Health Effects of Dioxins:
An Interview with Dr. Linda Birnbaum.
Excerpted from the Environmental Review Newsletter Volume One Number Eight, August 1994
Dr. Linda Birnbaum received the Ph.D. in
microbiology from the University of Illinois in 1972 and is
currently director of the Environmental Toxicology Division in the
Health Effects Laboratory of the U.S. Environmental Protection
Agency. She manages a research group of about one hundred
scientists, twenty students and post doctoral fellows at the
E.P.A. laboratories in Research Triangle Park, North Carolina. We
spoke with Dr. Birnbaum about the health effects of exposure to
the family of chemicals called dioxins.
ER: Dr. Birnbaum, what is dioxin?
LB: By dioxin-related compounds or dioxin itself we mean 2,3,7,8-tetrachlorodibenzoparadioxin (TCDD), which is the most toxic congener. [Congeners are related chemical substances. ed.] There are approximately one hundred chlorinated dioxins, seventy-five chlorinated dibenzofurans and 209 PCBs. [Dioxin and TCDD are used interchangeably in this article. ed.]
ER: Dioxins are actually a family of related compounds.
LB: Right. TCDD and related compounds work by binding to a specific cellular protein called the Ah or the dioxin receptor. Interaction of dioxin with this cellular receptor is necessary - it is the first step but it is not sufficient - to bring about dioxin effects.
ER: Where does dioxin come from?
LB: Nobody ever made dioxin because they wanted it, except perhaps
in a research lab. It is a byproduct of different industrial and
some combustion processes. If you look in core samples prior to
1900, you cannot detect any dioxins. Its presence in the
environment is a function of the industrial use of organics and
chlorines. [Organic chemicals are based on carbon atoms. ed.] Dioxin levels in the environment and in human samples peaked in
the late 1960s and early 1970s and have come down since then.
We need to recognize that some of the regulatory moves made in the late 1970s have had major impacts leading to reduction of the levels of these compounds.
ER: What does dioxin do at toxic levels?
LB: At high levels dioxin causes death in every species that has been looked at. If you give a guinea pig an oral dose of a microgram of TCDD per kilogram of body weight you will kill it. [One microgram is one millionth of a gram. ed.] You have to administer about 5,000 times that amount to kill a hamster.
ER: Are you familiar with Dixie Lee Ray, a former governor of Washington State?
LB: Yes. Wasn't she also head of a nuclear regulatory committee?
ER: Right. Chair of the Atomic Energy Committee under Nixon. She wrote a book where she says "No human has ever become chronically ill or died from dioxin exposure in the U.S."
LB: That is not true. There is data now on people in the United States who were occupationally exposed to dioxins who have decreased circulating testosterone levels and have increased incidence of diabetes. There are people in the United States who have had severe acne for forty-five years.
ER: On page eighty-nine she says "The only proven human illness to result from dioxin exposure is chloracne - a severe skin rash that is curable with proper treatment - and possibly a short term nerve disfunction."
LB: First of all, chloracne is not curable. The only treatment we know of is time. And what that means is that over time enough dioxin will eventually leave your body that you may get below the level where those effects occur. There is no treatment for chloracne.
ER: She wrote this book in the late 1980s. Were the testosterone level and diabetes results known then?
LB: Some of these studies showing effects on circulating male hormone levels and on diabetes have only been very recently published. So she would not have known that.
ER: What is your approach to the possibility of human health effects of dioxins?
LB: I am a conservative biologist in the sense that I believe that people are animals and that as animals, we have common ancestry. Nature - when it works out a mechanism - tends to use that mechanism over and over. So unless you have reason to suspect that something is not happening in humans that happens in animals, it likely is. When I see a chemical like dioxin which causes effects in multiple tissues at multiple stages of development in both sexes of multiple species, I am confident that people are going to also respond to it.
When you deal with a chemical that you see only one kind of response, maybe in one tissue of one sex of one species, that kind of chemical I do not have confidence that it is going to affect humans. You have to prove to me that things are not going to happen to people before you assume they do.
We do have quite a bit of information now about people responding to dioxin in the same way that animals respond to dioxin. For example, for biochemical effects, we know now that
this dioxin or Ah receptor is present in people as well as other animals. Some recent work from the University of Toronto where they looked at placentas from 150 women has shown there is a distribution of the strength of binding of dioxin to these receptors in different people. There are some people who are more sensitive and some people who are less sensitive. That is perfectly reasonable.
ER: There is usually a distribution of traits in any population.
LB: Right. People are not a highly inbred population. You expect variability in the population. Anyhow, we do not have any evidence that anyone ever died from dioxin or dioxin-like chemicals.
