Carcinogenicity and Pesticides - ACS Publications - American

Chapter 14

Pesticide Residues and Cancer Causation Bruce N. Ames

Downloaded by CORNELL UNIV on August 22, 2016 | http://pubs.acs.org Publication Date: October 31, 1989 | doi: 10.1021/bk-1989-0414.ch014

Department of Biochemistry, University of California, Berkeley, CA 94720

There are many sources of natural mutagens and carcinogens in the environment, such as natural toxic chemicals present in all plants as defenses against insects, chemicals formed on cooking or preparing food, and mold products and other dietary components. Since natural carcinogens appear to be extremely common, priority setting is required to separate important from trivial hazards. We discuss mechanisms of carcinogenesis and the implications of this for doing risk assessment. We also discuss reasons why animal cancer tests cannot be used to predict absolute human risks. Such tests, however, may be used to indicate that some chemicals might be of greater concern than others. Possible hazards to humans from a variety of rodent carcinogens are ranked by an index that relates the potency of each carcinogen in rodents to the exposure in humans. This ranking suggests that carcinogenic hazards from current levels of pesticide residues (or water pollution) are likely to be of minimal concern relative to the background levels of natural substances, though one cannot say whether these natural exposures are likely to be of major or minor importance. Many Chemicals are Carcinogens and Reproductive Toxins, and We Cannot Eliminate All of Them Over 50% of the chemicals tested to date in rats and mice have been found to be carcinogens at the high doses administered (1,2), the maximum tolerated dose (MTD). The exhaustive database of animal cancer tests developed by my colleagues and me (3,4) listed 392 chemicals tested in both rats and mice at the MTD. Of these, 58% of the synthetic chemicals and 45% of the natural chemicals were carcinogens in at least one species (.1,2). We concluded that the proportion of chemicals found to be carcinogens is strikingly high, a conclusion reached by others with smaller compilations. The earlier 0097-6156/89/0414-0223$06.00/0 ©1989 American Chemical Society Ragsdale and Menzer; Carcinogenicity and Pesticides ACS Symposium Series; American Chemical Society: Washington, DC, 1989.


224

CARCINOGENICITY AND PESTICIDES

I n n é s et a l . study (5) is sometimes c i t e d to support the conclusion that the proportion of carcinogens is low. The Innes study was a much smaller dataset (120 chemicals, 11 p o s i t i v e ) , and the t e s t s , though appropriate for t h e i r time, used only one species and were less thorough than modern tests (2). Even when one considers that some chemicals are selected for testing because they are suspicious, the high proportion of positives i s disturbing. From considerations of carcinogenesis mechanisms, it i s plausible that a high proportion of a l l chemicals we test i n the future, both natural and man made, w i l l prove to be carcinogens (1^; see section on Extrapolating R i s k s ) . High proportions of positives are also reported for teratogenic tests. F u l l y one-third of the 2800 chemicals tested i n laboratory animals have been shown to induce b i r t h defects at maximum tolerated doses (6). Thus, i t seems l i k e l y that a sizeable percentage of both natural and man-made chemicals w i l l be reproductive toxins when tested at the MTD. The world i s f u l l of carcinogens and reproductive toxins, and i t always has been. The important issue i s the human exposure dose, and, fortunately, almost a l l of these are usually tiny. The major preventable r i s k factors for cancer causation, such as tobacco, dietary imbalances (Z~I2)> hormones (ΙΑ), and viruses (ljSjljO, have been discussed by us ( 1,17-22) and others (7,14,23-25). Man-Made Chemical Pollutants Significant Amounts

Do

Not

Appear

to

be

Present

in

We have attempted to address the issue of p r i o r i t y setting among possible carcinogenic hazards (!)· Since carcinogens d i f f e r enormously in potency in rodent t e s t s , a comparison of possible hazards from various carcinogens ingested by humans must take t h i s into account. Our analysis makes use of an exhaustive database of animal cancer tests (currently 3500 experiments on 975 chemicals) (2>4) that calculates carcinogenic potency, the T D , e s s e n t i a l l y the dose of the carcinogen to give half of the animals cancer. The T D i s close to the high dose (MTD) actually given and thus involves a minimal extrapolation. To calculate our index of possible hazard we express each human exposure ( d a i l y l i f e t i m e dose i n mg/kg) as a percentage of the rodent T D dose (mg/kg) for each carcinogen. We c a l l t h i s percentage HERP (Human Exposure dose/Rodent Potency dose). As rodent data are a l l calculated on the basis of l i f e t i m e exposure at the indicated d a i l y dose rate ( 3 . » 4 ) , human exposure data are s i m i l a r l y expressed as l i f e l o n g d a i l y dose rates even though the human exposure i s l i k e l y to be less than d a i l y for a l i f e t i m e . The HERP values are not r i s k assessment, because i t is impossible to extrapolate to low doses (see section on Extrapolating R i s k s ) , but are a way of comparing possible hazards of exposures to put them i n perspective and to set p r i o r i t i e s (Table 1). This analysis suggests that the amounts of pollutants that humans are ingesting from pesticide residues or water pollutants appear to be t r i v i a l r e l a t i v e to the background of natural, and t r a d i t i o n a l ( e . g . , from cooking food) carcinogens (1,22). 5 0

5 0

5 0

t

n

e

Nature's Pesticides. Americans ingest i n t h e i r diet at least 10,000 times more by weight of natural pesticides than of man-made pesticide

Ragsdale and Menzer; Carcinogenicity and Pesticides ACS Symposium Series; American Chemical Society: Washington, DC, 1989.


14.

