Bruce AmesBruce AmesIn late spring, CH&S Editor Carl Gottschall visited Bruce Ames, professor of biochemistry and molecular biology and director of the National Institute of Environmental Health Sciences Center at the University of California at Berkeley. During their discussion of research in mutagenesis, toxicology, cancer prevention, and the degenerative diseases of aging, they were joined by Ames' colleague, Lois Swirsky Gold, director of the Carcinogenic Potency Project at U. C. Berkeley and the Lawrence Berkeley National Laboratory.
Carl Gottschall: I'd like to begin by asking you to repeat a little history. Tell me about the steps that led to the Salmonella test for mutagenicity, better known as the Ames Test.
Bruce Ames: I've always been half a geneticist and half a biochemist. I was mutating bacteria all of the time to change genes. At some point, actually when I was reading labels on potato chip bags, I realized that people weren't thinking about the possible consequences, if one of the many chemicals being used as food additives were a mutagen.
As a hobby I decided to develop a test for detecting mutagens and, as happens with a lot of hobbies, it ended up taking over my life. We sucessfully developed a bacterial test for detecting mutagens; that got me interested in carcinogens, which got me interested in aging.
Gottschall: What are your current interests and activities?
Ames: My main interest is in DNA damage (e.g., mutations): what's causing it and what the consequences are for cancer. In recent years I've gotten more interested in aging because cancer is one of the degenerative diseases of aging. Aging and the degenerative diseases associated with it have an oxidative component in common. In my laboratory, we are spending a lot of time trying to understand aging, and we're making progress.
There is a tremendous amount of oxidative damage just from living. Living is like getting irradiated, because when you generate energy what you're doing is burning fat or carbohydrates, which means you're pulling electrons off them. You add the electrons to oxygen to generate energy. If you add four electrons to oxygen you're home safe to water; if you add them one at a time you make superoxide, hydrogen peroxide, and hydroxyl radical, the very substances you get from radiation. We've shown that oxidative damage is a major component in the carcinogenic process, and it looks like a major part of heart disease also has an oxidative component. The same is true for cataracts and brain dysfunction. It isn't the only factor, but it's a major factor.
I've also been very interested in nutrition, because if you eat your fruits and vegetables you have half the cancer rate-for nearly all the major types of cancer-of people who don't eat fruits and vegetables. Gladys Block, an epidemiologist here at Berkeley, has reviewed about 200 epidemiological studies from all over the world in which people were asked about diet and cancer. If you compare the cancer rates of the quarter of the population eating the fewest fruits and vegetables (zero or one portion a day) and the quarter eating the most (four or five portions a day), the group eating the fewest has more than double the cancer rate for practically every type of cancer. Just a few portions of fruits and vegetables make a huge difference in cancer risk.
The protection afforded by fruits and vegetables appears to extend to people who smoke: the excess risk of lung cancer is cut in half. We think this is because vitamins and antioxidants come from fruits and vegetables: Vitamin C, carotenoids, and a good part of Vitamin E. Cigarette smoke is high in oxidants and depletes the body's antioxidants. But that's not the only story because you require 44 micronutrients including copper, manganese, vitamins, essential fatty acids, and essential amino acids. If you leave out any one of them, you can distort your biochemistry in complicated ways that can lead to cancer, heart disease, and premature aging.
We've been particularly interested in antioxidants, but other micronutrients are also important. For example, folic acid deficiency breaks chromosomes, and a sizable percentage of the population suffers from folate deficiency. If a pregnant woman doesn't have enough folate, her child has an increased risk of neural tube birth defects. Among older people who don't get enough folate, cognitive deficiencies can occur. There is a lot of evidence from various directions that folate deficiency affects brain function.
Gottschall: What supplemental vitamins or antioxidants do you take?
