Our Stolen Futurea book by Theo Colborn, Dianne Dumanoski, and John Peterson Myers


  Alonso-Magdalena, P, S Morimoto, C Ripoll, E Fuentes and A Nadal. 2006. The Estrogenic Effect of Bisphenol-A Disrupts the Pancreatic ß-Cell Function in vivo and Induces Insulin Resistance. Environmental Health Perspectives 114:106-112.

What is bisphenol A and how is it used?

What is insulin resistance and how does it relate to diabetes?

Insulin is a hormone that stimulates glucose transport into cells. In mice and in people, it is made by specialized cells in the pancreas called ß cells. Insulin production is triggered by rising glucose levels. Secreted into the bloodstream, it then binds with insulin receptors on the surface of muscle, liver and fat cells, and stimulates production and movement of glucose transporters to the cell membrane allowing diffusion of glucose into the cell.

In people (and mice) with insulin resistance, insulin is less effective at stimulating glucose uptake.

In early stage of insulin resistance, ß cells respond by producing more insulin. In later stages, they can't produce enough, and blood sugar levels rise, intially following meals but ultimately even during fasting, at which point it is diagnosed as Type II diabetes. Approximately 25% of the people who develop insulin resistance go on to develop Type II diabetes.

Insulin resistance is linked not just to Type II diabetes, but also contributes to central obesity, cholesterol abnormalities, and high blood pressure. This constellation of diseases is known collectively as the metabolic syndrome. Metabolic syndrome and its components like diabetes and obesity over the past 2 decades have become global epidemics.


Alonso-Magdalena et al. demonstrate that low-level, chronic exposure to bisphenol A (BPA) induces insulin resistance in adult mice. Their work provides the first experimental link between endocrine disruption and diabetes.

The doses they used in these experiments were within the range of current human exposure, 5000 times below the dose identifed by the US EPA as the lowest level causing effects.

Bisphenol A is used to make polycarbonate plastic, a material used in many consumer products like the bottles to the left. The chemical bonds of this plastic are weak and hence it readily breaks down and leaches into food and water.

Insulin resistance in people is a precursor of type II diabetes and a core part of one of the modern world's major epidemics, 'metabolic syndrome.' Insulin resistance in people is also tied to congestive heart failure.

The authors conclude that exposure to BPA increases the risk of type II diabetes, hypertension and dyslipidemia.

These results raise large questions about the possible contribution of bisphenol A to a major public health concern. The incidence of diabetes has increased in the last decades and at the present, it is reaching epidemic levels, with 125 million diabetics in the world. This heightens the importance of developing new public health standards for exposure to BPA. Current standards are based upon scientific studies from the 1980s.

What did they do? The research team injected mice with doses of bisphenol A and compared the effects to those caused by injecting estradiol. They looked at short-term reactions to single injections and chronic impacts of twice-daily injections over 4 days. Doses used were between 10 and 100 μg/kg, depending upon the experiment. In experiments on insulin intolerance, they also delivered BPA orally (100 μg/kg for 4 days).

To assess the impacts of these treatments, they took measurements of a series of cell parameters important for insulin and glucose regulation, using standard measurement techniques, including blood glucose concentration, plasma insulin levels and insulin content and release from islet cells.

To establish glucose tolerance, they fasted treated mice overnight and then treated them with glucose, comparing serum levels prior to glucose treatment with those measured subsequently.

To establish levels of insulin tolerance, they injected fed mice with insulin and tracked subsequent changes in serum glucose.

What did they find? Alonso-Magdalena et al. found both rapid and long-term effects. In summary, they showed:

  • Rapid (within 30 minutes) decreases, compared to controls, of blood sugar levels and sharp increases in blood insulin levels;
  • Long-term (over 4 days) increases in pancreatic ß cells insulin production and secretion, increases in serum insulin levels; and development of insulin resistance, impaired glucose tolerance and insulin intolerance.

Rapid effects: The experiments revealed rapid changes in glycemia and induction of higher insulinemia in response to low level exposures to both estradiol and bisphenol A.

Blood sugar levels in control animals rose within 30 minutes of injecting fasting mice with the control treatment, corn oil. This is expected and can be seen in the graph below. Within 30 minutes of injection, blood sugar levels increased in control mice by almost 30 milligrams per deciliter (mg/dl).



When bisphenol A or estradiol were added to the corn oil, the rise in blood sugar level was lower. This result was statistically significant at 10 ppb of both BPA and E2. At 100 ppb of BPA, blood sugar level actually decreased.

* p < 0.05 compared to control

They also measured plasma insulin 30 minutes after injecting the mice with corn oil with or without estradiol and bisphenol A (graph to right).

Insulin levels in serum were sharply elevated after a single application of only 10 μg/kg of estradiol or bisphenol A.

* p = 0.024
p = 0.004


These results demonstrate rapid interference with glucose regulation. Prior work by this laboratory indicates this rapid response is mediated by a 'non-classical membrane receptor' for estrogen, which bisphenol A can also activate. Results from other laboratories are consistent with rapid responses via an estrogen receptor located outside the cell nucleus.

Alonso-Magdalena et al. support this interpretation by showing that a compound, ICI182,780, known to suppress activation of a classical nuclear estrogen receptor, ER, does not shut down the rapid response to estradiol and bisphenol A found in this experiment.

Long-term effects (in 4 days):


In mice dosed by injection over 4 consecutive days with estradiol and bisphenol A at 100 μg/kg, pancreatic ß cells increased insulin production, as reflected by insulin content in the cell clusters called islets where ß cells are found in the pancreas (graph to left).

* = < 0.05

An smaller but still significant increase in insulin content was also seen at 10 μg/kg.