AMES

Pesticide Residues and Cancer Causation

225

residues (22). These natural "toxic chemicals" have an enormous variety of chemical structures, appear to be present in a l l plants, and serve as protection against fungi, insects, and animal predators (21,22). Though only a few dozen are found in each plant species, they commonly make up 5 to 10% of the plant's dry weight (22). There has been relatively little interest in the toxicology or carcinogenicity of these compounds u n t i l quite recently, although they are by far the main source of "toxic chemicals" ingested by humans. Most chemicals tested for carcinogenicity in rodent bioassays are synthetic compounds; however, the proportion of positive tests is about as high for natural pesticides as for synthetic chemicals (roughly 30%). Since over 99.99% of the pesticides we ingest are "nature's pesticides" (1^,21^,22), our diet i s l i k e l y to be very high in natural carcinogens. Their concentration i s usually in parts per thousand or more rather than parts per b i l l i o n , as is usual for synthetic pesticide residues or water pollutants (1_). The known natural carcinogens in mushrooms, parsley, b a s i l , parsnips, fennel, pepper, celery, f i g s , mustard, cabbage, b r o c c o l i , Brussels sprouts, carrots, pineapple, and c i t r u s juices are undoubtedly j u s t the beginning of the l i s t of natural carcinogens, since so few of "nature's pesticides" have been tested (1^,21,22). For example, a recent analysis (26) of lima beans showed an array of 23 natural alkaloids (those tested have b i o c i d a l a c t i v i t y ) that ranged in concentration in stressed plants from 0.2 to 33 parts per thousand fresh weight. None appear to have been tested for carcinogenicity or teratogenicity. Man-made Pesticide Residues. Intake of man-made pesticide residues from food i n the United States, including residues of i n d u s t r i a l chemicals such as polychlorinated biphenyls (PCBs), has been estimated by FDA. They assayed food for residues of the 70 compounds thought to be of greatest importance (27^). The human intake averages about 150 μ g / d a y . Most (105 ug) of t h i s intake i s composed of three chemicals (ethylhexyl diphenyl phosphate, malathion, and chlorpropham) shown to be noncarcinogenic in tests in rodents (1). Thus, the intake of carcinogens from residues (45 yg/day i f a l l the other residues are carcinogenic, which i s u n l i k e l y ) i s extremely t i n y r e l a t i v e to the background of natural substances (1,22). The l a t e s t figures from the FDA about actual exposures don't include every known man-made p e s t i c i d e , but they are a reasonable attempt at doing so. In a recent NRC/NAS report, Regulating Pesticides i n Food (28), i t i s suggested that some of the pesticides not covered by the FDA sampling, p a r t i c u l a r l y those used on tomatoes, should have t h e i r allowable l i m i t s lowered and presumably should be added to the FDA sampling program. Nevertheless, the estimate of 45 \xg of possibly carcinogenic pesticide residues consumed in a day i s l i k e l y to be a reasonable estimate, as i s our conclusion that the possible hazards from these residues are minimal compared to the background of nature's pesticides. To put the amounts of man-made pesticides i n perspective (1_), there are about 500 μg of carcinogens in a cup of coffee (hydrogen peroxide and methylglyoxal), 185 yg of carcinogenic formaldehyde i n a s l i c e of bread, about 2,000 \ig of formaldehyde i n a c o l a , 760 \ig of carcinogenic estragole in a b a s i l leaf, and a gram of burnt material from cooking our food.



0.008

0.03 0.06 0.07 0.1 0.1 0.2

0.003 0.006 0.003 0.03

0.0002* 0.0003* 0.0004

0.0004* 0.0002* 0.0003* 0.008* 0.6 0.004 2.1

0.001* 0.004*

Possible hazard: HERP (%) Environmental pollution Chloroform, 83 μg (U.S. average) Trichloroethylene, 2800 μg

Carcinogen dose per 70-kg person

Trichlorocthylcnc, 267 μg Chloroform, 12 μg Tetrachloroethylene, 21 μg Chloroform, 250 μg (average pool) Swimming pool, 1 hour (for child) Formaldehyde, 598 μg Conventional home air (14 hour/day) Benzene, 155 μg Formaldehyde, 2.2 mg Mobile home air (14 hour/day) e and other residues PCBs, 0.2 μg (U.S. average) PCBs: daily dietary intake DDE, 2.2 μg (U.S. average) DDE/DDT: daily dietary intake Ethylene dibromide, 0.42 μg EDB: daily dietary intake (U.S. average) (from grains and grain products) Natural pesticides and dietary toxins Dimcthylnitrosamine, 0.3 μg Bacon, cooked (100 g) Dicthylnitrosamine, 0.1 μg Urcthane, 43 μg Sake (250 ml) Symphytine, 38 μg Comfrey herb tea, 1 cup (750 μg of pyrrolizidine alkaloids) Aflatoxin, 64 ng (U.S. average, 2 ppb) Peanut butter (32 g; one sandwich) Dimcthylnitrosamine, 7.9 μg Dried squid, broiled in gas oven (54 g) Allyl isothiocyanate, 4.6 mg Brown mustard (5 g) Estragole, 3.8 mg Basil (1 g of dried leaf) Mushroom, one raw (15 g) {Agancus bisporus) Mixture of hydrazines, and so forth Safrolc, 6.6 mg Natural root beer (12 ounces; 354 ml) (now banned) Dimcthylnitrosamine, 1 μg Beer, before 1979 (12 ounces; 354 ml)

Tap water, 1 liter Well water, 1 liter contaminated (worst well in Silicon Valley) Well water, 1 liter contaminated, Woburn

Daily human exposure

Table 1. Ranking possible carcinogenic hazards


(+) 0.2 0.003 (0.2) 96 (?) (?) (436) (0.2)

0.2 (+) 22 (?) (0.2) 0.02 (41) 1.9

0.2

52 20,300 56

(9.6) 13 (5.1)

1.7 (") 1.5

90 941

Mice

941 90 (126) 90 (44) 53 (44)

5

(") (119) 101 (119) 1.5 (157) 1.5

(119) (")

Rats

Potency of carcinogen. TD o (mg/kg)

8

n

25

w

2 ο ο

η

9

Ragsdale and Menzer; Carcinogenicity and Pesticides ACS Symposium Series; American Chemical Society: Washington, DC, 1989. 50

5Q

5 0

Ethyl alcohol, 18 ml Ethyl alcohol, 30 ml Comfrey root, 2700 mg Symphytine, 1.8 mg Food additives AF-2 (furylfuramide), 4.8 μg Saccharin, 95 mg Drugs Phenacetin, 300 mg Metronidazole, 2000 mg Isoniazid, 300 mg Phénobarbital, 60 mg Clofibratc, 2000 mg Occupational exposure Formaldehyde, 6.1 mg Ethylene dibromide, 150 mg (44) (5.1)

1.5 1.5

5 0

(2137) 506 30 5.: (?)

