pigment canthaxanthin (Color Index

No. 40850) (Figure 8.3) occurs naturally in an edible

mushroom (Cantharellus cinnabarinus), sea trout, algae,

daphnia, salmon, brine shrimp, and several species of flamingo. It has been available commercially as a synthetic

food color since 1969.

Toxicological studies conducted with three dogs of

each sex fed 100 and 400 mg/kg body weight canthaxanthin for 1 week showed no adverse effects on body weight

or general health of the animals as compared to those of

the control dogs (WHO, 1975). Similarly, when tested at

1000-ppm levels, canthaxanthin did not produce any adverse teratological effects in three generations of rats.

Along with β-carotene and apocarotenal, the FDA

permanently lists canthaxanthin as an uncertified color additive (Dziezak, 1987).

8.5.5 Food Uses and Consumption Patterns

Typical usages of certified and exempt colorants in various

food systems are summarized in Tables 8.5 and 8.6. Data

Table 8.5

Food Applications of Synthetic Colors Regulated in the United Kingdom and the United States


FD&C no.

Yellow/orange colors


Yellow No. 5

Yellow 2G

Quinoline Yellow

Sunset Yellow FCF

Yellow No. 6

Orange RN

Orange G

Red colors


Ponceau 4R


Red 2G


Red No. 3

Allura® Red

Citrus Red No. 2

Red No. 40

Citrus Red No. 2

Blue colors


Patent Blue V

Brilliant Blue FCF

Green colors

Green S

Fast Green FCF

Blue No. 2


General-purpose, powdered desserts, confectionery, ice cream, dairy products, soft

drinks, pickles, sauces, fish, bakery products


General-purpose, soft drinks, desserts, ice cream, dairy products, confectionery

General-purpose, soft drinks (not recommended if calcium ions present), ice cream,

canned foods, confectionery, baked goods, desserts



Confectionery, soft drinks, ice cream, desserts, canned fruit

Soft drinks, confectionery, jellies, canned goods, fish, lake to color cheese rind, and

coated confections

Canned foods, soft drinks, jams, ice cream, powdered desserts

Meat products, sugar confectionery, jams

Only red color used with maraschino cherries and glacé; also used in meat products,

confectionery, and canned foods


Coloring of orange skins only

Blue No. 1


General-purpose, confectionery, drinks, icings

General-purpose, confectionery, drinks, icings

Green No. 3

General-purpose, often blended with yellow to produce leafy green hues

General-purpose, oftend blended to produce various shades

Brown colors

Brown FK

Chocolate Brown FB

Chocolate Brown HT

Coloring of fish in brine without precipitation

Baked cereal products, sugar confectionery, desserts

General-purpose, baked products, vinegar, confectionery

Black colors

Brilliand Black BN

Black 7984

General-purpose color used in blends, also in fish roe products and confectionery


Source: From Marmion (1984), Newsome (1990), Rayner (1991), and Ghorpade et al. (1995).

on the major categories of processed foods manufactured

with certified colors and the levels of colorant used are

presented in Table 8.7. These figures were obtained by the

Certified Color Industry Committee from a survey of sales

records of certified colorants used by the various segments

of the color industries (CCIC, 1968).

Ten percent of the food in the United States contains

added color (NAS/NRC, 1989). Because of the public concern over the increasing use of food additives in processed

foods, the NAS and the NRC conducted an extensive survey of more than 12,000 individuals and estimated their

average daily intake of food additives, including certified

Copyright 2002 by Marcel Dekker. All Rights Reserved.

and exempt food colorants (NAS/NRC, 1979). The results

are summarized in Table 8.8. The average daily intake of

the certified FD&C colorants by Americans above the age

of 2 years ranged from 3.1 to 100 mg/day/person, whereas

that of colorants exempt from certification ranged from

0.43 to 250 mg/day/person. Because the U.S. food supply

is very complex and different food colorants can be used

interchangeably in foods to achieve similar technical effects, the NAS estimated actual intakes to be approximately one fifth of the reported amount. There is some

concern that the intake in children is higher than that in the

general population, as they are heavier consumers of foods

Table 8.6

Food Applications of Colorants Exempt from Certification

Anthocyanins (blue-red shades)

Soft drinks, alcoholic drinks, sugar confectionery, preserves, fruit toppings and sauces, pickles, dry mixes, canned and frozen foods, dairy


Annatto extracts (orange shades)

Oil-soluble bixin: dairy and fat-based products, processed cheeses, butter, margarine, creams, desserts, baked and snack foods

