Hormone messengers work every second of your life to keep your body functioning. They regulate a variety of cell processes, stimulating cells to release chemicals, send electrical or chemical signals, divide, produce proteins and, during foetal development, even determine what function a cell will ultimately have.
All endocrine signalling in the body begins with the same step: a hormone binds to a specific hormone receptor, which is embedded either in the cell membrane, or floats inside the cell’s cytoplasm or nucleus. Binding changes the shape of the receptor, changing how the receptor interacts with other molecules in the cell, initiating or halting a biochemical process.
The concern with endocrine disrupting chemicals (EDCs) is that man-made substances may also be able to initiate and halt cellular processes. Because EDCs are introduced to the body by the environment, and not made as part of the body’s own signalling needs, they may initiate, halt or prevent change to physiological processes at the wrong times.
In an adult, wrong signals at the wrong time can cause problems for health. In the foetus, because hormones control how cells develop into organs, signalling errors can cause permanent alterations to tissue arrangement and behaviour. Hence there is concern about implications of EDCs for health.
There is, predictably enough, scepticism about whether or not EDCs pose a significant health threat. A common argument, which we partially addressed in last month’s H&E, is that natural hormones “swamp” any possible effects by EDCs.
The reasoning goes that, since data shows most EDCs have only around 1/1000th of the affinity of native hormones for binding to receptors, then an EDC has to be present in concentrations 1,000 times higher than endogenous hormones if it is to bind to enough receptors to induce the same effect as the hormone it mimics.
The argument assumes these concentrations are much higher than actually exist and concludes that, even if EDCs are having an effect, it must be so small in comparison to the signals being sent by native hormones that its consequences for physical function and health are negligible.
In the previous issue of H&E, we presented evidence that receptors on the outside of cells (membrane receptors) are much more receptive to some man-made molecules than the better-studied nuclear receptors. In this case the sceptical conclusion is false, even if the significance of the consequences is not yet clear.
Below we will see why the swamping argument is unsound, even in cases where an EDC has a much lower affinity for a receptor than does a natural hormone, and examine why this means standard toxicological testing protocols are worthless for assessing harm from ED effects.
Big responses from small doses
At any given time, a signalling molecule may or may not be attached to its receptor. They bump into each other, they might attach or they might not, and if they are attached may break apart. When thousands or millions of hormones and receptors are constantly bumping into each other, an equilibrium is eventually established. During equilibrium, a certain percentage of receptors at any given moment are attached to hormones, though precisely which ones are attached among those many thousands changes constantly.
It is where this equilibrium is established which determines a signalling molecule’s affinity for a receptor. Affinity is said to be low if, at a given concentration, it occupies relatively few receptor sites. Affinity is high if at the same concentration a higher number of receptor sites are occupied. BPA has 1,000th of the affinity of natural oestrogen for the nuclear oestrogen receptor. This means that, at a given concentration, oestrogen will occupy 1,000x as many receptors as BPA will.
One of the interesting things about an equilibrium is that increasing concentration of the signalling molecule does not necessarily make that much difference to how many receptors are occupied. In the case of oestrogen, for example, increasing the concentration from 0.001NM by a factor of 10 to 0.01NM results in almost 10x as many receptors being occupied (see figure 1, click image to enlarge).
However, increasing concentration by another factor of 10, to 0.1 NM, only increases receptor occupancy by a factor of 5. Increasing concentration by 10 again, to 1NM barely doubles receptor occupancy. And increases beyond that make hardly any difference at all. By this point the receptors saturated.
This behaviour follows the law of mass action and is an essential precondition of hormone signalling. If small changes could not have big effects we would need to produce millions upon millions of molecules to have an effect, a process too energy-consuming and cumbersome to allow an organism to keep up with changes in its internal and external environment. Cell signalling would be too insensitive to exist and life would be dead on arrival.
Increasing the sensitivity of signalling systems still further, biological systems also amplify hormone signals through multi-step cascade reactions. One initial signal will cause several molecules to be produced in the cell. Each of these in turn cause several more molecules to be produced, these several again, so that within 7 or 8 steps 1 molecule can cause production of 1 hundred million molecules, as in the case of adrenaline (epinephrine) causing production of glucose-1-phosphate.
Physiological responses to signals also saturate. This can be due to a number of reasons, including finite energy, raw materials, or shutdown processes in the cell which prevent further signalling. In the case of breast cancer tumour cells, where exposure to oestrogen causes them to proliferate, they respond maximally at 0.1NM concentration (see figure 2, click image to enlarge).