(131) (")

(?) (?) (?) (?)

1246 (542) (150) (+) 169

29 2143

9110 9110 626 1.9

5 0

5 0

5 0

5 0

Potency of carcinogens: A number in parentheses indicates a T D value not used in the calculation of the human exposure-rodent potency (HERP) index because it is the less sensitive species. Symbols: - , negative in cancer test; +, positive for carcinogenicity in test(s) not suitable for calculating a T D ; ?, not adequately tested for carcinogenicity. T D values shown are averages calculated by taking the harmonic mean of the TD s of the positive tests in that species from the Carcinogenic Potency Database. Results are similar if the lowest T D value (most potent) is used instead. For each test, the target site with the lowest T D value has been used. The average T D has been calculated separately for rats and mice, and the more sensitive species is used for calculating the possible hazard. The database, with references to the source of the cancer tests, is complete for tests published through 1984 and for the National Toxicology Program bioassays through June 1986. We have not indicated the route of exposure or target sites or other particulars of each test, although these are reported in the database. Daily human exposure: We have tried to use average or reasonable daily intakes to facilitate comparisons. In several cases, such as contaminated well water or factory exposure to ethylene dibromide (EDB), average intake is difficult to determine, and we give the value for the worst found. The calculations assume a daily dose for a lifetime; where drugs are normally taken for only a short period, we have bracketed the HERP value. For inhalation exposures, we assume an inhalation of 9,600 L/8 h for the workplace and 10,800 L/14 h for indoor air at home. Possible hazard: The amount of rodent carcinogen indicated under carcinogen dose is divided by 70 kg to give milligrams per kilogram of human exposure, and this human dose is given as the percentage of the T D dose in the rodent (in milligrams per kilogram) to calculate the HERP index. (Reprinted with permission from réf. 1. Copyright 1987 American Association for the Advancement of Science.)

"Asterisks indicate HERP from carcinogens thought to be nongenotoxic.

Formaldehyde: Workers' average daily intake EDB: Workers' daily intake (high exposure)

Phenacetin pill (average dose) Metronidazole (therapeutic dose) Isoniazid pill (prophylactic dose) Phénobarbital, one sleeping pill Clofibratc (average daily dose)

[0.3] [5.6] [14] 16* 17*

5.8 140

AF-2: daily dietary intake before banning Diet Cola (12 ounces; 354 ml)

Beer (12 ounces; 354 ml) Wine (250 ml) Comfrcy-pcpsin tablets (nine daily) Comfrcy-pepsin tablets (nine daily)

0.0002 0.06*

2.8* 4.7* 6.2 1.3



228


An Alternative to Using Synthetic Pesticides is to Raise the Level of Natural Plant Toxins by Breeding. It i s not clear that the l a t t e r approach, even where f e a s i b l e , i s preferable. One consequence of disproportionate concern about tiny traces of synthetic pesticide residues, such as ethylene dibromide (_1), i s that plant breeders are developing highly insect-resistant plants, thus creating other r i s k s . Two recent cases are i n s t r u c t i v e . A major grower introduced a new variety of highly insect-resistant celery into commerce. A f l u r r y of complaints to the Centers for Disease Control from a l l over the country soon resulted, because people who handled the celery developed a severe rash when they were subsequently exposed to sunlight. Some detective work uncovered that the pest-resistant celery contained 9000 ppb psoralens (light-activated mutagenic carcinogens) instead of the l e v e l of 900 ppb psoralens present in normal celery (29,30). It i s unclear whether other natural pesticides in the celery were increased as w e l l . Solanine and chaconine (the main natural alkaloids in potatoes) are cholinesterase i n h i b i t o r s and were widely introduced into the human diet about 400 years ago with the dissemination of the potato from the Andes. They can be detected in the blood of a l l potato eaters. Total alkaloids are present in potatoes at a l e v e l of 15,000 Ug per 200-g potato, which i s only about a s i x - f o l d safety margin from the toxic l e v e l for humans (_1). Neither a l k a l o i d has been tested for carcinogenicity. By contrast, the pesticide malathion, the main synthetic organophosphate cholinesterase i n h i b i t o r present in our diet (17 ug/day), has been thoroughly tested and i s not a carcinogen in rodents. Plant breeders have produced an insectresistant potato; however, i t had to be withdrawn from the market because of i t s acute t o x i c i t y to humans, a consequence of higher levels of solanine and chaconine. There i s a tendency for laymen to think of chemicals as being only man-made, and to characterize them as t o x i c , as i f every natural chemical was not also toxic at some dose. Even a recent NRC/NAS report (28) states: "Advances i n c l a s s i c a l plant breeding. . .offer some promise for nonchemical pest control in the future. Nonchemical approaches w i l l be encouraged by tolerance revocations i f more p r o f i t a b l e chemical controls are not a v a i l a b l e . . . . " The report was p a r t i c u l a r l y concerned with some pesticides used on tomatoes. Of course, tomatine, one of the alkaloids in tomatoes, i s a chemical too, and was introduced from the new world 400 years ago. It has not been tested in rodent cancer bioassays, i s present at 36,000 μ g / 1 0 0 g tomato, and i s orders of magnitude closer to the toxic l e v e l than are man-made pesticide residues found on tomatoes. The idea that nature i s benign and that evolution has allowed us to cope with the toxic chemicals i n the natural world (31) i s not compelling (20) for several reasons: ( i ) There i s no reason to think that natural selection should eliminate the hazard of carcinogenicity of a plant toxin that causes cancer past the reproductive age, though there could be selection for resistance to the acute effects of p a r t i c u l a r carcinogens. For example, a f l a t o x i n , a mold toxin that presumably arose early in evolution, causes cancer in trout, r a t s , mice, and monkeys, and probably people, though the species are not equally sensitive (18,32). Many of the common metal s a l t s are carcinogens (such as lead, cadmium, beryllium, n i c k e l , chromium, selenium, and arsenic) despite t h e i r presence during a l l of



14.

