||Wozniak, AL, NN Bulayeva and CS Watson. 2005. Xenoestrogens at Picomolar to Nanomolar Concentrations Trigger Membrane Estrogen Receptor-alpha-Mediated Ca++ Fluxes and Prolactin Release in GH3/B6 Pituitary Tumor Cells. Environmental Health Perspectives 113:431-439.
Most studies of the mechanisms of action of xenoestrogens have focused on changes in gene expression resulting from interactions of the contaminants with the classic estrogen receptor located within the cell nucleus. These have consistently found that xenoestrogens are much weaker than estradiol, the native hormone. This observation has lead some to conclude that such 'weak estrogens' are unlikely to have an effect on human health.
A growing body of literature is revealing that even through this 'traditional' mechanistic pathway, some xenoestrogens alter biological functions at levels far beneath current safety standards, for example, bisphenol A.
With this paper, Wozniak et al. directly challenge the concept of 'weak estrogens.' They show that several xenoestrogens alter a signaling pathway in cells that is initiated by an estrogen receptor located on the cell membrane's surface. The altered pathway is centrally involved in the control of a wide array of developmental and physiological functions, including affecting release patterns of prolactin, a hormone that influences behavior and fertility. Each contaminant has its own signature effect. In several cases, they are as powerful as estradiol at disrupting the process.
What did they do? Wozniak et al. studied a process mediated by an estrogen receptor located on the surface of cell membranes that controls calcium influx into the cells.
They carried out a series of experiments exposing cells in culture to extremely low doses of estradiol and, one by one, similar doses of a series of xenoestrogens.
The xenoestrogens used were bisphenol A, nonylphenol, dieldrin, DDE, endosulfan , DES and coumestrol.
The first round of experiments examined the calcium response of the cells to different levels of xenoestrogen. Wozniak et al. tracked changes in calcium concentration within the cell by using a measurement of the cell's fluoresence that changes as calcium concentration changes.
The native hormone estradiol interacts with a version of the estrogen
α receptor on the cell membrane surface to cause calcium to flow into the cell. This response is very rapid-- detectable within less than one minute-- and it stops when estradiol is removed.
Increases in calcium within the cell initiate a wide array of processes beginning within the cell as signaling pathways are altered and downstream genes are affected. One example of many involves rapid secretion of prolactin, a hormone involved in the control of lactation, cell proliferation, the cell's immune system, and, at an organismal level, maternal and paternal behavior.
The instruments they used allowed them to track, second-by-second, these changes in cell fluoresence--and thus calcium concentration within the cells. To describe the impact of exposure on calcium influx, they compared the fluoresence measurement of a cell before exposure, R0, to a five minute average of its fluoresence afterward, R. They then calculated an index of change as (R-R0)/R0.
Additional experiments, described below, then used compounds known to interfere with the effects of estradiol to tease apart details of how the xenoestrogens were working.
What did the find? Estradiol and all the xenoestrogens tested increased calcium flowing into the cells, detectable within 30 seconds of exposure. The patterns of impact varied among the compounds, as can be seen in four examples below, which display the calcium influx change from the base rate for different concentrations of the estrogenic compound.
Several aspects of these dose-response curves are noteworthy:
- Significant increases in calcium inflow were seen even at the lowest levels tested, parts per trillion.
- Not all responses increased as dose increased. For example, the maximum response to bisphenol A was at 10-9 and calcium inflows for DDE dipped back to the basal rate at 10-10. A non-monotonic pattern also was seen with dieldrin (not shown).
- Some compounds had greater effects than estradiol at the same concentration. For example, the response to BPA at 10-9 was greater than to estradiol at 10-9.
These graphs show a measurement of the relative change in calcium influx compared to the base rate (no exposure) across different exposure levels. Doses are molar concentrations of the compound. Verticle lines on each bar are standard errors, reflecting the amount of variability in the measurements at that dose. Bars labeled * differed significantly from the base rate.
The calcium influx index is calculated as described above: (R-R0)/R0. In principal, it could have increased or decreased compared to the base rate.
Having established that the xenoestrogens increased calcium influx, they set about to understand how it happened and what subsequent effects it might have.
One experiment showed that the changes in calcium measured within the cells were not caused by calcium already stored inside. They did this by first using a technique known to empty the intra-cellular calcium stores. After the calcium reservoirs had been emptied, they carried out exposure experiments like those above, and found that the system still responded as it had before.