Water-soluble norbixin: sugar and flour confectionery, cheese, smoked fish, ice cream and dairy products, desserts, custard powders,

cereal products, and bread crumbs

β-Carotene (yellow to orange)

Butter, margarine, fats, oils, processed cheeses, water-dispersible forms in soft drinks, fruit juices, sugar and flour confectionery, ice

cream, yogurts, desserts, cheese, soups, and canned products

β-Apo-8′-carotenal (orange to orange-red)

Cheese, sauces, spreads, oils, fats, ice cream, cake mixes, cake toppings, snack foods, and soft drinks

Canthaxanthin (orange-red to red)

Sugar confectionery, sauces, soups, meat and fish dishes, ice cream, biscuits, bread crumbs, salad dressings

Paprika (orange to red)

Meat products, snack seasonings, soups, relishes, salad dressings, processed cheeses, sugar confectionery, fruit sauces and toppings

Saffron (yellow)

Baked goods, rice dishes, soups, meat dishes, sugar confectionery

Crocin (yellow)

Smoked white fish, dairy products, sugar and flour confectionery, jams and preserves, rice and pasta

Lutein (yellow)

Salad dressings, ice cream, dairy products, sugar and flour confectionery, soft drinks

Beet powder (bluish red)

Frozen and short shelf life foods, ice cream, flavored milks, yogurts, dry mix desserts, jelly crystals

Cochineal (orange)

Soft drinks and alcoholic drinks

Cochineal carmine (bluish red)

Soft drinks, sugar and flour confectionery, flavored milks, desserts, sauces, canned and frozen products, pickles and relishes, preserves

and soups

Sandalwood (orange to orange-red)

Fish processing, alcoholic drinks, seafood dressings, bread crumbs, snack seasonings, meat products

Alkannet (red)

Sugar confectionery, ice cream, alcoholic drinks

Chlorophyll (olive green)

Sugar confectionery, soups, sauces, fruit products, dairy products, pickles and relishes, jams and preserves, pet foods, drinks

Copper chlorophyll (green)

Flour and sugar confectionery, soups, sauces, pickles, relishes, fruit products, ice cream, yogurts, jelly, desserts, dry mix desserts, sauces

and soups, soft drinks

Caramel (yellowish tan to red-brown)

Alcoholic and soft drinks, sugar and flour confectionery, soups, sauces, desserts, dairy products, ice cream, dry mixes, pickles, and relishes

Malt extract (light brown)

Alcoholic and soft drinks, sugar and flour confectionery, soups, sauces, desserts, dairy products, ice cream, dry mixes, pickles, and


Turmeric (bright yellow)

Ice cream, yogurt, frozen products, pickles and relishes, flour and some sugar confectionery, dry mixes, yellow fats

(table continues)

Copyright 2002 by Marcel Dekker. All Rights Reserved.

Table 8.6


Riboflavin (yellow)

Cereal products, sugar-coated confectionery, sorbet, ice cream Vegetable carbon black (black)

Sugar confectionery, shading color

Orchil (red)

Soft drinks, alcoholic drinks, sugar confectionery

Safflower (yellow)

Soft drinks, alcoholic drinks

Titanium dioxide

Sugar-coated confectionery

Ferrous gluconate

Ripe olives

Iron oxides

Sugar-coated confectionery, pet foods, meat and fish pastes

Silver, gold, and aluminum

Surface coating of sugar confectionery, cake decorations


Not all colors are permitted for food use in the United States.

Source: From Marmion (1984), Newsome (1990), Rayner (1991), and Ghorpade et al. (1995).

that contain more coloring. Unfortunately, data on the intake of food colors by various segments of the population

are not available.

On the basis of U.S. food consumption patterns and

the amount of FD&C colors certified by the FDA during

the 1978–1981 period, Marmion (1984) estimated the

Table 8.7 Major Categories of Processed Foods Manufactured

Using Certified Colors and Levels of Color Used

Level of color used (ppm)


Candy and confections

Beverages (liquid and powdered)

Dessert powders


Maraschino cherries

Pet foods

Bakery foods

Ice cream and sherbets

Sausage (surface)

Snack foods


























Includes nuts, salad dressings, gravy, spices, jams, jellies, and food packaging.

Source: From CCIC (1968).