AMES


229

evolution. ( i i ) It i s argued by some that humans, as opposed to rats or mice, may have developed resistance to each s p e c i f i c plant toxin or chemical i n cooked food (3_1). This i s u n l i k e l y , because, as discussed (1^,21,22), both rodents and humans have developed many types of general defenses against the large amounts and enormous variety of toxic chemicals in plants (nature's p e s t i c i d e s ) . These defenses include the constant shedding of the surface layer of cells of the digestive system, the glutathione transferases for detoxifying alkylating agents, the active excretion of hydrophobic toxins out of l i v e r or i n t e s t i n a l c e l l s , numerous defenses against oxygen r a d i c a l s , and DNA excision repair. The fact that defenses usually are general, rather than s p e c i f i c for each chemical, makes good evolutionary sense and i s supported by various studies. Experimental evidence indicates that these general defenses are effective against both natural and synthetic compounds (33^), since basic mechanisms of carcinogenesis are not unique to e i t h e r . ( i i i ) Most natural p e s t i c i d e s , l i k e manmade pesticides, are r e l a t i v e l y new to the modern d i e t , because most of our plant foods were brought to Europe within the l a s t 500 years from the Americas, A f r i c a , and A s i a , and vice versa. ( i v ) The argument that plants contain anticarcinogens which protect us against plant carcinogens is i r r e l e v a n t : plant antioxidants, the major known type of ingested anticarcinogen, do not distinguish whether oxidant carcinogens are synthetic or natural i n o r i g i n , and thus help to protect us against both. (v) It has been argued that synthetic carcinogens can be synergistic with each other. However, t h i s is also true of natural chemicals and i s irrelevant to the argument that synthetic pesticide residues in food or water p o l l u t i o n appear to be a t r i v i a l increment over the background of natural carcinogens. TCDD (Dioxin) Compared to Alcohol and B r o c c o l i . Common sense suggests that a chemical pollutant should not be treated as a s i g n i f i c a n t hazard i f i t s possible hazard l e v e l i s far below that of common food items. TCDD is a substance of great public concern because i t i s an extremely potent carcinogen and teratogen in rodents, yet the doses humans are exposed to are very low r e l a t i v e to the effective l e v e l in rodents. It i s analyzed i n some d e t a i l below to i l l u s t r a t e t h i s point. TCDD can be compared to a l c o h o l , as an example. Alcohol i s an extremely weak carcinogen and teratogen, yet the doses humans are exposed to are very high r e l a t i v e to the effective dose i n rodents (or humans). Indeed, a l c o h o l i c beverages are the most important known human teratogen, and the effective (5 drinks/day) dose l e v e l of alcohol i n humans in mg/kg i s s i m i l a r to the l e v e l causing b i r t h defects i n mice. By contrast, there i s no convincing evidence that TCDD i s carcinogenic or teratogenic i n man, though i t i s i n rodents. If one compares the teratogenic potential of TCDD to that of alcohol for causing b i r t h defects, after adjusting for t h e i r potency i n rodents (1_), then a d a i l y consumption of the EPA "reference dose" (formerly "acceptable dose limit") of TCDD, 6 fg/kg/day, i s equivalent i n teratogenic potential to the amount of alcohol ingested d a i l y from 1/3,000,000 of a beer, the equivalent of drinking one beer (15 grams ethyl alcohol) over a period of 8,000 years. A d a i l y s l i c e of bread, or glass of orange j u i c e , contains much more natural alcohol than t h i s . Alcoholic beverages i n man are c l e a r l y carcinogenic, though only one of several tests on ethyl alcohol i n rats was positive (12,13).



230


This test should be replicated as confirmation that ethyl alcohol is the active ingredient, though the evidence for that i s f a i r l y strong (12). A comparison of the carcinogenic potential of TCDD with that of a l c o h o l , adjusting for potency in rodents, shows the equivalence for the TCDD reference dose of 6 fg/kg/day i s 1 beer every 345 years. Since the average consumption of alcohol in the U.S. i s equivalent to more than one beer per day per c a p i t a , the great concern over TCDD at levels in the range of the reference dose seems unreasonable. The assumption of a worst-case l i n e a r dose response, often used for carcinogens, i s not plausible for TCDD, yet extrapolations to man using those assumptions have generated great concern. TCDD binds to a receptor i n mammalian c e l l s , the Ah receptor, and the evidence suggests strongly that a l l of the TCDD effects are through t h i s binding (34). Moreover, there i s a wide v a r i e t y of natural substances that bind to the Ah receptor, and as far as they have been examined they have a l l of the properties of TCDD. A cooked steak contains p o l y c y c l i c hydrocarbons, which bind to the Ah receptor and mimic TCDD. In addition, our diet contains a variety of flavones and other substances from plants, which bind to the Ah receptor. The most interesting of such substances i s indole carbinol (IC), which i s present in large amounts in b r o c c o l i (500 mg/kg), cabbage, cauliflower, and other members of the Brassica family (35). IC induces the same set of enzymes as TCDD (36). When given before aflatoxin or other carcinogens, i t protects against carcinogenesis, as does TCDD (37). However, when i t i s given after aflatoxin or other carcinogens, i t i s a strong promoter of carcinogenesis, as i s TCDD (38). This stimulation of carcinogenesis has also been shown for cabbage i t s e l f (39). When IC i s exposed to acid pH (equivalent to that of the stomach), i t i s converted to a series of dimers and trimers that are similar to TCDD in size and shape. These bind to the Ah receptor, and induce the set of TCDD-inducible enzymes, thus mimicking TCDD (36) (Bradfield, C ; Bjeldanes, L . , University of C a l i f o r n i a , Berkeley, personal communication, 1988). The 360 fg of TCDD/day EPA "reference dose" should be compared with 50 mg of IC per 100 g of b r o c c o l i , (one portion). Since the a f f i n i t y of the indole derivatives i n binding to the Ah receptor is less by a factor of about 8,000, the b r o c c o l i portion might be roughly 20 m i l l i o n times the possible hazard. Though these IC derivatives appear to be much more of a possible hazard than TCDD, i t i s not clear whether at these low doses either i s any hazard. Another study (40) also shows that when sunlight oxidizes tryptophan, a normal amino a c i d , i t converts i t to a v a r i e t y of indoles (similar to the b r o c c o l i IC dimers), which bind to the Ah receptor and mimic the action of TCDD. It seems l i k e l y that many more of these "natural dioxins" w i l l be discovered in the future. Teratogens. The trace amounts of man-made pesticides found in polluted wells or on food should be a n e g l i g i b l e cause of b i r t h defects, when compared to the l e v e l of the background of known teratogens such as alcohol. Most agents causing b i r t h defects would also be expected to be harmless at low doses. Important r i s k factors for b i r t h defects in humans include: age of mother, alcohol, smoking, and r u b e l l a v i r u s . Cooking Food.