ER: Dr. Ray said in an industrial accident in 1976, 37,000 people in Seveso Italy had acute exposure and suffered no long-term adverse effects and no cancer.
LB: We do not know that they suffered no long-term adverse effects. In fact, there was a publication within the last six months showing increased cancer in that population. You could not see increased cancer when she looked. It was too soon.
ER: Because there's a long latency with cancer.
LB: Right. And some of the other effects of dioxin are very subtle kinds of effects that you would not see unless you looked for them. We do not expect dioxin to cause obvious birth effects in children. There is a big difference in species sensitivity to dioxin. But the dose that kills a fetal hamster and a fetal guinea pig is the same.
ER: So there is a developmental difference in response to dioxin?
LB: No. The point is that for some reason, adult hamsters do not die from dioxin and adult guinea pigs do. We don't understand all the steps that lead up to death, but there is something different. Most rats die after a dose somewhere between ten and one hundred micrograms per kilogram. [ten to one hundred parts per billion. ed.] But there is another strain of rats maintained in Kupio, Finland, that you have to go to doses like the hamster to kill it [one hundred times more.ed.] But if you look at dioxin effects on the developing embryo, if you look at biochemical changes, if you look at lots of other changes - the two rat strains behave identically. So there is something in this Kupio strain of rats which blocks the death response. With dioxin for any given response you may have a species or a strain that is an outlier. But most animals appear to respond similarly for most responses.
ER: You mentioned birth defects...
LB: Cleft palate is a sensitive response following prenatal exposure to dioxin in mice, and that occurs in levels where you do not see any adverse effects on the mother. If we go to rats,
hamsters, or guinea pigs, we do not see cleft palate except at doses where we are actually killing the fetus. So for some reason the mouse fetus is very sensitive to cleft palate; the other species are not.
I do not expect to see cleft palates in humans. We can take out from the embryo the region that is destined to become the palate. We can take that out of an embryonic mouse, an embryonic rat, or an embryonic human. And in culture, I can treat that tissue with TCDD and prevent palate fusion. I would get a cleft palate in vitro. And what I can see is that the amount of TCDD that causes clefting in the rat and the human is identical, and I need a lot more to do it in the rat and the human than I do in the mouse.
So they looked at children when they were born to women who were exposed in Seveso. Were these kids of normal size? Did they have their arms and their legs and their eyes? And the answer is yes. Well, those are't the kinds of effects we would expect. There have been two major PCB poisonings in the Far East - one in Japan in 1968 and one in Taiwan in 1979. In both of these populations, rice oil was contaminated by PCBs, by transformer fluid, and the PCBs were contaminated by dibenzofurans. So you had a mixed exposure. There were stillbirths. There was definite embryo and fetal toxicity in those populations. There were children born with chloracne. There were children born with effects on their skin and their nails just like we see in monkeys whose mothers were exposed to dioxin. Some of the children were small for size, just like we see in animals who were exposed. And there were a number of other adverse effects on these kids. Many of these effects are correlated with the amount of dibenzofurans that the women ate. So we do have some clear cut episodes of human poisonings to dioxin-like chemicals.
ER: The dibenzofurans are related to the dioxins - they are dioxin-like, you would call them.
LB: Right. The proper name for these compounds is polychlorinated dibenzo-dioxins or polychlorinated dibenzo-furans. And if you saw the structures lined up next to each other, even if you did not know chemistry, you would say, yes they look alike. If you do structural modeling of these molecules you can overlay them.
ER: Dr. Ray made no mention of the Japanese and the Taiwan information in her book. Was this information available to her?
LB: Absolutely. But many people have said, Well, I can't look at that because that was not dioxin. The point is, you have to look at it because many of these chemicals, while they are not 2-3-7-8 TCDD, have the same mechanism of action and bring about the same effects. That has been a more recent agreement among many scientists that we have to look at all of these related chemicals. If we only look at TCDD, we could miss the boat.
ER: Again, dioxins are a family of related nasty compounds.
LB: It is a family of related chemicals, and many of them do the same thing the same way. So you have to look at the sum total of these. A scheme has been developed called the toxic equivalency factor scheme, or toxic equivalency approach. This has been accepted by the international community as well as EPA where you give dioxin (TCDD) a value of one, and then you give the other related compounds some value which is a fraction of one. Then to calculate the total potency of the mixture of dioxins, you would multiply that potency factor times how much of it is in the mix. And this scheme works very well.
There have been a couple of complex mixtures of dioxins and people have done chemical analysis and said what is in the mixture and totaled it up and said, OK, this mixture is equivalent to so much TCDD. Then they dosed animals with the mixture, and they dosed animals with the equivalent amount of TCDD, and they got the same response.