A second experiment showed that to see the response at all it was necessary to have calcium present in the fluid outside the cells. After eliminating calcium from the extra-cellular fluid, and then repeating the exposure experiments, they found that the system no longer responded. The response reappeared after they restored calcium to the outside fluid. Hence extra-cellular calcium is necessary.
In a third experiment, they demonstrated that the response is mediated by the same type of calcium channel (for calcium inflow into the cell) involved in the response to estradiol itself. They did this by exposing the cells to a compound known to disable that channel. The response disappeared.
Because one of the well established consequences of calcium influx into cells is rapid release of the hormone prolactin, they then determined whether the xenoestrogens had the same effect on prolactin secretion, using a standard concentration for each compound of 10-8 molar. Some did--DES, bisphenol A, nonylphenol and coumestrol. However the organochlorine pesticides had either a delayed effect or, in the case of DDE, no effect on prolactin secretion at that concentration (although it did at others).
They then studied the dose-response characteristics for the prolactin secretion response. Their results for four of the xenoestrogens are shown in the graph below. Each graph displays the increase in prolactin secretion, as a percent of control, for the range of doses tested.
These graphs show the average response to exposure to a range of concentrations of 4 estrogenic compounds, expressed as a percent of the response of the control. Thus a value of 100 is equivalent to the control's response. Small vertical lines on the bars are standard errors.
* indicates statistical significance (p<0.05)
||Of all the compounds tested, including estradiol, bisphenol A had the largest impact at the lowest dose tested. Exposure to bisphenol A at 10-12 caused a doubling of the amount of prolactin secretion. The smallest effects were seen for exposure to the phytoestrogen coumestrol. The only concentration that differed significantly from the control as the highest,10-8, at which it doubled prolactin.
As with calcium influx, Wozniak et al. observed several non-monotonic dose-response relationships over the range of doses used in experiments on prolactin secretion.
In the final experiment, Wozniak et al. showed that by knocking out the calcium channel, they eliminated the prolactin response. To do this they used the same chemical used above to show that knocking out the calcium channel eliminated the influx of calcium.
What does it mean? This study demonstrates that xenoestrogens alter a major biochemical signaling pathway with effects comparable to, and sometimes greater than, similar concentrations of estradiol. Via this pathway, they are not "weak estrogens," they are comparable in potency. Moreover, the changes are observed for contamination levels well within those that have been measured in human tissue and, for the environmental contaminants (bisphenol A, nonylphenol and the organochlorine pesticides) far below levels deemed safe by current regulatory standards.
This study considers biochemical changes in single cells manipulated in vitro. Many levels of organizational complexity lie between this simple but elegant series of experiments and understanding how these changes may affect human health.
Their results raise many issues, nonetheless. In their experiments, For example, Wozniak et al. show that prolactin levels increase with xenoestrogen exposure. As they comment in their discussion, prolactin increases in people have been associated with delays in puberty, interference with ovulation, decreases in libido and fertility, and increases in cell proliferation. Prolactin is also an important signaling molecule for the immune system and the lining of the pregnant uterus.
And changes in prolactin activity are but one of the biochemical pathways altered by calcium influx, which is an important initiating event in signaling systems that alter gene expression in humans of a wide array of genes. For example, in a related set of experiments, Quesada et al. showed that bisphenol A, by increasing calcium influx, causes phosphorylation of CREB. This transcription factor is involved in the molecular mechanisms of long-term memory formation, the control of natural cell death during brain development, as well as altering gene expression important to weight regulation.
Studies of endocrine disrupting compounds have focused over the past 30 years for the most part on their interactions with nuclear hormone receptors and the adverse effects that follow. As documented in Our Stolen Future, on these pages, and in a wide array of studies published, this focus has demonstrated many impacts on laboratory animals and wildlife, and has raised serious questions about the consequences for human health. Links to human health are becoming stronger as epidemiologists increasingly adjust their measurement tools to take into account the complex challenges that endocrine disruption creates for their science, including low-dose impacts, mixtures, non-monotonic dose-response curves and extremely long latencies between exposure and detectable effect.
With this paper, Wozniak et al. confirm that it will also be necessary to integrate a whole new set of signaling pathways into the assessment of health risks from endocrine disrupting chemicals. Their work indicates that via this cell membrane receptor pathway, EDCs can work on a much faster time-scale than the traditional pathways through hormone receptors in the nucleus of the cell and can be equally as potent as endogenous signaling molecules. Assessing the health risks from EDCs was not easy before. This raises the complexity and seriousness of the challenge.