Copyright 2002 by Marcel Dekker. All Rights Reserved.

consumption of certified color additives as 0.024 lb/day/


Labeling Issues

When color is added to a food, the label must state artificially colored or artificial color added. The term natural

color may not be used even if the color is derived from nature. This labeling is intended to protect the consumer so

that there can be no misunderstanding about whether it is

the actual color of the item or one that has been enhanced

by the addition of color of any kind. The alternate way to

label the product is to declare on the label “colored with

_________” or “__________ color,” where the blanks are

filled in with specific name(s) of the colorant(s) used. The

declaration of the specific colorant FD&C Yellow No. 5

(tartrazine) on the label has been required since 1986,

since some individuals are sensitive to it. However, the

Nutrition Labeling and Education Act passed by the U.S.

Congress in November 1990 required that all certified colors be shown on the label after November 1991.

8.5.6 Food Colorants and Hyperkinesis

Food additive–induced hyperkinesis is characterized by

several types of aberrant behavior whereby individuals

show one or more signs of the following: hyperactivity and

fidgetiness, compulsive aggression, excitability, impul-

Table 8.8


Consumption of Certified FD&C and Exempt Food


Average daily intake


Certified FD&C colorants

FD&C Red No. 3

FD&C Red No. 40

FD&C Blue No. 1

FD&C Blue No. 2

FD&C Yellow No. 5

FD&C Yellow No. 6

FD&C Green No. 3

Orange B

FD&C Red No. 3 Lake

FD&C Red No. 40 Lake

FD&C Blue No. 1 Lake

FD&C Blue No. 2 Lake

FD&C Yellow No. 5 Lake

FD&C Yellow No. 6 Lake















Colorants exempt from certification


Annatto extract


Paprika oleoresin


Turmeric oleoresin


Cochineal extract (carmine)

Grape skin extract (enocianina)

Beet powder (dehydrated beets)

Carrot oil















Data represent the 99th percentile of persons over 2 years of age in the

“eaters group” (those who consumed one or more foods containing the

additive in question during the 14-day survey period). Ninety-nine percent of the population sampled was estimated to have intakes equal or

below the value shown. Total sample size was 12,000 persons.

Source: From NAS/NRC (1979).

siveness, impatience, short attention span, poor sleep habits, and gross and fine muscle incoordination. Such

behavior is generally accompanied by learning disabilities

in the form of difficulty in reasoning, lack of auditory and

visual memory, and difficulty in understanding ideas and

concepts. Several clinical trials have confirmed that additive-free diets can indeed improve the behavior of hyperkinetic children (Brenner, 1977; Conners, 1980).

Such studies, however, should not be expected to

provide definite conclusions about the hyperkinetic effects

of food additives, especially the synthetic food colors.

Copyright 2002 by Marcel Dekker. All Rights Reserved.

Among other findings, these studies are highly subject to

the placebo effect (Spring and Sandoval, 1976; Harley and

Matthews, 1977; Wender, 1977). Harley and colleagues

(1978), while studying nine boys selected from a group of

46 hyperactive subjects, found that only one subject responded with increased undesirable behavior when challenged with cookies and candy bars containing 26 mg of a

blend of eight approved food colors. In a similar study,

Weiss and coworkers (1980) observed that, among 27 hyperactive children (22 males and 5 females, aged 2.5–7

years) who previously responded favorably to additivefree diets, only 2 showed statistically significant adverse

responses when challenged daily with a mixture of about

35 mg of certified FD&C colors. One 3-year-old boy had a

mild response on several criteria. The food colors used in

this study were FD&C Yellow No. 5, FD&C Yellow No. 6,

FD&C Red No. 40, FD&C Red No. 3, FD&C Blue No. 1,

FD&C Blue No. 2, and FD&C Green No. 3.

Yet another aspect of such double-blind studies was

that the food colors elicited hyperkinetic behavior rapidly

and briefly. This is contrary to Feingold’s (1975) initial

claim that the effects persisted for several days. Thus if

two different observers were to note behavioral changes

several hours apart, they might report opposite effects. For

example, Williams and associates (1978) have reported

that only the teachers, who were in a better position to observe early behavioral changes than were the parents,

noted the improvement or worsening of the hyperactive

behavior of test children. Goytte and colleagues (1978)

and Conners (1980) have observed hyperkinetic effects

within 3 hours after the food additive challenge. Levy and

coworkers (1978) reported that significant effects in terms

of deterioration in behavior could be detected only when

measured within a few hours after a tartrazine challenge;

the effect could not be observed after 24 hours.