Significance for Endocrine Disruption
Combining Figure 1 and Figure 2 shows just how big an effect a small change can have (see Figure 3, click image to enlarge). At the low 10% increase in receptor occupancy increases cell proliferation by a factor of 9; at the higher end, a 10x increase only increases response by 1.1x. The first 10% of receptor occupancy induces 90% of the maximum response in our breast cancer cells.
We have to remember that oestrogen receptor signalling is already physiologically active: EDCs are not being introduced into a system devoid of hormones, but into a system where signalling is already taking place. Since negative feedback loops control response, at any given time hormones must occupy a percentage of receptors close to the level which induces a significant physiological effect. In most circumstances, receptors are occupied to near the threshold level of effect.
This means that an EDC such as BPA does not have to be present at 1,000x higher concentration than oestrogen to potentially have a significant effect. All it needs to do is be present at a concentration which will tip the balance. As Figure 3 shows, there are occasions when a 2% increase in receptor occupancy can increase response by 15%. If BPA only needs to occupy this few receptors, it could plausibly have an effect at only 50-100x the concentration of oestrogen.
ED potency is therefore not exclusively determined by its affinity for a receptor, but also by background physiological conditions determining existing receptor occupancy. Since hormone signalling in the body varies at different times, the potency of an ED is not an absolute value, but will change depending on timing of exposure.
Even a substance with relatively low affinity for a receptor could tip the balance without having to be present in as high a concentration as calculations of relative affinity might suggest. Equally, if there is already a substantial percentage of receptors occupied, then occupying more receptors will not make much difference to the overall total effect. Adding BPA to an environment where there is already active oestrogen signalling will not have much observable effect.
Lessons for toxicological testing
It is because all the action takes place at low concentrations, at the bottom end of the dose-response curve, that EDCs are of concern – even if they have relatively low affinity for receptors in comparison to native hormones. It is also why standard toxicological testing protocols are worthless for detecting ED effects.
Toxicological testing is carried out at high doses, with a maximum dose often defined as one which causes substantial harm to a large percentage of animals in a test group. These effects are not hormonally-mediated – we know it is happening at levels far higher than saturation for receptors, so cannot induce any physiological response this way. (And besides, you can kill breast cancer cells with oestrogen, even if the cells have been bred to have no oestrogen receptors.)
Dosing is then brought down to levels 50-100 times lowers than causes serious harm. This dosing, however, is still far too high to reveal anything about hormonal effects, because even 1/100th of an acutely toxic dose is high enough that the hormone receptors are saturated. Reducing the dose can make the acute effect vanish, but will reveal nothing about what the substance does at the hormonal level. This is why toxicological testing of potential ED effects has to take place at physiologically relevant concentrations.
Hormone signalling also demands that testing be done on total hormone load, since it is total receptor occupancy which determines the effect which an EDC has. It also means the amount of EDCs we can safely be exposed to vanishes towards zero: if we are exposed to just 10 chemicals, each with the 1,000th the affinity of oestrogen for the oestrogen receptor, then we have an oestrogen load 100th as active as oestrogen itself.
As we have seen, this cannot be a safe dose in a physiologically active environment because there will be a significant number of occasions when it will make the difference between a physiologically irrelevant and significant response. If we were dealing with high levels of oestrogen, and high receptor occupancy, it would not matter. But we are not – it is the small changes which count.
Finally, all of this shows why windows of development are so significant. In adults, hormone signalling is mainly concerned with keeping the body in balance. Short-term, occasional interference is likely to be corrected. However, during foetal development hormone signals are responsible for tissue arrangement and behaviour. Short-term exceeding of tipping points can set processes going out of sync, permanently altering tissue development and behaviour.
With thanks to Tony Tweedale for input and ideas.
Welshons WV, Thayer KA, Judy BM, Taylor JA, Curran EM, vom Saal FS. Large effects from small exposures. I. Mechanisms for endocrine-disrupting chemicals with estrogenic activity. Environ Health Perspect. 2003 Jun;111(8):994-1006. Review. PubMed PMID: 12826473; PubMed Central PMCID: PMC1241550.
vom Saal FS, Welshons WV. Large effects from small exposures. II. The importance of positive controls in low-dose research on bisphenol A. Environ Res. 2006 Jan;100(1):50-76. Epub 2005 Oct 27. PubMed PMID: 16256977.