Ames: I'm not a big believer in megadoses of vitamins. My wife is Italian and I eat a good Mediterranean diet with lots of fruit and vegetables and not too much meat or too much fat . When I remember, I take a supplement with the minimum daily requirement of everything. I think that's insurance. Vitamin pills are being taken by people who eat the best diet. The poor aren't taking vitamin pills or eating adequate fruits and vegetables.
Gottschall: What else do we know about risk factors for cancer?
Ames: The public is being mislead about the causes of cancer. We know the important preventable factors are nutrition and smoking. We know that caloric intake has a big effect on cancer in rodents. Feed rats too many calories and you have a huge cancer rate compared with that of rats restricted for calories. Much of cancer prevention is in our own hands. For example, the Seventh Day Adventists (whose religion emphasizes not smoking and eating a good diet) have half of the cancer rate of the average American.
However, the public is being inundated with stories that cancer is caused by exposure to low levels of some industrial chemical at a few parts per billion. The problem, I think, is that there are a lot of wrong assumptions in the interpretation of animal cancer tests. People knew that high doses of industrial chemicals could give cancer to people. For example, -naphthyl amine gave a high percentage of the exposed workers bladder cancer. But doses were huge. So we know synthetic chemicals can give cancer. Then people jumped to the conclusion that synthetic chemicals caused most of the cancer in the world, which turns out to be wrong. Testing of these chemicals was begun in rats, which is reasonable because you don't want workers to be guinea pigs. Half the synthetic chemicals tested came out positive in high-dose rat tests. What was their conclusion? That synthetic chemicals are causing most of the cancer in the world.
Unfortunately, the people doing the tests didn't run the controls: 99.9% of the chemicals humans ingest are natural. Every plant is filled with hundreds of chemicals that plants make to kill off insects; these are nature's pesticides. People thought that natural chemicals could not possibly induce cancer, but if you look at the natural chemicals that have been tested, half of them come out positive in high-dose cancer tests.
There is a flood of papers suggesting that these results can be a high-dose artifact. Basically, when you use the maximum-tolerated dose (MTD) of the chemical, as in a rodent cancer test, you're killing cells, getting a lot of cell division, and that's a risk factor for cancer. You're also getting inflammation, and that's a risk factor for cancer. These tests tell you nothing about low-dose risks. And if you only test synthetic chemicals you're only going to worry about synthetic chemicals, but if you test chemicals in broccoli, half of them will probably come out positive. Of the 1000 natural chemicals in a cup of coffee, 28 have been tested and 19 of them came out positive. That doesn't mean a cup of coffee is dangerous. It just shows that carcinogens, as defined by high-dose tests in rodents, are everywhere.
Gottschall: What about pollution as a cancer cause?
Ames: Pollution mostly seems a big distraction to me. Specific occupations might cause a few percent of cancer, mostly high doses of asbestos in heavy smokers and in shipyard workers during World War II. We're wiser now about asbestos than we were in the old days, but we were fighting a war and you had to insulate those ships, so asbestos was a very useful material. But you don't want to get it out of every school in the country when the amounts people are exposed to are tiny; you may cause more harm then good. All the soil in California has little bits of asbestos. It's a matter of economics, how to save the most lives for the money and not to distract people with trivia.
Gottschall: That brings up the question of how you allocate resources within a finite pie.
Ames: Resources are limited; every economist knows that. Whenever I talk with environmentalists, they say "Oh, take it out of the Defense Department." But again people get confused. If there are 1000 potential risks, we need to know which ones are important and which aren't. We should be drumming into people's heads that when it comes to cancer prevention in the U.S., there are two really important things: eating good diets and not smoking. After we figure out how to get the most mileage out of that, we can worry about second-order effects.
Gottschall: I am glad you could join us, Dr. Gold. How can we convince people that the risks are arranged in some sort of a hierarchy?