The failure of some earlier studies (Harley et al.,

1978; Levy et al., 1978; Swanson and Kinsbourne, 1979a,

1979b) to detect effects of food additive challenges may

have been due to the low doses (1 to 26 mg of food colors)

used in their experiments. Thus when much higher doses

(up to 150 mg) of certified FD&C food color blends were

used, Swanson and Kinsbourne (1980) were able to document impaired behavior on a laboratory learning test based

on the Conners scale in all 20 confirmed hyperactive children. Once again, the measurements were made within 3.5

hours after the challenge with a blend of nine dyes. The

nonhyperactive children, in contrast, did not show any adverse effects. The amount of food color used in this study,

according to an FDA estimate, was at the 90th percentile

for artificial food colors consumed by 5- to 12-year-old

children in the United States (Sobotka, 1976). Swanson

and Kinsbourne (1980) have commented that the time

course of the appearance of the effect, i.e., the initial appearance of 0.5 hour after administration, peaking at 1.5

hours and lasting up to at least 3.5 hours, suggested that

the food additive effect was nonimmunological.

The nonimmunological nature of the hyperkinetic

effect of food colorants was also observed in animal studies. Mailman and associates (1980) showed that the administration of 50–300 mg/kg body weight FD&C Red

No. 3 (erythrosine) to rats attenuates the suppressive effect

of punishments monitored by the number of electric

shocks received by the animals in an approach-and-avoidance test. This observed effect in rats is similar to that seen

with barbiturate and benzodiazepine drugs, which also aggravate hyperkinesis in humans. In contrast, amphetamine

reverses such effects (Cantwell, 1975). Levitan (1977) has

reported membrane interactions with FD&C Red No. 3

dye, whereas Logan and Swanson (1979) have observed

that this dye also decreases the uptake of several neurotransmitters by homogenates prepared from rat brains.

However, at least with dopamine, a major portion of the

observed effects may have been an artifact resulting from

its nonspecific interaction with brain membranes (Mailman et al., 1980). The dye also irreversibly increases acetylcholine release when applied to isolated neuromuscular

synapses in the frog (Augustine and Levitan, 1980). The

neurotransmitter release, which normally depends on the

presence of calcium ions in the presynaptic terminals, was

also independent of its concentration.

The studies cited essentially support the basic

premise of Feingold that food additives do induce certain

behavioral changes in humans. However, at least with the

artificial food colors, the hyperkinetic syndrome in humans may be induced or exacerbated in a subset of children. The evidence also shows that the basic mechanism

may involve the central nervous system. Thus one important aspect of food toxicological mechanisms, the behavioral toxicity of food components, is underscored by the

experience with artificial food colors. These studies perhaps will create a greater interest in this field and the unexplored aspects of behavioral food toxicology.



Sweeteners present the consumers with one of the most

important taste sensations. This is reflected by the world

production of sugar, which increased from 8 million tons

in 1900 to over 90 million tons in 1990s. For nutritional

and health reasons, however, there is a growing need for

sugar substitutes in food that are nonnutritive, i.e., noncaloric and noncarcinogenic.

Copyright 2002 by Marcel Dekker. All Rights Reserved.

Noncaloric sweeteners lead all other food additives

in dollar sales, and their use appears to be growing. Concerns about sweeteners seem to grow along with their acceptance. The banning of cyclamate, the subsequent

controversy about saccharin, and spurious reports about

aspartame (Roberts, 1990) have raised consumer concerns

about the safety of sweeteners and other additives. The cyclamate banning sparked a systematic review of all additives. Interest in artificial sweeteners continues because of

the strong interest in dieting and because saccharin, aspartame, and other sweeteners are frequently in the news

(Lecos, 1985; Stamp, 1990; Jones, 1992). The toxicological properties of three of the most important sweeteners,

viz., cyclamates, saccharin, and aspartame, are described

in the following sections.

8.6.1 Cyclamates

The sodium and calcium salts of N-cyclohexylsulfamic

acid (Figure 8.4), commonly known as cyclamates, were

introduced commercially in 1950. These synthetic sweeteners are about 30 times as sweet as sucrose, and about

one tenth as sweet as saccharin. Cyclamate’s sweetness

coupled with a less bitter aftertaste than saccharin made it

very viable commercially as an artificial sweetener. Cyclamates are sold generally as a mixture of 1 part saccharin

and 10 parts cyclamate. A commercial brand, Sucaryl,

containing such a mixture was widely available in the

United States prior to its banning in 1970. Since cyclamate

was 20 times less expensive than saccharin, although less

sweet, the use of cyclamate-saccharin mixtures soared.