So except when you have something like an industrial poisoning as occurred in Seveso Italy, where the primary exposure is to TCDD, our environmental exposure is to a whole mixture of these dioxin-like compounds.
ER: What is the human body burden of dioxins?
LB: How much is in us? The numbers are not real hard, but I would say the numbers in the U.S. population today are in the neighborhood of three to five parts per trillion in the blood on a lipid adjusted basis.
ER: Lipid adjusted means dioxin is not water soluble?
LB: It is very insoluble in water. It is very fat soluble.
ER: So it accumulates in fat cells?
LB: Correct. Or into fatty environments. You calculate up from that amount of TCDD. If you do it on a toxic equivalency approach you go to the neighborhood of about fifty to sixty parts per trillion dioxin equivalents. If you put that on a whole body burden instead of on a lipid adjusted basis, it comes out that we are in the neighborhood of about ten parts per trillion of toxic equivalencies.
ER: How are we exposed to dioxins?
LB: The primary route of exposure for people is food. We estimate that greater than ninety percent is from the very low levels present in the food, and it is primarily from animal products. In this country it is primarily beef, dairy, and then a little bit of fish, a little bit of poultry, little bit of pork.
ER: From the fat in the animals. How does dioxin get into those animals?
LB: It accumulates up the food chain. Much of the dioxin today is
getting into the environment through combustion processes so it is emitted up smokestacks or chimneys. It is a sticky molecule; very water insoluble. It attaches to particles, gets blown around in the air, and settles down. If it settles on a river, it goes to the sediment and can get into the fish who get eaten by the birds and by us. If it settles on the land or the grass, it gets taken up by cattle and so on.
ER: It accumulates in their fat?
LB: It accumulates in their fat and then we eat them. We (EPA with USDA and FDA) are looking at some dioxin levels in cattle. But we do not have the kinds of measurements we need to find out whether cattle grazing near an incinerator, for example, would have higher dioxin levels in their body than cattle grazing far away from incinerators. We do not have that information.
If you lived in Scandinavia, for example, or if you were a heavy fish eater, more of your body burden would come from fish.
ER: So there's no particular part of the environment that is clean if you wanted to hide from it?
LB: The recommendations from Department of Health and Human Services for a healthy heart are to eat more fruits and vegetables.
ER: Eat lower on the food chain.
LB: That is correct. Less animal fat. That is supposed to be good for your heart. It would also help lead to a lower body burden of dioxins.
ER: What is the effect if we are carrying ten parts per trillion in our bodies? I am thinking of the recent paper on endometriosis in monkeys.
LB: The levels of dioxin in the monkeys that developed endometriosis were about five to ten times that. And if I tell you that the level that you and I have is ten parts per trillion, that is the mean of the population. There are people in the population who are much higher than the mean, just within the distribution of the normal population, and that is not talking about populations which we know are heavily exposed.
I think the endometriosis result is of concern because we know endometriosis is a disorder not only of the hormone system, but also it is related to immune dysfunction. And one of the things that dioxin does is mess up the immune system.
ER: In the older literature, dioxin effects on the immune system
is one of the more clear cut findings.
LB: We have to be real careful here. Because in just about any species, high levels of dioxin do cause atrophy of the lymphoid tissue. But at much lower levels, for example in mice, you can see suppression of a number of different immune system effects. You see a block in the ability to develop an antibody response to antigen challenge. You see effects on cell mediated immunity. You see effects on host defense to tumors and parasites and viruses. There are a number of different effects of dioxins.
ER: So different aspects of the immune system seem to be affected?
LB: Correct. In mice. There is some data in monkeys showing that monkeys appear to be very sensitive to the immunosuppressive effects of dioxin. Rats are very different. Levels that are not overtly toxic to the rat do not cause suppression of the immune system. However, if we go to prenatal exposure, we find that prenatally exposed rats may be even more sensitive than prenatally exposed mice. So the difference may have something to do with the state of the immune system when you expose it to dioxin.
The data in humans on effects on the immune system is very weak. Researchers have not done the right tests. They are just starting to do that now. And there is some suggestion that children who have been born to women who have higher levels of dioxin-like compounds in their body do appear to have higher incidence of respiratory disease, ear infections, and a suggestion of some changes in the kinds of T cells they have. This data is pretty new, and not necessarily hard.
