Friday, September 26, 2008

Just another brick in the wall: more on bisphenol A


My neighbor, the “real” doctor, called the other day, asking for “The Neighborhood Toxicologist.”

“So, what are you doing about your bicycle bottles,” she asked.

She’d just read the latest study and related commentary on the potential dangers of bisphenol A in the
Journal of the American Medical Association. It’s rare that I get to advise Katta, most often it’s me calling her – how does Sophie’s staph infection look? What do you think of this little black spot on my arm? I just called an ambulance for Ben, do you think you could come take a look at him while we wait?

I leaned into my expertise. “Well,” I said, “you know those aren’t
polycarbonate. It’s just the polycarb that has bisphenol A. Those bicycle bottles are polyethylene,” I said with some authority – impressing myself with my own recall. “As far as I know no-one’s found anything bad about those,” I pause, “not yet anyway.” Not unless you consider the filmy black crude (I’m guessing something biological rather than chemical) that inevitably coats the insides of those bicycle bottles – even if all you’ve ever had in them is water.

What’s confusing about the polycarbonate issue is that it provides s a perfect (or maybe imperfect) opportunity for the public to crab about the wishy-washyness of scientists. Most folks just want an answer – yea or nay, good or bad. But with bisphenol A you get two conflicting answers from two federal organizations, the FDA and the National Toxicology Program.

While the National Toxicology Program (under the National Institute of Environmental Health Sciences) concludes, as far as anyone can conclude, that bisphenol A “is of “some concern” for effects on development of the prostate gland and brain and for behavioral effects in fetuses, infants and children” (for details check out their final report, NTP-CERHR Monograph on the Potential Human Reproductive and Developmental Effects of Bisphenol A ,
) the FDA gives the A-okay all-clear for the chemical. According to their recently issued draft report, “…FDA concludes that an adequate margin of safety exists for BPA at current levels of exposure from food contact uses for infants and adults .”

So what gives? The FDA’s overall findings suggest that the available studies are “inadequate” (problems with dosing, species, timing – you name it.) It’s true that all of these can impact the outcome and that even the very best study on a particular contaminant can be rendered relatively irrelevant because the concentrations say, were screamingly high (for example beyond those anyone would ever be exposed to unless they ate their pretty blue bottles); or that the method of exposure is irrelevant (say, injecting a chemical – essentially mainlining it – rather than feeding it to experimental animals); or the so-called mechanism of action – how a chemical causes toxicity – is unique to a particular test species (though this one goes both ways – the sedative thalidomide offers a tragic example of why chemicals need to be tested in several different species.)

Unfortunatley, sometimes we just have to do the best with what we’ve got when it comes to data. Sometimes knowing what’s lacking informs experimental design, so studies that are “most appropriate” can be done. While I won’t review the review that reviewed the review (FDA’s most recent
draft) I would like to point out that there are no conflicts about BPA’s femininity. The chemical is indeed estrogenic – scientists knew that long before it ever became a part of those polycarbonate bottles. Estrogen, as we all know is a pretty powerful hormone.
And estrogenic chemicals can bind with, and activate estrogen receptors (referred to below as ERα and ERβ) which means that, like estrogen, they can also elicit all or some of the biological outcomes triggered by estrogen.

But contaminants like BPA must compete with both estrogen in the body and other ingested estrogens, here’s
FDA again, “In fact, BPA has an approximately 1000 - 10,000 fold lower affinity for ERα and ERβ as compared to E2, whereas genistein, a phytoestrogen, has a much higher affinity than BPA for ERα and ERβ. Accordingly, if equal concentrations were available, the assumed order of binding to the ERs would be E2, genistein, and then BPA.”

Here’s where even I’m a little confuzuled as my daughter used to say. Though I hesitate to reveal my ignorance – and I do pledge to take this on and fully understand the implications one day – are they saying that it doesn’t matter that BPA binds a powerful receptor because there are several other more “natural” chemicals that will beat it out? When we know that too much estrogen, or estrogen exposure at the “wrong time” could be bad (what I mean by “wrong time” is that there are times during say, development – particularly development in the male when natural concentrations of estrogen may be very low)? Why not take the cautious approach that adding another estrogen to the mix could also be bad – particularly one that is apparently easy to avoid – stop using BPA containing bottles (although that still leaves can linings.)

What follows is an excerpt from a
review by Alex Vidaeff and Lowell Server explaining why just knowing the relative potency of estrogens isn’t necessarily enough:
“It has been said that xenoestrogens and phytoestrogens, being weak estrogens with a low level of environmental contamination, are not sufficient to produce adverse effects. The opinions were mainly based on the observations derived from DES-exposed cohorts where only “sufficient” doses of DES generated adverse effects
[71] . Such considerations, based on an estrogen potency threshold, or dose-response effects, may underestimate environmental estrogens activity. Hazard identification and assessment in this area cannot rely solely on linear measurements of estrogen activity. Undoubtedly, the xenoestrogens are weaker estrogens than estradiol or even estriol, but studies focusing on binding activity may overlook the complexity of ER action as described above, and the fact that factors other than the binding affinity of the ligand for the receptor may affect gene expression…... When vom Saal et al. [70] observed an increase in prostate size after prenatal exposure to estrogens in mice, the dose-response curve was an U-shaped curve, whereby lower doses also resulted in larger effects. This supports the possibility that even low doses of estrogen in fetal life may affect the expression of genes involved in the morphogenesis of the prostate gland and possibly other genital tissues.”


And then there’s that JAMA article. What alarmed Dr. Katta wasn’t the squabbling over laboratory studies or the reproductive and developmental impacts in rats – but the more recent finding that very real concentrations of bisphenol A in human urine samples (yes we drink the stuff in and pee it out in small but measurable amounts) was positively associated with heart-disease and type 2 diabetes in adult humans in addition to the prostate and brain effects which are of concern to the National Toxicology Program.


But remember, an association is just that – the two things tend to travel together. In this case those with more BPA in their urine tended to have a higher incidence of disease but that doesn’t mean disease was caused by BPA – maybe those with more disease just eat more canned food compared with fresh potentially healthier food (can lining is another source of BPA.) It will take further laboratory studies to confirm any cause and effect linkages. But what’s notable about the study was that there are already rat data linking the chemical to insulin resistance – which in turn is key in the development of type 2 diabetes.


If you’ve read to this point – you must, by now get the idea of how complicated it can be to figure these things out. Oh only if we could just sit a bunch of infants down and have them chug warm milk from polycarb bottles – and then wait and see what happens.
Oops we’ve already done that.