In 1955, the NAS reported cyclamate safe for human

consumption. Since it was used in the food supply before

1958, cyclamate was classified as GRAS with the passage

of the Food Additives Amendment Act. In the early 1960s,

because of its popularity and increasing consumption, especially in carbonated beverages, the FDA again requested

the NAS to assess the safety of cyclamate at the then-current levels of use. The NAS reported to the FDA that although reasonable quantities of cyclamate posed no hazard

to humans, additional studies were needed to resolve some

questions about its safety.

In 1969, Oser (1975) revealed the results of their experiments whereby Wister-derived rats were fed a 10:1 cyclamate/saccharin mixture, and after 78 weeks, some of

the animals were fed cyclohexylamine. The rats were on

treatment for 104 weeks, at 500, 1120, or 2500 mg mixture/kg body weight. Papillary carcinomas of the bladder

were reported in 12 (9 males, 3 females) survivors that

were fed 2500 mg/kg; none was found with the lower

doses or the control. When the FDA learned of the results,

cyclamates were removed from the GRAS list on October

Figure 8.4

Chemical structures of nonnutritive sweeteners.

18, 1968. The FDA further ordered in 1969 the halting of

the production of general-purpose products containing cyclamates. In addition, all beverages and packaged mixes

containing cyclamates were ordered removed from the

marketplace beginning January 1, 1970.

Two other studies in mice have shown cyclamates to

be tumorigenic. Bryan and Erturk (1970) implanted cholesterol pellets containing one part sodium cyclamate and

four parts cholesterol into the bladder of 60- to 90-day-old

mice; all surviving animals were killed at 13 months.

Bladder tumors were observed in 34/69 of treated animals

compared to 13/106 of control (p < 0.001). Rudali and colleagues (1969) administered sodium cyclamate in drinking

water of mice throughout their lifespan at a dose of 20 to

25 mg/kg/day. In treated female XVII/G mice, 16/20 lung

tumors predominantly developed as compared to 3/16 in

the control group (p < 0.01). Among treated male C3H ×

RIII mice, 28/34 predominantly showed liver tumors compared to the 16/28 in the control (p < 0.05). A shorter latent period also was observed in the cyclamate-treated

animals. However, these tumors were diagnosed on the basis of gross appearance only.

More significantly, the experiments of Hicks and coworkers (1975) demonstrated the cocarcinogenicity of sodium cyclamate in Wistar rats. In this case, the animals

were given a single 2-mg intramuscular subcarcinogenic

dose of N-methyl-1-nitrosourea (NMU), then fed sodium

cyclamate in the diet equivalent to 1 or 2 g/kg/day. Among

Copyright 2002 by Marcel Dekker. All Rights Reserved.

those given the higher dose, bladder tumors developed in

34/69 of the treated group; in this group, 18 kidney tumors

also were detected. Neither type of tumor was seen in 124

controls (p < 0.001, in both cases). No tumors were found

among animals treated with NMU or cyclophosphamide,

both known to cause this type of tumor. These observations pointed to an important aspect of the potential health

hazards of cyclamates: i.e., the possible influence of other

exogenous compounds.

In contrast to these positive effects, 22 other studies

by investigators around the world using sodium cyclamates, and in some, calcium cyclamate or a mixture of

these sweeteners and saccharin, have yielded negative results. Almost all of these studies employed a dose of 5%

sodium cyclamate in the diet. These studies employed rats

and mice but also hamsters, monkeys, and beagle dogs.

However, in the opinion of the National Cancer Institute

committee (TCRDCC, 1976), at least 10 of these studies

were deficient in various factors, so that their negative

conclusions cannot be given much credence. Nevertheless,

five studies with negative results were, at a minimum,

properly executed.

Because of the negative findings in several feeding

studies and the questioned human applicability or uncertainties of those studies showing positive results, even

though the experimental protocols and results were valid,

the Temporary Committee for the Review of the Data on

the Carcinogenicity of Cyclamate (TCRDCC, 1976) con-

cluded, “The present evidence does not establish the carcinogenicity of cyclamate nor its major metabolite,

cyclohexylamine, in experimental animals.” Thus, the

committee implied that the available evidence neither

proved nor disproved the suspected carcinogenicity of cyclamates. Indeed the committee stated that the studies of

Bryan and Erturk (1970), Hicks and associates (1975), and

Oser (1975) were not invalid but that these studies underscored the uncertainties associated with bioassay techniques. The work of Oser (1975) did not address itself to

the question of cyclamate carcinogenicity because a saccharin-cyclamate mixture was used. The Bryan and Erturk

(1970) technique involving bladder implantation cannot be

related to the mode of human exposure, since “highly artificial conditions were being used.” The cocarcinogenicity

study of Hicks and colleagues (1975) involving exposure of

the animals to a subcarcinogenic dose of N-methyl-Nnitrosourea prior to feeding cyclamate has not yet been shown

to be valid as a means of detecting bladder carcinogens.