Lois GoldLois Gold: In my judgment, an educational effort is needed about the scientific evidence from epidemiological studies showing that the major causes of cancer are from our own behavior: smoking and diet. I think people often externalize and want to assign blame. To recognize that the known causes of cancer require us to alter personal behavior is unpleasant. In combination with that is the historical development of a mind-set that views causality in terms of social factors. It's education, it's occupation, or something outside of us-a sociological phenomenon-that explains problems. This contributes to accepting the unproven assumption that pollution is important in cancer causation.
Ames: So we're all victims. The previous generation didn't think that way at all; they thought if they didn't succeed, it was their fault. But in any case, what's really bizarre is that Lois and I have been kicking out the foundations of what we (and others) think is a house of cards, and it's still standing with no foundation.
Gold: No, it's being chopped away. Look at the recently proposed new EPA guidelines (see "Regulatory Update," p. 8), which recognize that in order to estimate risk from low-dose human exposures more is required than the standard high-dose rodent test. We need a better understanding of mechanism of carcinogenesis to evaluate a potential hazard at doses hundreds of thousands of times lower than the rodent dose.
Ames: Okay, we're slowly winning the scientific wars, but the public's been brainwashed for 20 years. Rachel Carson wrote Silent Spring, a diatribe against DDT. She never once mentioned that DDT was one of the most valuable chemicals ever invented by man. It saved 30 million lives, and Paul Müller won a Nobel Prize for it. You could argue that penicillin was more valuable, but DDT was one of the triumphs of humanity and it never hurt a person as far as any epidemiology has ever shown. The levels of exposure don't make sense as a likely cause of human cancer. One doesn't want to kill eagles, but today we're in the fifth generation of pesticides after DDT. To say it was an awful thing for humanity is without scientific basis, but it's that mind-set again.
Gold: There is also a paradigm that modern industrial chemicals can be "toxics." But this ignores the fact that any chemical is toxic at some dose. The results of animal cancer tests are at near-toxic doses, but there is a failure to recognize how limited these results are for understanding effects in humans at tiny doses.
Ames: There is an idea that humanity is just now changing the world, but for eons Earth has been changing, and comets have been hitting our planet and wiping out the dinosaurs. The Earth has never been static; everything is evolutionary.
Gold: And then something like HIV appears, and suddenly whole populations are at terrific real risk of premature death.
Ames: We must put risks into perspective; people can't keep 1000 risks straight. You have to help them to put risks into perspective based on their order of importance.
Gottschall: Don't you think part of the difficulty is that people will say that it's one thing to accept risks voluntarily-I choose to smoke, drink, and act with reckless abandon-but these other things I don't have any choice about and therefore they should be weighted differently?
Ames: Yes, but if you don't know, for example, that a cup of coffee is chock-full of carcinogens, how do you have voluntary and nonvoluntary risks? Life is complicated trade-offs, you can't just intervene and think that you're doing good.
If you serve people coffee at dinner parties, should they sue you because you are feeding them carcinogens? Where does this end? Should the government be regulating how many pats of butter each of us can eat? In the end you have to say important public health policies like chlorinating the water is a government responsibility. But by the time you start getting into trivia, you get special interest groups who want you to do some things that don't make any sense.
Gottschall: Would you briefly explain your HERP (human exposure/rodent potency) list?
Ames: We compared possible hazards for daily exposures to rodent carcinogens using an index (HERP) that relates human exposure to carcinogenic potency in rodents. We found that residues of synthetic pesticides or environmental pollutants rank low when they are compared with the large background of naturally occurring chemicals in common foods. We also found a similar result when we compared 32 average daily exposures to natural pesticides and synthetic pesticide residues in the diet. (Ed. note: Additional information is available in Gold, L. S.; Slone, T. H.; Stern, B. R.; Manley, N. B.; Ames, B. N. "Rodent carcinogens: setting priorities"; Science 1992, 258, 261-65; or electronically on the WWW at http://potency. berkeley.edu/cpdb.html).
Gottschall: These results provide a hierarchy of risks rather than an absolute index of risks?