In contrast to the rapid effects, described above, long-term effects were blocked by the anti-estrogenic compound ICI182,780. This indicates that the long-term effects are mediated by classical nuclear estrogen receptors.

After four days of treatment, animals then challenged with glucose secrete more insulin if they have been exposed to estradiol or bisphenol A.

This can be seen in the graph to the right, showing the results of glucose administration at three different levels to animals in the control group vs. those treated with estradiol or bisphenol A.

Administered 7 mM glucose, the BPA-treated animals secreted more insulin than did control, whereas with 16 mM glucose both E2 and BPA treated animals secreted more.



  Alonso-Magdalena et al. then confirmed that the higher rates of insulin secretion (above) caused higher levels of insulin in the animal's serum, a condition called hyperinsulinemia (graph to right): Four days after treatment, estradiol-treated mice had blood insulin levels 1.7-fold higher than controls; insulin levels in BPA-treated mice were 1.5-fold higher.

In these last mice, the blood glucose levels of the estradiol-treated mice were slightly but not significantly lower (p=0.65) than controls, even though the serum insulin levels were 1.7x higher. For BPA-treated mice, despite a 1.5x greater amount of serum insulin, blood glucose levels were virtually the same (p=0.93). This pattern, according to Alonso-Magdalena, is "a clear symptom of insulin resistance." With higher insulin levels in the treated mice, blood glucose levels should be signficantly lower, if their response to insulin were normal.

In the final series of tests, the research team tested for glucose and insulin tolerance:

  • To test for glucose tolerance (graph below to left), they administered glucose to animals that had been fasting for 12 hours and measured glucose serum levels in the subsequent two hours. The treated animals had received injections of BPA or estradiol (not shown) at the rate of µg/kg/day over 4 days prior to glucose administration.
  • To test for insulin tolerance (graph below to right), they administered insulin via injection to fed mice and monitored glucose serum levels over the next 45 minutes. In the example shown, BPA was delivered orally in corn oil at a rate of 100 µg/kg/day over 4 days prior to insulin administration.

Serum glucose rose higher in the BPA-treated animals that had been fasting, indicating slower cellular uptake of glucose, a pattern consistent with insulin resistance (graph above left).

Following insulin injection in the controls, serum glucose levels dropped. No decline was seen in animals treated orally with BPA (graph above right). The same effect is observed when E2 and BPA were injected (not shown).

What does it mean? Based in this series of experiments, Alonso-Magdalena et al. conclude that "E2 and in a remarkable manner, BPA treated animals develop insulin resistance."

Previous work has shown that estradiol is involved in insulin regulation, and that abnormal levels can lead to insulin resistance in people. Hence it is not surprising that an estrogen mimic like bisphenol A can affect insulin regulation.

What is striking, however, is the low level at which bisphenol A caused changes in glucose and insulin metabolism in these experiments. Alonso-Magdalena et al. show rapid effects at a dose 5 times lower than the level considered safe by the US EPA, below the level reported in infant cord blood at birth, and approximately 5000 times beneath the dose currently identified by the US EPA as the lowest level at which an adverse effect is seen (the LOAEL). These results are only (as of October 2005) the most recent in a flood of studies over the past 8 years that have shown adverse effects well beneath EPA's LOAEL.

These results are also consistent with laboratory studies showing that BPA, while a weak estrogen in actions mediated by receptors located in the cell nucleus, is just as powerful as estradiol when acting through estrogen receptors on the cell membrane.

According to Dr. Fred vom Saal, an expert on developmental impacts of bisphenol A at low levels, these results demonstrate important changes in insulin and glucose metabolism at BPA levels within the range experienced by 95% of the American public.

One exceptional aspect of this paper, compared to most of the more than 100 animal studies now available on low dose impacts of BPA, is that Alonso-Magdalena et al. performed their experiments on adult mice, not fetal mice. Almost all other low-dose impacts have been observed following fetal exposure. Usually it takes higher doses to affect adults, because protective enzyme systems have matured and because developmental processes are complete and hence no longer vulnerable to disruption. This suggests it will be very important to examine the consequences of fetal exposure to BPA on subsequent insulin regulation.

These results raise pivotal questions about bisphenol A's possible link to 'metabolic syndrome,' a collection of adverse human health conditions--including insulin resistance, Type II diabetes, hypertension and obesity--that public health authorities now acknowledge is a world-wide epidemic, affecting over 50 million people in the United States alone. This syndrome has reached epidemic proportions over the last 30 years, the same time span over which human exposure to bisphenol A became ubiquitous.

Almost no epidemiological research has been conducted on the human health effects of BPA. Hence firm conclusions about the consequences of this exposure for people are impossible. A great deal is known about the physiological and metabolic processes involved in diabetes and insulin resistance, and mouse experiments have proven highly informative for questions about human health. This strongly suggests that research on the effects of BPA on human insulin metabolism should be a high priority.

Animal experiments with bisphenol A have now shown low-level exposures cause a wide array of adverse effects, including:

Virtually all of this research has been published within the past decade. With this paper, Alonso-Magdalena et al. add insulin resistance, and by implication, the plausibility of links to Type II diabetes to the list of BPA adverse impacts. It is astonishing that the EPA and the FDA continue to use a safety standard based on research findings from the 1980s.





OSF Home
 About this website
Book Basics
  Synopsis & excerpts
  The bottom line
  Key points
  The big challenge
  Chemicals implicated
  The controversy
New Science
  Broad trends
  Basic mechanisms
  Brain & behavior
  Disease resistance
  Human impacts
  Low dose effects
  Mixtures and synergy
  Ubiquity of exposure
  Natural vs. synthetic
  New exposures
  Wildlife impacts
Recent Important    Results
Myths vs. Reality
Useful Links
Important Events
Important Books
Other Sources
Other Languages
About the Authors
Talk to us: email