The cooking of food generates a

variety

of


mutagens

14.

AMES

231


and carcinogens. The t o t a l amount of browned and burnt material eaten in a t y p i c a l day i s at least several hundred times more than that inhaled from severe outdoor a i r p o l l u t i o n (22). Nine heterocyclic amines, isolated on the basis of t h e i r mutagenicity from proteins or amino acids that were heated in ways that reproduce cooking methods, have now been tested; a l l have been shown to be potent carcinogens in rodents (41,42). Many others are s t i l l being isolated and characterized (41,42). Three mutagenic nitropyrenes present i n d i e s e l exhaust have now been shown to be carcinogens (43), but the intake of these carcinogenic nitropyrenes has been estimated to be much higher from g r i l l e d chicken than from a i r p o l l u t i o n (41,42,44). Gas flames generate N 0 , which can form both the carcinogenic nitropyrenes (1J3,15) and the potently carcinogenic nitrosamines in food cooked in gas ovens, such as f i s h . It seems l i k e l y that food cooked in gas ovens may be a major source of dietary nitrosamines and nitropyrenes.


2

Occupational Exposures. Pesticides could be of possible significance to chemical workers or applicators. Occupational exposures to chemicals can be high and can often be s i g n i f i c a n t (jL,45). The potential carcinogenic hazards to U.S. workers has been ranked using the PERP index (analogous to the HERP index (1) except that U.S. OSHA Permitted Exposure Levels replace actual exposures) (45). The PERP values d i f f e r by more than 100,000 f o l d . For 12 substances, the permitted levels for workers are greater than 10% of the rodent T D dose. P r i o r i t y attention should be given to reduction of the allowable worker exposures that appear most hazardous in the PERP ranking. 5 0

We Cannot Extrapolate Carcinogenesis

Risks

Without

Understanding

Mechanisms

of

Extrapolating Rodent Cancer Test Results to Humans. It i s prudent to assume that i f a chemical i s a carcinogen in rats and mice at the maximum tolerated dose (MTD), i t i s also l i k e l y to be a carcinogen in humans at the MTD. However, u n t i l we understand more about mechanisms, we cannot r e l i a b l y predict r i s k to humans at low doses, often hundreds of thousands of times below the dose where an effect i s observed in rodents. Thus, quantitative r i s k assessment is currently not s c i e n t i f i c a l l y possible (1,17,20). Carcinogenesis Mechanisms and the Dose-Response Curve. The f i e l d of mechanisms in carcinogenesis i s developing r a p i d l y and i s essential for r a t i o n a l r i s k assessment. Both mutations and c e l l p r o l i f e r a t i o n ( i . e . , promotion) are involved in carcinogenesis (_1,46). There i s an enormous spontaneous rate of damage to DNA from endogenous oxidants and methylation, which we have discussed in r e l a t i o n to cancer and aging (47-50). There i s also a basal spontaneous rate for c e l l proliferation ( Ζ Ί Δ ) · Thus, increasing either mutation or c e l l p r o l i f e r a t i o n should frequently be carcinogenic. Additional complications are that several mutations appear necessary for carcinogenesis, and there are many layers of defense against carcinogens. These considerations suggest a sub-linear dose-response r e l a t i o n , which i s consistent with both the animal and human data



232


(51), and that m u l t i p l i c a t i v e interactions w i l l be common in human cancer causation. Most importantly, administering chemicals in cancer tests at the MTD might commonly cause c e l l p r o l i f e r a t i o n (1^,46,52,53). If a chemical i s non-mutagenic and i t s carcinogenicity is due to i t s t o x i c i t y causing c e l l p r o l i f e r a t i o n resulting from near-toxic doses, one might commonly expect a threshold (1,46,52). The fact that high doses of a chemical cause tumors doesn't necessarily mean that small doses w i l l . Most chemicals may, in f a c t , be harmless at low l e v e l s . A l i s t of carcinogens i s n ' t enough. The main rule in toxicology i s that the "dose makes the poison": at some l e v e l , every chemical becomes t o x i c , but there are safe levels below that. In dealing with carcinogens, a s c i e n t i f i c consensus evolved in the 1970s that we should treat carcinogens d i f f e r e n t l y , that we should assume that even low doses could possibly cause some harm, even though we don't have the methods to measure effects at low levels. This idea evolved because most carcinogens appeared to be mutagens, agents that damage the DNA. The precedent of r a d i a t i o n , which i s both a mutagen and carcinogen, gave credence to thinking that there possibly could be effects of chemicals even at low doses. The idea that most of the c l a s s i c a l carcinogens were mutagens damaging DNA (about 90% in our studies) (54,55) and work on oncogenes (56) reinforced the mutagen-carcinogen connection. However, in recent years there has been a change in the p i c t u r e . About half of a l l chemicals tested in animals are carcinogens, and only about half of these appear to be mutagenic. It i s now standard in cancer tests to be rigorous about giving the maximum tolerated dose of the chemical for the lifetime of the animal, and t h i s factor, or the p o s s i b i l i t y that c l a s s i c a l carcinogens comprised a s p e c i a l group, may account for the change in percentages. It seems quite reasonable that non-mutagens cause cancer because dosing at the MTD accelerates the promotional step of carcinogenesis (57,58). Promotion, or c e l l p r o l i f e r a t i o n , can also be accelerated by viruses, such as the human carcinogenic viruses h e p a t i t i s B, a major cause of l i v e r cancer in the world (16), or human papilloma virus 16 (HPV16), a contributor to cancer of the cervix (15). Both cause chronic c e l l k i l l i n g and consequent c e l l proliferation. Promotion can also be caused by hormones, which cause c e l l p r o l i f e r a t i o n . Hormones appear to be major r i s k factors for certain human cancers such as breast cancer (14). The promotional step of cancer causation can also be accelerated by chemicals. Alcohol, for example, causes c i r r h o s i s of the l i v e r leading to cancer. The c l a s s i c a l chemical promoters such as p h é n o b a r b i t a l and tetradecanoyl phorbol acetate would be expected to be, and are i n f a c t , carcinogens in animals when tested thoroughly (58). There i s increasing evidence to show that low doses of promoters are not active (46^,52). It seems l i k e l y , therefore, that a high percentage of a l l of the chemicals in the world, both man-made and n a t u r a l , w i l l be c l a s s i f i e d as carcinogens, but most of these may be acting as promoters and therefore may not be of interest at doses much below the toxic dose ( .1 ). Epidemiological F a l l a c i e s : Storks Bring Babies and Cancer and B i r t h Defects