Finally, the insensitivity of bioassay techniques used

renders judgment regarding weak carcinogens impossible.

For example, it has been calculated that a total of 51,968

animals would be required to detect a 1% difference in

bladder tumor incidence between test and control groups;

this total assumes a survival of 18 to 24 months, and a

bladder tumor incidence in the control (spontaneous) of

1%. Even so, some major uncertainties still remain.

In 1973 Abbott Laboratories sought FDA’s permission to remarket the sweetener in foods designed for special dietary purposes. In the intervening years, there have

been several attempts to reinstate cyclamate. One new petition included nearly 500 new toxicological assessments

attesting to cyclamate’s safety. However, even after a series of petitions and court challenges, cyclamate was still

not permitted in the food supply. In 1985, the NAS reviewed all existing data and concluded that cyclamate and

its incriminated metabolite, cyclohexylamine, were not

themselves carcinogens but could be cancer promoters

(NAS/NRC, 1985). The Society of Toxicology published

an article stating that the FDA’s cyclamate decision was an

example of how not to do and interpret animal studies

(Munro, 1987).

In 1984, another petition to reinstate cyclamate was

submitted to the FDA. This included over two dozen studies indicating that high doses of cyclamate throughout the

lives of laboratory animals did not cause cancer. Further,

15 epidemiological studies showed no significant increase

in the relationship between bladder cancer risk and the use

of artificial sweeteners, both saccharin and cyclamate

(Calorie Control Council, 1985; Morris and Przybyla,

1985). The FDA’s Cancer Assessment Committee has now

Copyright 2002 by Marcel Dekker. All Rights Reserved.

exonerated cyclamate as a carcinogen (Newberne and

Conner, 1986).

Despite these extensive studies, cyclamate is still not

allowed as a food additive in the United States. It is, however, used in over 40 countries around the world. It is also

considered a safe additive by the WHO and the EEC

(Malaspina, 1987). The current ADI for cyclamate of 0–11

mg/kg body weight is derived by using the NOEL for cyclohexylamine-induced testicular toxicity in rats during

subchronic administration, assuming approximately a

level of 18% transformation of cyclamate to cyclohexylamine (which assumes 30% metabolism of the 60% cyclamate that is not absorbed and thus can be metabolized by

the intestinal microflora) and applying a safety factor of

200. The NOEL of 100 mg/kg body weight for cyclohexylamine was consistent in the different studies reviewed by

Bopp and associates (1986) and is one of the best validated

NOEL values available for food additives.

8.6.2 Saccharin

Saccharin (Figure 8.4) was discovered accidentally in

1879 (Fahlberg and Remsen, 1879) and began to be sold

commercially in 1900. It was originally used as an antibacterial agent and food preservative. It is 200 to 700

times sweeter than sucrose but must be used below a concentration of 0.1% because of its bitter aftertaste. This flavor quality was overcome somewhat in the 1950s when a

1:10 mixture of saccharin and cyclamate gained popularity

because the sweetness imparted by the mixture was greater

than the sum of the sweetness of each component, and the

bitterness of saccharin was held below its threshold of perception for most consumers.

The safety of saccharin has probably been investigated more and certainly has been reviewed and debated

more than that of any food additive. The suspicion regarding its safety occurred early in its commercial introduction. It was banned in 1912, but the ban was lifted during

World War I because of the shortage of sugar.

The first report that suggested the possible carcinogenicity of saccharin was that of Fitzhugh and colleagues

(1951), who reported an increased incidence of lymphosarcoma in rats fed a 5% saccharin diet. Because the control

also had a high incidence of tumors, this initial finding

was inconclusive (U.S. FDA, 1977). Therefore, in 1955

and again in 1968, the Food Protection Committee of the

U.S. National Academy of Sciences (1968) concluded that

a saccharin intake of as much as 1 g/day by an adult could

be considered safe.