Ames: It's more a ranking for where to put your attention. You would certainly put your attention on cups of coffee before you'd put your attention on dietary residues of pesticides. You get more carcinogen in one cup of coffee than from a year's worth of potentially carcinogenic pesticide residues. Pesticides just aren't something that inherently seem very interesting as possible causes of human cancer. We know that pesticides lower the price of fruits and vegetables, which can have a huge effect on lowering cancer rates because people buy more when they are cheaper.
Gold: There's also the fact the cancer death rates aren't going up, in spite of the common perception that they are or that there is an epidemic of cancer. In fact, cancer death rates have come down 14% since 1950 if you exclude lung cancer (90% of which is due to smoking), and life expectancy has increased.
Ames: Yes, cancer death rates are going down.
Gottschall: What are your thoughts on current research and regulatory agency funding?
Ames: We need to remember that technology is making us healthy and wealthy, and we're very optimistic that we're making inroads in understanding aging. Biomedical research is taking off, and yet it's starving for research money. We're putting 15 times as much money into EPA regulations, which seem to be mostly counterproductive when you look at them. So how do you get reasonable rules that provide appropriate protection to the public but not shoot yourself in the foot, as we've done so far? Is Superfund really worth doing anyway? It takes $30 billion a year, but 80% goes to lawyers and consultants, and there is no incentive to get the job done.
Gold: But I want to interject the optimistic side of this. I think that the strict worst-case scenario, linear extrapolation is going out of vogue. There is a shifting in the paradigm among regulators and the scientific community that says one really has to know much more about a rodent, high-dose carcinogen detected under specific experimental conditions, before you can say it is a cancer risk at a low level in humans. The uncertainties are enormous.
Gottschall: Lois, you've touched on another question that I wanted to raise. Along with this linear extrapolation from high doses to zero is the belief, suggestion, or in some cases, the statute that there is no threshold.
Ames: Think about the way the body's defenses work. Every time you eat a new plant you are getting 100 new chemicals. How do we deal with that? Our liver is a marvelous defense factory against chemical insult. If you ingest an oxidizing agent, whether synthetic or natural, you induce antioxidant defenses. Each defense system is inducible: You make more of it if you use it. You can take human cells in culture, give them a little radiation, and they adapt and become more resistant at high doses. The same thing is true in rats, and they're also more resistant to hydrogen peroxide. Our bodies work by general defenses against chemical insult so we are buffered.
If you're going to get a big dose of radiation, maybe you're better off getting a little every day so you induce your defenses, but it's all very complicated. I am not advocating that everybody get a little dose of radiation. I work in the lab all the time; if I go out and burn myself on a beach, then that's serious. Maybe a little daily sunshine exposure to work up a tan would prevent a burn.
Gold: There are two types of extrapolation from experimental results in rodents to human exposures: one a qualitative extrapolation between species and the other a quantitative extrapolation from high to low dose. I think it takes some examples for people to understand that there may be thresholds and that a more biological basis for risk assessment is necessary. One example is chemicals such as d-limonene that produce kidney tumors only in male rats, based on 2u-globulin nephropathy, which is a species-specific phenomenon. This illustrates that qualitative species extrapolation may not be warranted. For extrapolation from high to low dose, examples of a threshold are bladder tumors secondary to production of urinary tract calculi and subsequent inflammation with regenerative hyperplasia; for example, uracil. Another example of thresholds is bladder tumors in male rats from high doses of sodium saccharin that are produced by cytotoxicity and increased cell division.
Ames: I personally do not think that linear response is credible. Linear just seems very "unbiological" to me because we have all these inducible defenses. EPA is linearly extrapolating down 740,000 times from the dose that they tested in a rat, and they draw the line in the sand. That's a pretty big extrapolation off into the darkness.
Gold: So without ever doing a cancer test, you can just take the maximum-tolerated dose and divide it by 740,000. You'll come up with an estimate equivalent to a regulatory, one-in-a-million, virtually safe dose, and you will be right within a factor of 10. This demonstrates both the importance of data on exposure levels and how little information is provided by the bioassay results.