Welshons WV, Nagel SC, vom Saal FS. Large effects from small exposures. III. Endocrine mechanisms mediating effects of bisphenol A at levels of human exposure. Endocrinology. 2006 Jun;147(6 Suppl):S56-69. Epub 2006 May 11. Review. PubMed PMID: 16690810.
This video explains the differences between the mainstream proposals for cumulative risk assessment (CRA, as covered in H&E #40), the attempt to assess the risk to human health and the environment posed by simultaneous exposure to multiple chemicals. It explains the role of concepts such as common health outcomes and modes of action within the framework of CRA.
This presentations were part of a workshop organised by the US Standing Committee on Emerging Science for Environmental Health Decisions. For more information visit their website; for updates on new workshops and other information, sign up for their newsletter.
The next Standing Committee workshop is on Individual Exposomes (8-9 Dec 2011, Washington DC, more information here).
Strength in Numbers: Three Separate Studies Link in Utero Organophosphate Pesticide Exposure and Cognitive Development. A summary of three studies (Bouchard et al. 2011; Engel et al. 2011; Rauh et al. 2011) which deliver compelling new data linking in utero exposure to organophosphates and early cognitive development. Organophosphates are finding increasing use as alternatives to halogenated flame retardants.
Twenty Years Of Green Chemistry. In this article, Paul Anastas, often described as “the father of green chemistry” outlines successes in the 20 years since the term was first coined for the “design of chemical products and processes that reduce or eliminate the use and generation of hazardous substances”.
Banned chemical levels high in pregnant women. Evaluating levels of PBDEs in blood samples of 286 pregnant women, researchers found that every 10-fold increase was tied to a 4.1-ounce drop in birth weight. “The good news is, these chemicals aren’t being used anymore. But the bad news is, they’re in things we don’t replace very often,” said study author Kim Harley.
Triclosan, an Antibacterial Chemical, Raises Safety Issues. The maker of Dial Complete hand soap says that it kills more germs than any other brand. But its active ingredient triclosan is so prevalent that a survey by the US Centers for Disease Control and Prevention found the chemical present in the urine of 75 percent of Americans over the age of 5. Is it safe?
Scented laundry products emit hazardous chemicals through dryer vents. Findings, published online this week in the journal Air Quality, Atmosphere and Health, show that air vented from machines using the top-selling scented liquid laundry detergent and scented dryer sheet contains hazardous chemicals, including two that are classified as carcinogens.
Environmental Exposures and Mammary Gland Development: State of the Science, Public Health Implications, and Research Recommendations. Assessment of mammary gland development should be incorporated in chemical test guidelines and risk assessment. Inconsistent reporting methods hinder comparison across studies, and relationships between altered development and effects on lactation or carcinogenesis are still being defined. This editorial explains the context of the recommendations, while this study proposes 4 specific ways in which mammary gland animal assays can be improved.
Epigenetic perspective on the developmental effects of bisphenol A. This review highlights research indicating potential consequences of prenatal BPA exposure for brain, behaviour, and immune outcomes and discusses evidence for the role of epigenetic pathways in shaping these developmental effects.
Temporal Changes in the Levels of Perfluorinated Compounds in California Women’s Serum over the Past 50 Years. This is the first study to investigate temporal changes of PFCs over the past 50 years. For longer chain perfluorocarboxylic acids (PFCAs), there was a continuous build-up in serum of Californian women from the 1960s to 2009. PFOS and PFOA are the PFCs present at the highest concetrations.
Associations of persistent organic pollutants with abdominal obesity in the elderly. In animal experiments, persistent organic pollutants (POPs) have induced visceral obesity. This cross sectional and prospective human study found that low exposure to less-chlorinated POPs was associated with increased abdominal fat, and highly-chlorinated POPs the opposite. Another study (Rönn et al. 2011) found a complex relationship between pesticides and PCBs and obesity, while further evidence emerged that obese people exposed to POPs are at the highest risk of becoming diabetic (Airaksinen et al. 2011).
Prenatal exposure to bisphenol A and phthalates and infant neurobehaviour. In this study, prenatal exposure to DBP was associated with improved behavioral organization in 5-week-old infants while prenatal exposure to DEHP was associated with nonoptimal reflexes in male infants. There was no evidence of an association between prenatal BPA exposure and changes infant neurobehavior.