Pollution

Causes

The number of storks in Europe has been decreasing for decades. the same time, the European b i r t h rate has also been decreasing.


At We


14.

AMES


233

would be f o o l i s h to accept t h i s high c o r r e l a t i o n (59) as evidence that storks bring babies. The science of epidemiology t r i e s to sort out from the myriad chance correlations those that are meaningful and involve cause and effect. However, i t is not easy to obtain convincing evidence by epidemiological methods because of inherent methodological d i f f i c u l t i e s (24). There are many sources of bias in observational data and chance v a r i a t i o n i s also an important factor. For example, because there are so many different types of cancer or b i r t h defects, by chance alone one might expect some of them to occur at a high frequency in a small community. Toxicology provides evidence to help decide whether an observed c o r r e l a t i o n might be causal or accidental. There i s no convincing evidence from epidemiology or toxicology that p o l l u t i o n i s a s i g n i f i c a n t source of b i r t h defects and cancer. For example, the epidemiological studies on Love Canal, dioxin in Agent Orange ( 6 0 , 6 0 , Contra Costa County r e f i n e r i e s (62,63), S i l i c o n Valley (64), Woburn (.1,20, or DDT provide no convincing evidence that in any of these well publicized exposures p o l l u t i o n was the cause of human harm. Even in Love Canal, where people were l i v i n g next to a toxic waste dump, the epidemiological evidence for an effect on public health i s equivocal. Analysis of the toxicology data on many of these cases suggests that the amounts of the chemicals involved were much too low r e l a t i v e to the background of natural and t r a d i t i o n a l carcinogens to be credible sources of increased cancer to humans ( 1). A comparative analysis of teratogens using a HERP-type index expressing the human exposure l e v e l as a percentage of the dose l e v e l effective in rodents would be of i n t e r e s t , but i t has not been done in a systematic way. Environmental exposure to ethylene dibromide and other pesticide pollutants i s thousands of times lower than the exposure to these same agents in the workplace (1,45) (Table 1). Thus, i f parts per b i l l i o n of these pollutants were causing cancer or b i r t h defects, one might expect to see an effect in the workplace. The studies on these chemicals so far do not provide any evidence for a causal association (32), though epidemiological studies are inherently insensitive. Historically, cases of cancer due to workplace exposure resulted mainly from exposures to chemicals at very high l e v e l s . For example, the EDB levels that workers were allowed to be exposed to were shockingly high (Table 1). (I t e s t i f i e d in C a l i f o r n i a in 1981 that our calculations showed that the workers were allowed to breathe in a dose higher than the dose that gave half of the rats cancer.) C a l i f o r n i a lowered the permissible worker exposure over 100-fold. Despite the fact that the epidemiology on EDB in highly exposed workers does not show any s i g n i f i c a n t e f f e c t , the uncertainties of our knowledge make i t important to have s t r i c t rules about workers because they can be exposed to extremely high doses. Technology is Not Doing Us In Modern technologies are almost always replacing older, more hazardous technologies. The reason that b i l l i o n s of pounds of the solvents TCE (one of the most important i n d u s t r i a l non-flammable solvents) and PCE (the main dry-cleaning solvent in the U . S . ) are used i s because of t h e i r low t o x i c i t y and the fact that they are not flammable. Is it advisable to go back to the age when industry or dry cleaners used



234


flammable solvents and were frequently going up in flames? Eliminating a carcinogen may not always be a good idea. For example, ethylene dibromide (EDB), the main fumigant in the U.S. before i t was banned, was present in t r i v i a l amounts in our food: the average d a i l y intake was about one-tenth of the possible carcinogenic hazard of the aflatoxin in the average peanut butter sandwich, a t r i v i a l r i s k in i t s e l f (1) (Table 1). Elimination of fumigation results in insect infestation and subsequent contamination of grain by carcinogen-producing molds. This might r e s u l t in a regression in public health, not an advance, and would also greatly increase costs. The coming a l t e r n a t i v e s , such as i r r a d i a t i n g food, could be more hazardous than EDB, as well as more expensive. S i m i l a r l y , modern pesticides replaced more hazardous substances such as lead arsenate, one of the major pesticides before the modern era. Lead and arsenic are both n a t u r a l , highly t o x i c , and carcinogenic. Pesticides have increased crop yields and brought down the price of foods, a major public health advance. Every l i v i n g thing and every industry "pollutes" to some extent. How much does society wish to spend to get r i d of the l a s t part per b i l l i o n of TCE out of wells in S i l i c o n V a l l e y , or PCE from drycleaning plants? We are currently spending enormous amounts of money trying to eliminate lower and lower amounts of p o l l u t i o n ; one estimate i s about 80 b i l l i o n d o l l a r s annually (20); for comparison, the amount spent on a l l basic s c i e n t i f i c research i s $9 b i l l i o n . The fact that s c i e n t i s t s have developed methods to measure parts per b i l l i o n (one part per b i l l i o n i s one person in a l l of China) of carcinogens and are developing methods to measure parts per t r i l l i o n does not mean that s i g n i f i c a n t p o l l u t i o n i s increasing, or that the p o l l u t i o n found i s a cause of human harm. Conclusion Everyone knows that spending a l l of one's effort on t r i v i a without focusing on important problems is counterproductive. If we divert too much of our attention to traces of p o l l u t i o n and away from important public health concerns such as smoking (400,000 deaths per year), alcohol (100,000 deaths per year), eating unbalanced diets (e.g., too much saturated fat and c h o l e s t e r o l ) , AIDS, radioactive radon coming up from the s o i l into our homes, and high-dose occupational exposure, we do not improve public health, and the important hazards are l o s t i n the confusion. It i s the inexorable progress of modern technology and s c i e n t i f i c research that w i l l continue to provide the knowledge that w i l l result i n steady progress to decrease cancer and b i r t h defects and lengthen l i f e s p a n . Acknowledgments I wish to thank Lois Gold and David Freedman for helpful discussion and criticisms. This work was supported by NCI Outstanding Investigator Grant CA39910 and by NIEHS Center Grant ES01896. This paper has been adapted from r e f . 65. Literature Cited 1.

