The Kuiper paper is an extensive comparison using binding and gene activation assays with the two estrogen receptors. Estrogen receptor beta was only discovered in 1996. We’d done some work in 1997 on estrogen receptor alpha and they just combined everything and directly compared the two receptors and showed that for most compounds, they were fairly similar. There’s been lots of work done since, and the early paper is cited because it was the first extensive comparison with synthetic estrogens, steroidal compounds, and hydroxy PCBs.

PCBs had been banned back in the 1970s, but there was as much scientific and regulatory activity in 1994 as when they were banned. There’s still as much activity, in fact even more. Although PCB levels are decreasing in biological tissues, albeit slowly, there’s still a huge concern about the potential adverse effects of PCBs. What that review did was summarize PCB science: What do they do? What are the possible adverse effects? What do we know for sure? What are the uncertainties? In terms of the structure, activity, and comparing PCBs with dioxins, I developed the so-called dioxin equivalents—or toxic equivalency factors—of PCBs compared with TCDD, which is the most toxic dioxin. I hypothesized that this might be a useful way of doing risk assessment of PCBs, and this was relatively novel. We had done a lot of quantitative structure activity studies of various PCBs, and so I used our data and results from other laboratories. As a result, I could estimate that a compound has a toxic equivalency factor of .1, which means it is one-tenth as active as TCDD. I assigned these factors to all the dioxin-like PCBs, and subsequently others have refined this approach and used it for risk assessment of PCBs. That’s why a lot of people cite that paper, but in addition I pointed out that these dioxin-like PCBs are not necessarily the only concerns. There have been some studies suggesting that non-dioxin-like PCBs might also be neurotoxic. That’s an area that is still being extensively investigated.

Whether the effects in humans are real or not is the key question. I have a problem with the assumption that they are, because the results tend to be inconsistent between studies. What is consistent is that many papers seem to come up with an effect, usually as a result of in utero or in early postnatal exposures to PCBs. There’s been some confusion as to what problems these compounds actually do cause. That’s an area being investigated by several groups. To be frank, when it comes to this question of whether PCBs cause neurodevelopmental deficits in children, that still requires further study. I’m not convinced either way, but it’s an important problem to resolve. One problem is that the measured responses differ in many studies. If scientists can’t get a correlation with total PCBs, they’ll look at dioxin-like PCBs and so on. These inconsistencies must be resolved.

A decade ago we knew these compounds could cause a tremendous number of different responses. They induce all kinds of biochemical responses in animal models and in cell culture models. How that relates to human effects at high doses and low doses is unclear. Even the good neurotoxicology data that suggests that PCBs could have a neurodevelopmental effect require administration of high doses in animal models. Even though humans are obviously exposed to more PCBs than they want or need, the concentrations are still relatively low. What these low doses do, if anything, has still not been worked out. My opinion is that they aren’t very significant, but others disagree, and that’s why the science continues. The important thing is that we recognize that this kind of chemical, which is highly fat soluble and highly resistant to degradation, should not be used in applications where it can get into the environment. In effect, we’re still living with the legacy of what happened before the 1970s when PCBs were banned. The current leakage of these compounds into the environment is minimal, but because these compounds are so long-lived, it’s going to be a long time before they’re eliminated. So in a sense, the right regulatory decisions were made in the 1970s and all we can do is minimize our current exposures.

I started out in the 1970s, doing environmental studies on PCBs and dioxins.. I worked my way through the chemistry, the toxicology, and, in a sense, the molecular biology and biochemistry of PCBs. We developed structure-activity relationships and then my PCB work was minimal. We started back on PCB research with hydroxy PCBs. Currently, my research is focused on the molecular biology of hormone action and in cancer. Most of our work now is totally cancer- or molecular biology-related.

Well, that’s an interesting story. We got into breast cancer research mainly because these dioxin-like compounds and PCBs that are dioxin-like exhibit anti-estrogenic activity. That’s not necessarily a good response, unless you have breast cancer. You can actually administer dioxin or one of these dioxin-like PCBs to an animal with a mammary tumor and tumor growth is inhibited. We wanted to understand the mechanism of how dioxin-like compounds block estrogen action and estrogen-induced mammary tumors. We then tried to develop analogues that are relatively non-toxic but still anti-estrogenic. Subsequently we developed different analogues, which work through different pathways. They’re generally active not only in breast cancer, but also in colon, prostate, pancreatic, and a number of other cancers. These compounds are potent inhibitors of tumor growth, and this all resulted from our studies with dioxin and dioxin-like PCBs that have this unusual property of being anti-estrogenic.

We currently do a lot of cancer-related research on many tumors, and find that dioxin-like compounds that work through the aryl hydrocarbon receptor—the dioxin receptor—exhibit anticancer activity. This receptor was initially discovered as a protein that bound all these toxic compounds. Now we know that this receptor binds all sorts of phytochemicals and other naturally-occurring chemicals that are chemo-protective. It looks like potential drugs that can interact with this receptor may be useful for tumors other than breast cancer. We’ve done some work on prostate and pancreatic cancer that looks promising. So this receptor, which was initially a target for environmental contaminants, is now a potential receptor for drug development.

Getting grants and papers turned down is always the most trying, particularly when you think something is good and the reviewers don’t. I think that’s the most trying time for all scientists.

Our PCB research illustrates the way that science sometimes works. We investigated the chemistry, toxicology, and molecular biology of PCBs and we were interested in them because they were an important group of environmental contaminants. We now find that these studies have led us to the development of new potential anti-cancer compounds. Hopefully this research will make some positive contribution to science and to human health.

Stephen H. Safe, D. Phil.
Texas A & M University
College Station, TX, USA