We have been trying to urge some people conducting epidemiology studies to consider doing some vaccination challenges. In other words, if people respond like the mice and like the monkeys, if you challenge them with a vaccine they have never seen before, they might have trouble mounting a response. The immune effects of dioxin appear to be very sensitive. Recent data from our lab has shown that a dose of ten nanograms per kilogram (that is what human body burdens are), if you give mice ten nanograms per kilogram of dioxin, and a week later you challenged them with influenza, they die. Normally they would not die.
ER: They would not die. Would they get sick?
LB: They might get sick, but they recover.
ER: So their immune system quit on them?
LB: We are not sure it is directly an immune system effect. What many of us believe is that dioxin is disrupting growth control and regulation of growth and differentiation.
Dioxin functions like a hormone in that hormones have multiple effects which are going to be tissue specific and developmental stage specific, and some effects may even be species specific. If you look at dioxin with that kind of mind set, then
the variation you see in response becomes very understandable.
ER: We do not understand how dioxin works but we can understand why it may work many different ways.
LB: We can put it in perspective. Dioxin is not like cyanide which is flat out toxic and does the same thing anywhere you give it. Cyanide is not a hormone. It has a direct toxicity.
ER: Can you explain the Rier paper on endometriosis in monkeys?
LB: The monkeys were exposed to five or twenty-five parts per trillion dioxin in their food for four years. At the end of that time there was no more dioxin in their food, but that does not mean of course that it was not in their bodies. The animals were held for an additional seven to ten years.
ER: Why was that necessary?
LB: It was not necessary. There were other studies being done on these animals: immune responses and the animals were bred to see if there were any problems with fertility. After seven years, one of the high dose animals died. The animal appeared in pain, had problems with menses. And when they did an autopsy, they found it had fulminating endometriosis.
ER: What is fulminating endometriosis?
LB: People do not get fulminating endometriosis, but monkeys do. It is where you have a massive response.
ER: With cells growing all over the body cavity?
LB: Yes. About a year later another high dose monkey died and they found the same thing. And at that point a team of veterinarians and gynecologists did laparoscopic surgery on all the monkeys. They found a dose-related increase in both the incidence and severity of endometriosis in the monkeys.
There had been a report from the Canadians about a PCB exposure of monkeys where they seemed to see increased endometriosis. And then there were two reports in the literature suggesting that women with endometriosis have higher levels of persistent organochlorines in their blood. But that was not specifically PCBs or dioxins. It was organochlorines in general.
Now I will tell you two pieces of information which may turn out to be very important. There is a cohort of Rhesus monkeys that were at Bowman Gray School of Medicine that was inadvertently exposed to PCBs in the late 1970s. Those monkeys are just in the process of being sacrificed, and we are having histology done on all of them. [Histology is microscopic examination. ed.] We are seeing some endometriosis. I cannot tell you yet if there is going to be more in these monkeys than there are in control animals. I don't have that information. But we are looking at this experiment of nature.
ER: Was there long-term exposure?
LB: If you are exposed to dioxins and PCBs, the exposure itself does not have to be long term because these chemicals are persistent. So you could have a single poisoning and then you are exposed internally the rest of your life.
ER: So one chronic exposure would do it?
LB: Right. One acute exposure can lead to chronic exposure.
Now the other thing is that in the pharmaceutical industry, there is a rodent model for endometriosis. In the rat you can surgically plant pieces of uterine tissue in the peritoneum and then watch what happens to that tissue over time. [The peritoneum is the lining of the body cavity. ed.] And if you treat the animal with estrogens, the implanted tissue grows faster. If you remove the ovaries, which basically removes the estrogen, it grows slower.
ER: This is a rodent model for human disease?
LB: Correct. Rats and mice do not normally get endometriosis because to get endometriosis you have to be able to get backflow of menstrual fluid. You cannot get that in rats and mice. It has to do with the shape of the uterus. But you can initiate endometriosis by taking pieces of endometrial tissue.
In our laboratory, Audrey Cummings in our reproductive biology group has gotten that model to work in the rat. And she has also developed it in the mouse, which no one has done before. In both animals we can see that if we treat them with dioxin, we promote the growth of these endometrial lesions. We are currently repeating the study. We have seen dose response, and we have done the experiment in both species. [Dose response means symptoms increase as the dioxin dose is increased. ed.] I am pretty confident about the results but we are repeating it to be one hundred percent sure. And again, with endometriosis being related to immune dysfunction, it all makes sense.
ER: How does it make sense?
LB: If we have immune dysfunction going on from dioxin, that may be part of the reason dioxin is promoting the endometriosis.
ER: So the immune system is not cleaning up these renegade cells that are growing where they should not be?