However, the Wisconsin Alumni Research Foundation (WARF, 1972, 1973) reported the result of a two-generation study in which the parent (male and female)

Sprague-Dawley rats (F0) and their offspring (F1) were fed

0.05%, 0.5%, and 5% saccharin in the diet throughout

their lifespan. In the males of the F1 generation a statistically significant increased incidence of bladder tumors

(treated: 8/14, control, 0/14, p = 0.001) developed. A similar study by the U.S. FDA (1973) also showed a statistically significantly increased incidence of bladder tumors

among males of the F1 generation when fed a 7.5% saccharin diet for 2 years (treated, 7/23; control, 1/25, p =

0.018). Thus, when the results were published, the U.S.

FDA (1972) issued a regulation restricting general use of


Despite these results, a committee of the National

Academy of Sciences (NAS, 1974) charged by the FDA to

study the evidence concluded that partly because of the

possible presence of impurities such as orthotoluenesulfonamide (OTS), the evidence did not conclusively establish the carcinogenicity of saccharin; the committee

recommended further tests. OTS is the most common impurity in commercial saccharin preparation.

Between 1973 and 1977, 11 single-generation studies in many laboratories in the United States and in other

countries did not confirm these results. However, a twogeneration study using Charles River–Sprague-Dawley

rats conducted by the Canadian National Health and

Welfare Ministry (Arnold, 1977) confirmed the FDA and

WARF studies (treated, 12/45; control, 0/42, p = 0.002).

These studies also demonstrated that under the conditions

of the experiments, the F0 generation showed a significant

incidence of bladder tumors among male rats (treated,

7/38; control, 1/36, p = 0.033).

Whereas the saccharin samples used in the FDA and

WARF studies contained OTS estimated at 20 to 368 ppm

and 12.5 to 18.0 ppm, respectively (NAS, 1974), the samples used by the Canadian study contained none. OTS was

shown not to produce bladder tumors when fed at a level

of as high as 250 mg/kg/day in the diet or in drinking water with added 1% ammonium chloride (Arnold, 1977).

The latter was added to prevent the formation of alkaline

urine, which has been associated with the formation of

bladder calculi; ammonium chloride was correlated with

the formation of bladder tumors in mice.

The U.S. National Research Council/National Academy of Sciences Committee for a Study on Saccharin and

Food Safety Policy (NAS/NRC, 1978) reviewed the pertinent toxicological studies. The committee concluded the



Saccharin is absorbed rapidly in the GI tract, is

distributed throughout the body, and crosses the

placental wall. It is not metabolized in humans

as far as present analytical methods permit, and

Copyright 2002 by Marcel Dekker. All Rights Reserved.







only to a small extent in animals; it is eliminated mainly in the urine.

Saccharin is a bladder carcinogen in male rats

only as shown by two-generation studies in

which the rats were fed continuously with saccharin (5% or greater) in utero and throughout

life. In addition, in one two-generation study in

rats, the males of the parent generation also

showed a significant increase in bladder cancer.

Saccharin promotes bladder tumor development

in the presence of some other chemical carcinogens. Thus, the carcinogenic risk from saccharin as a tumor promoter may be considerably

greater than that by itself, since humans are subject to multiple exposures to environmental carcinogens. However, in this case, the state of the

science at present does not permit estimation of

the human risk.

Factors in design of the two-generation studies

do not present doubtful interpretation of results.

These factors include: doses studied (maximum

tolerated dose), exposure in utero, high sodium

diet (owing to use of sodium cyclamate) of

treated compared to control animals, and the

possible presence of microcalculi in the urinary

bladders of treated compared to control rats.

A few studies suggested an increase in benign

uterine tumors and ovarian lesions in saccharintreated rats.

OTS, the main impurity in some commercial

saccharin preparations, is not a carcinogen in

rats. That other impurities in saccharin may be

carcinogenic is a remote possibility for the following reasons:

a. Even though the saccharin used in two

studies (FDA and WARF) had different patterns of impurities, the same carcinogenic

responses (bladder cancer) were obtained.

b. The very much purer saccharin used in the

Canadian study produced no other tumors

in males but those in the bladder, and

c. If the impurities in the latter were carcinogenic, they would have to be very potent, since they are present at very low


Sixteen short-term assays for genetic effects

produced negative results, whereas results of

five others were positive. These variations in results might be expected because the assays evaluated various types of genetic effects and

because saccharin is a weak carcinogen. Thus,

these results are compatible with the in vivo car-

cinogenic effects. Definitive interpretation of

the health risk to humans has not been provided

by these results.