Ames, Β. N.; Magaw, R.; Gold, L . S.

Science 1987, 236, 271-280.


14. AMES 2. 3.

4. 5.


6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25.

26.


235

Gold, L. S.; Bernstein, L . ; Magaw, R.; Slone, T.H. Environ. Health Perspect. 1989, in press. Gold, L. S.; Sawyer, C. B.; Magaw, R.; Backman, G. M.; de Veciana, M.; Levinson, R.; Hooper, N. K.; Havender, W. R.; Bernstein, L . ; Peto, R.; Pike, M. C.; Ames, Β. N. Environ. Health Perspect. 1984, 58, 9-319. Gold, L. S.; Slone, T. H.; Backman, G.; Magaw, R.; Da Costa, M.; Ames, Β. N. Environ. Health Perspect. 1987, 74, 237-329. Innes, J. R. M.; Ulland, B. M.; Valerio, M. G.; Petrucelli, L . ; Fishbein, L . ; Hart, E. R.; Pallotta, A. J.; Bates, R. R.; Falk, H. L . ; Gart, J. J.; Klein, M.; Mitchell, I.; Peters, J. J. Natl. Cancer. Inst. 1969, 42, 1101-1114. Schardein, J. L . ; Schwetz, Β. Α.; Kenal, M. F. Environ. Health Perspect. 1985, 61, 55-67. Lipkin, M. Cancer Res. 1988, 48, 235-245. Yang, C. S.; Newmark, H. L. CRC Crit. Rev. Oncol./Hematol. 1987, 7, 267-287. Pence, B. C.; Buddingh, F. Carcinogenesis 1988, 9, 187-190. Reddy, B. S.; Cohen, L. Α.; Eds. Diet, Nutrition, and Cancer: A Critical Evaluation, Vols. I and II; CRC Press: Boca Raton, FL, 1986. Joossens, J. V.; Hill, M. J . ; Geboers, J . ; Eds. Diet and Human Carcinogenesis ; Elsevier Science Publishers B.V.: Amsterdam, 1986. Ames, Β. N. Review of Evidence for Alcohol-Related Carcinogenesis. Report for Proposition 65 Meeting, Sacramento, CA, December 11, 1987. IARC Monographs on the Evaluation of Carcinogenic Risks to Humans : Alcohol Drinking, Vol. 44, International Agency for Research on Cancer: Lyon, 1988, in press. Henderson, Β. E.; Ross, R.; Bernstein, L. Cancer Res. 1988, 48, 246-253. Peto, R.; zur Hausen, H.; Eds. Banbury Report 21. Viral Etiology of Cervical Cancer; Cold Spring Harbor Laboratory: Cold Spring Harbor, NY, 1986. Yeh, F.-S.; Mo, C.-C.; Luo, S.; Henderson, Β. E . ; Tong, M. J . ; Yu, M. C. Cancer Res. 1985, 45, 872-873. Ames, Β. N.; Magaw, R.; Gold, L. S. Science 1987, 237, 235. Ames, Β. N.; Magaw, R.; Gold, L. S. Science 1987, 237, 12831284. Ames, Β. N.; Gold, L. S.; Magaw, R. Science 1987, 237, 13991400. Ames, Β. N.; Gold, L. S. Science 1987, 238, 1634. Ames, Β. N.; Gold, L. S. Science 1988, 240, 1045-1047. Ames, Β. N. Science 1983, 221, 1256-1264. Doll, R.; Peto, R. The Causes of Cancer; Oxford University Press: Oxford, England, 1981. Higginson, J. Cancer Res. 1988, 48, 1381-1389. Peto, R. In Assessment of Risk from Low-Level Exposure to Radiation and Chemicals; Woodhead, A. D.; Shellabarger, C. J.; Pond, V.; Hollaender, Α.; Eds.; Plenum: New York, 1985; pp. 316. Harborne, J. B. In Natural Resistance of Plants to Pests. Roles of Allelochemicals (ACS Symposium 296); Green, M. B.; Hedin, P. Α.; Eds.; American Chemical Society: Washington, D.C., 1986; Ch. 3.


236


27.