LB: What the human fertility doctors tell me is that while usually there is some pain associated with endometriosis, severity of pain is not a good indicator of extent of endometriosis. In fact women with the most severe endometriosis often have the least pain. And that is because the pain is associated with the inflammatory response. And the more severely affected women do not have a big inflammatory response because they tend to be immunosuppressed.
ER: Can we apply this result from rhesus monkeys to humans?
LB: I get very nervous when people want to use monkeys as models for people willy nilly. But when you consider reproductive function, in fact the monkeys are very good models. They have similar kinds of placenta, they have similar kinds of hormonal cycling. They are much more similar to humans than rats and mice.
On the other hand, if you are looking for metabolism of a certain drug or chemical, sometimes mice or rats will metabolize it the same way people do - not necessarily monkeys. But when you are looking at this kind of reproductive effect, I think monkeys are in fact a good model.
ER: What else do we know about the toxicity of dioxin?
LB: Dioxin affects just about every system you can think of at one point in time or another.
ER: Neural tissue?
LB: That is not clear. Exposure of adult rats, mice and monkeys to dioxin has not been shown to cause any neural behavioral effects. Prenatal exposure to rhesus monkeys has been reported to have an effect on learning. And there is some data with prenatal exposure not to dioxin, but to some of the dioxin-like PCBs in mice suggesting some subtle effects on behavior.
Some of the non dioxin-like PCBs and complex PCB mixtures do appear to have neural behavior affects. But whether dioxin and dioxin-related compounds definitely have effects on the nervous system is not yet clear. However, during development, high levels of the dioxin or Ah receptor are present in a developing nervous system.
ER: In her book Dixie Lee Ray said the toxic effects of dioxin have been exaggerated. She said "Dioxin has been called the most toxic chemical known to man, even though no deaths or serious harm to humans can be attributed to it."
LB: I think the reason it is called the most toxic chemical is because such extremely low doses cause adverse affects. Vernon Houk, who had been head of one of the health units at the Center for Disease Control and had been the federal official responsible for the evacuation of Times beach, ended up saying three or four years ago that he believed that the reason we had not seen deaths in people from dioxin was because, thank God, people had not been exposed to enough.
The point is that the extreme toxicity of dioxin is based on the fact that such extremely low levels cause effects. Even though you and I have dioxins in our body, again, we are talking about trace kind of amounts. Twenty years ago we would not have known it was in our bodies because we would't have been able to measure it.
ER: Right. Our analytical instruments keep improving.
LB: That's right. Many researchers are beginning to look at dioxin more like lead, in that there are going to be biological responses at almost any level of exposure, and many of these responses are not going to be that bad. But in aggregate, exposures result in an individual who is let us say, more stressed or less competent.
ER: Less immuno-competent.
LB: Less immuno-competent, perhaps less reproductively competent. With lead exposure, you may get a five point shift in the I.Q. for the worse. And on an individual basis you will never pick that up. How would you know that your I.Q. was five points lower than it should have been? You cannot know that. But the point is, if the whole population shifted five points to the left, that has major implications for the population. For example, people are saying, what does a twenty percent decrease in testosterone mean? For an individual male it may not mean much, but for the population as a whole, it may mean that you have more men with fertility problems.
The data indicates that dioxin increases type II diabetes. Age is a more important risk factor than dioxin, but if you consider the whole population, maybe more people are going to get diabetes at a younger age - or more people overall than would have gotten it before. With endometriosis - maybe we are just nudging the population so that more people are going to be affected.
But remember, even if we now say dioxin, we mean the whole family of dioxin-like compounds. We are not exposed just to dioxins. We are exposed to benzene and butadiene and heavy metals and pesticides. And many of these have effects on for instance, our sugar metabolism or hormone levels or something else. [Diabetes is a disorder of sugar metabolism. ed.] We need to have more of an aggregate perspective on toxicant exposure.
ER: This goes back to the idea of a toxic equivalency scheme.
LB: Right. But that toxic equivalency is just focusing on dioxin-like effects.
ER: I was thinking of applying that to the whole spectrum of toxins.
LB: People have suggested that kind of thing but frankly, that scientifically is less than satisfactory. The comfort that scientists have in the toxic equivalency factor approach for dioxin-like compounds is that we have a structurally related class of chemicals that work by the same mechanism and do the same thing. [They all bind to the Ah receptor. ed.]
ER: With lead and other toxins, mechanisms of action will be different.
LB: That is right. But in a sense, do you care? If your I.Q. is down five points, do you care whether it is lead or dioxin? Scientific issues can be extraordinarily complex. There are no simple answers to difficult questions. I get frustrated about a lot of people who are quick to comment without having done any work in the area.
Copyright 1994 Environmental Review