The FDA in 1977 proposed a total ban of saccharin,

invoking both the Delaney clause and the general safety

clause of the Food, Drug, and Cosmetic Act. Since saccharin was at that time the only available artificial sweetener,

the expected ensuing public furor resulted in the passage

of the Saccharin Study and Labeling Act, which was

signed into law in November 1977. Briefly, the act required the posting of warning signs regarding the health

hazards of saccharin in stores, established an 18-month extension before the ban could be enforced, and called upon

the National Academy of Sciences to study, among other

points, the question pertinent to the decision regarding the

safety of saccharin. The U.S. Congress has continued to

enact extensions of the moratorium on the ban.

Since 1977 many further studies have been done to

assess long-term hazards of using saccharin. Metabolic

studies showed it to be metabolically unchanged after being slowly absorbed and rapidly excreted. This is important evidence against its carcinogenicity, as no known

carcinogens are excreted unchanged. A 1983 study (IFT

Expert Panel, 1986) involving 2500 second-generation

male rats revealed that high doses of saccharin caused

changes in rat bladder tissue if the rat was exposed to saccharin during the suckling period but not if exposure occurred during the fetal period or after the suckling period.

The incidence of tumors was clearly a function of dose, as

number of tumors declined sharply as dose decreased.

From this experiment, the risk of consumption of two cans

of diet soda daily was extrapolated, and the increased risk

of human bladder cancer was calculated at less than 1 in 1

million. Variation in risk using different methods of extrapolation based on rodent experiments ranged from 0.2

cancer to 144,000 cancers in the next 70 years in the

United States (Munro, 1987).

Epidemiological studies in Scandinavia, Japan,

England, and the United States have not revealed an overall association between saccharin ingestion and bladder

cancer. Data from these studies include individuals whose

exposure to artificial sweeteners began decades ago (Concon, 1988; Newberne and Conner, 1986). One study found

no elevated risk for the population in general, but they did

find a positive association for several subgroups (Hoover,

1980). These included white males who were heavy smokers and nonwhite females with no known exposure to bladder carcinogens. The American Medical Association’s

Council on Scientific Affairs recommends a moratorium

on the saccharin ban, since evidence for the carcinogenic-

Copyright 2002 by Marcel Dekker. All Rights Reserved.

ity of saccharin in humans has not been forthcoming. It

also recommends careful monitoring of any adverse effects of saccharin and warns that young children and pregnant women carefully consider the use of saccharin

(AMA, 1985, 1986).

Legally, saccharin is now classified as a cocarcinogen (tumor promoter) with very low potency. Extrapolations suggest that saccharin at 30–300 mg/day (0.43–4.3

mg/kg/day) does not increase human cancer risk (Byard,

1984). It is allowed in the United States under the congressional moratorium on banning its use. Because the rat is

the only species that has been reported to show an increase

in the incidence of bladder tumors at high dietary concentrations of sodium saccharin, the JECFA concluded in


from the long-term feeding studies . . . the dose-related carcinogenic activity of sodium saccharin on

the urinary bladder was specific to the male rat

and . . . exposure during the neonatal period was

critical for the subsequent development of these tumors in the absence of an initiator or stimulus such

as freeze ulceration. The critical events during the

neonatal phase that lead to an increase in the population of initiated cells have not been identified.

Saccharin is approved for use in 80 countries and has been

determined to be safe by both the FAO/WHO JECFA and

the Scientific Committee for Foods of the European Economic Community (Arnold and Clayson, 1985; Arnold

and Munro, 1983).

Saccharin provides a good example of the difficulty

of evaluating the actual risk for humans on the basis of animal experiments without a complete knowledge of the

mechanism of action. The approach adopted by the expert

committees is considered appropriate since the epidemiological studies on saccharin did not show any evidence that

its ingestion increases the incidence of bladder cancer in

the human population.

8.6.3 Aspartame

Aspartame, a dipeptide formed from the two naturally occurring amino acids phenylalanine and aspartic acid (Figure 8.4), is about 150 to 200 times sweeter than sucrose.

Its sweet taste and lack of the bitter aftertaste often associated with artificial sweeteners make aspartame advantageous from a sensory viewpoint. Since it hydrolyzes

under acid conditions or at high temperatures, its use is

limited to cold nonacid products that do not require prolonged storage.

Tài liệu bạn tìm kiếm đã sẵn sàng tải về

FD&C Yellow No. 6

Tải bản đầy đủ ngay(0 tr)