Gartrell, M. J.; Craun, J. C.; Podrebarac, D. S.; Gunderson, E. L. J. Assoc. Off. Anal. Chem. 1986, 69, 146. National Research Council, Board on Agriculture. Regulating Pesticides in Food; National Academy Press: Washington, D.C., 1987. Berkley, S. F.; Hightower, A. W.; Beier, R. C.; Fleming, D. W.; Brokopp, C. D.; Ivie, G. W.; Broome, C. V. Ann. Intern. Med. 1986, 105, 351-355. Seligman, P. J.; Mathias, C. G. T.; O'Malley, Μ. Α.; Beier, R. C.; Fehrs, L. J . ; Serrill, W. S.; Halperin, W. E. Arch. Dermatol. 1987, 123, 1478-1482. Davis, D. L. Science 1987, 238, 1633-1634. IARC Monographs on the Evaluation of Carcinogenic Risks to Humans. Overall Evaluations of Carcinogenicity: An Updating of IARC Monographs Volumes 1-42 (Supplement 7); International Agency for Research on Cancer: Lyon, France, 1987. Jakoby, W. B., Ed. Enzymatic Basis of Detoxification, Vols. I and II; Academic Press: New York, 1980. Knutson, J. C.; Poland, A. Cell 1982, 30, 225-234. Bradfield, C. Α.; Bjeldanes, L. F. J. Agric. Food Chem. 1987, 35, 46-49. Bradfield, C. Α.; Bjeldanes, L. F. J. Toxicol. Environ. Health 1987, 21, 311-323. Dashwood, R. H.; Arbogast, D. N.; Fong, A. T.; Hendricks, J. D.; Bailey, G. S. Carcinogenesis 1988, 9, 427-432. Bailey, G. S.; Hendricks, J. D.; Shelton, D. W.; Nixon, J. E . ; Pawlowski, Ν. E. J. Natl. Cancer. Inst. 1987, 78, 931-934. Birt, D. F.; Pelling, J. C.; Pour, P. M.; Tibbels, M. G.; Schweickert, L . ; Bresnick, E. Carcinogenesis 1987, 8, 913-917. Rannug, Α.; Rannug, U.; Rosenkranz, H. S.; Winqvist, L . ; Westerholm, R.; Agurell, E.; Grafstrom, A.-K. J. Biol. Chem. 1987, 262, 15422-15427. Sugimura, T.; Sato, S.; Ohgaki, H.; Takayama, S.; Nagao, M.; Wakabayashi, K. In Genetic Toxicology of the Diet; Knudsen, I., Ed.; Alan R. Liss: New York, 1986; pp. 85-107. Sugimura, T. Science 1986, 233, 312-318. Ohgaki, H.; Hasegawa, H.; Kato, T.; Negishi, C.; Sato, S.; Sugimura, T. Cancer Lett. 1985, 25, 239-245. Kinouchi, T.; Tsutsui, H.; Ohnishi, Y. Mutation Res. 1986, 171, 105-113. Gold, L. S.; Backman, G. M.; Hooper, N. K.; Peto, R. Environ. Health Perspect. 1987, 76, 211-219. Pitot, H. C.; Goldsworthy, T. L . ; Moran, S.; Kennan, W.; Glauert, H. P.; Maronpot, R. R.; Campbell, H. A. Carcinogenesis 1987, 8, 1491-1499. Park, J.-W.; Ames, Β. N. Proc. Natl. Acad. Sci. USA 1988, 85, 7467-7470 and 85, 9508. Ames, Β. N. In Medical, Biochemical and Chemical Aspects of Free Radicals 1988; Yoshikawa, T., Ed.; Elsevier Science Publishers, B.V.: Amsterdam, 1989, in press. Adelman, R.; Saul, R. L . ; Ames, Β. N. Proc. Natl. Acad. Sci. USA 1988, 85, 2706-2708. Richter, C.; Park, J.-W.; Ames, Β. N. Proc. Natl. Acad. Sci. USA 1988, 85, 6465-6467.

28. 29. 30.


31. 32.

33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50.


14. AMES 51. 52. 53. 54. 55. 56.


57. 58. 59. 60.

61. 62.

63.

64.

65.


237

1987 Annual Cancer Statistics Review Including Cancer Trends: 1950-1985 (NIH Publication No. 88-2789); National Institutes of Health: Bethesda, MD, 1988. Farber, E. Environ. Health Perspect. 1987, 75, 65-70. Swenberg, J. Α.; Richardson, F. C.; Boucheron, J. Α.; Deal, F. H.; Belinsky, S. Α.; Charbonneau, M.; Short, B. G. Environ. Health Perspect. 1987, 76, 57-63. McCann, J.; Choi, E . ; Yamasaki, E . ; Ames, Β. N. Proc. Natl. Acad. Sci. USA 1975, 72, 5135-5139. McCann, J.; Ames, Β. N. Proc. Natl. Acad. Sci. USA 1976, 73, 950-954. Stowers, S. J.; Maronpot, R. R.; Reynolds, S. H.; Anderson, M. W. Environ. Health Perspect. 1987, 75, 81-86. Butterworth, Β. E.; Slaga, T. J . ; Eds. Banbury Report 25. Nongenotoxic Mechanisms in Carcinogenesis; Cold Spring Harbor Laboratory: Cold Spring Harbor, NY, 1987. Iversen, O. Η., Ed. Theories of Carcinogenesis; Hemisphere: Washington, D.C., 1988. Sies, H. Nature 1988, 332, 495. Lathrop, G. D.; Machado, S. G.; Karrison, P. G.; Grubbs, W. D.; Thomas, W. F.; Wolfe, W. H.; Michalek, J. E . ; Miner, J. C.; Peterson, M. R.; Ogerskok, R. W. Air Force Health Study: Epidemiologic Investigation of Health Effects in Air Force Personnel Following Exposure to Herbicides. First Follow-Up Examination Results; U.S. Air Force: Brooks Air Force Base, TX, 1987. Gough, M. Dioxin, Agent Orange: The Facts; Plenum Press: New York, 1986. Austin, D. F.; Nelson, V.; Swain, B.; Johnson, L . ; Lum, S.; Flessel, P. Epidemiological Study of the Incidence of Cancer as Related to Industrial Emissions in Contra Costa County, California (NTIS Publication No. PB84-199785); U.S. Government Printing Office: Washington, D.C., 1984. Smith, A. H.; Waller, K. Air Pollution and Cancer Incidence in Contra Costa County: Review and Recommendations. A Report Prepared for the Contra Costa County Department of Health Services; 1985. California Department of Health Services, Epidemiological Studies and Services Section, Pregnancy Outcome in Santa Clara County, 1980-1985; California Department of Health Services: Berkeley, CA, 1988. Ames, Β. N. In Important Advances in Oncology 1989; DeVita, Jr., V. T.; Hellman, S.; Rosenberg, S. Α.; Eds.; J. B. Lippincott: Philadelphia, 1989; pp. 237-247.

RECEIVED

July 7, 1989


Carcinogenicity and Pesticides - ACS Publications - American

Recommend Documents