Ames: But remember that they're only looking at a few chemicals that they test. Think about caffeine again. You get 100 mg of caffeine in a cup of coffee; there is not a good cancer test on caffeine in mice, and yet we're comparatively close to a toxic level. So if you look at the natural world, you see all these things we are getting huge doses of, and nobody's done a cancer test on them. Yet we're worried about parts-per-billion levels of synthetic chemicals. No toxicologist can really defend the idea that we should be more worried about synthetic rather than natural chemicals. It's all wrong assumptions, one after another.
Gottschall: The expense of going through all these tests for all these different chemicals is, of course, prohibitive. What about results extrapolated from a microbe to a mouse to a man as well as computer simulations?
Ames: If you understand things very well, you can do some computer simulations within that small group, but because of the liver metabolism and jumping from rats to people, it's not going to tell us the answers. I think epidemiology is good at telling big risks, and that's what we care about.
Gottschall: That's what we should care about.
Ames: For the government to be effective in cancer prevention, it should be putting money into figuring out how to get people, especially the poor, to eat good diets and stop smoking. Instead, it is putting vast amounts of money into these regulations, many of which just don't make sense.
We know that huge doses of chemicals can be dangerous, so if you're buying a bottle of pesticide and dipping your hands in it, you can get a high dose. California has the strictest rules in the world on occupational exposure to pesticides, and we should have strict rules.
But there's no human activity where there aren't accidents and deaths. Do you want to ban all the cars because there are accidents? What about airplanes? Each new generation of technology is safer. Horses are more dangerous and put out a hundred times more pollution per mile than cars. Airplanes are much safer than cars per mile. Somehow people forget that it is technology that's making the world safer, not worse.
Gold: But you need rules.
Ames: You need rules, but the question is how to get rules so the bureaucratic monopolies don't take over. I believe that it's important to make the process as competitive as possible. This will stimulate innovation and we will save more lives for the money.
Gottschall: What about the argument that those people who are involved with something shouldn't make the decisions, because they are too close to the issues to be impartial? Ultimately, of course, this leads to people who know nothing about the subject making decisions.
Ames: Bureaucracy's self-interest is to expand its turf rather than to solve the problem, and certainly not for the least amount of money. That's why you want to decentralize government and make it competitive. You can't always do that, so it does come to economic decisions. The current precautionary principle to slow down technology is a recipe for disaster, because technology makes us wealthier and healthier.
Gold: But you do need rules, because self-interest doesn't always dictate the safest advice. We have shown, for example, that for many rodent carcinogens, the exposures that are permitted in the workplace are close to the dose that gave tumors to rodents (in mg/kg/day).
Ames: Even without government regulations, we have ways of controlling these problems. Lawsuits, for example, help. If a company has bad gas tanks or if they poison workers, they are sued. The potential for litigation is an incentive for control of problems.
Gottschall: Is there any other area that you think CH&S readers would find interesting?
Ames: It's important to remember that life expectancy keeps increasing. I think if people behave rationally it's going to get longer and do so faster than anybody thinks.
Gold: And the science is just blossoming now. People will know so much more in ten years than we do now.
Ames: Biomedical research is fiercely competitive and is improving and creating new ways to address our problems. It's better health through chemistry.
Gottschall: I think it's important to stress that not just longevity is increasing; so is the quality of life throughout those years.
Ames: My aunt is about to have her 100th birthday, and my father-in-law just turned 93. I am 67, and I feel that I am doing the best science that I've ever done. If they behave moderately, people are living long productive lives. If they drink or smoke too much or if they don't eat any vegetables, they're going to do themselves in, but there's no amount of government manipulation that's going to help that.
Gottschall: So you endorse the concept of everything in moderation?
Ames: Even moderation!
ACS Home Page