THE ENDOCRINE SYSTEM - By Dr. A. Michael Warhurst

The endocrine system is a complex interplay between a number of hormones, including the sex hormones the oestrogens and androgens, and other hormone systems such as the thyroid system.

Oestrogens such as oestradiol (structure in fig. 1) are the hormones that influence the development and maintenance of female sex characteristics, and the maturation and function of the sex organs. Chemicals which can imitate an oestrogen are known as oestrogenic chemicals. Androgens such as testosterone serve a similar purpose in males.

THE HORMONAL OR ENDOCRINE SYSTEM

This system is the main subject of this web site, particularly the reproductive hormones. Hormones generally carry fairly long-lasting messages, in contrast to the rapid signalling of the nervous system. In 1997 more complexity was added to the attempts to understand the endocrine system, and potential effects of chemicals on it, when it was discovered that the 'female' hormone oestradiol is essential for male fertility (Sharpe, 1997). If this wasn't enough, it is now clear that there is not just one oestradiol receptor; there are several, and they all behave slightly differently, with varying levels in different tissues (Petersen et al., 1998).

THE IMMUNE SYSTEM

The immune system is responsible for resisting infectious agents, environmental substances, and foreign or damaged cells (Dean and Murray, 1991). The capacity for immune response is determined early in development, so damage early in life will persist (Colborn, 1996). The immune and endocrine systems are linked by a variety of signalling systems, and much research on wild animals has demonstrated that endocrine disrupting chemicals can reduce the effectiveness of the immune system (Kavlock et al., 1996). A lot of this research has focused on marine animals and birds; for example, high levels of PCBs and DDT in the blood of dolphins was associated with decreased immune function and increased incidence of infection (Kavlock et al., 1996).

THE NERVOUS SYSTEM

The function of the nervous system is the rapid communication of signals between different parts of the body, and the processing of these signals. The most complex part of the nervous system is the brain, where most of the processing occurs. Although most of the nervous system is protected from potential toxicants by the blood-brain barrier, some chemicals can penetrate this barrier (Anthony and Graham, 1991). In addition, this barrier is not fully developed until after birth, which is why the developing fetus is so sensitive to chemicals such as alcohol.

A US EPA workshop concluded that a variety of endocrine disrupting chemicals, including some PCBs, dioxins, DDT and chlorinated pesticides, are capably of producing neurotoxicity (Kavlock et al., 1996). Effects reported in the literature included changes in behaviour, learning, and memory attention. 'Our Stolen Future' (Colborn et al., 1996) describes several of these studies. Most of these toxic effects are believed to occur due to the endocrine disrupting chemicals influencing the development of the nervous system.

THE REST OF THE BODY

Other parts of the body are also affected by hormones. For example, blood vessels are influenced by oestrogen levels, with oestrogen exposure leading to relaxation of the walls of the vessels. Ruehlmann et al (1998) have found that 4-octyl phenol, p-nonylphenol and o,p'-DDT have the same effect as oestrogen.

DEVELOPMENT

The timing of administration of a hormone disrupting chemical is particularly important for fetuses and young children, as the body's systems are most sensitive when they are under construction (Manson and Wise, 1991). During this period, different events are occurring on different days, so providing small windows of time when the developing organism is extremely sensitive to hormone disrupting chemicals, for example during the formation of the testes (see the health page). Diethylstilbestrol (DES), a synthetic oestrogen, was taken by more than 5 million pregnant women between the late 1840s and the early 1970s, and led to many reproductive abnormalities in both male and female offspring (Toppari et al., 1996). Experiments on rats have since demonstrated how DES exposure at particular stages of fetal development leads to reproductive system damage, similar to that exhibited by the human victims (Toppari et al., 1996).

It is known that the developing fetus is the most sensitive stage in human development of organs (such as the testes). The effects of many of these chemicals are additive; there is more debate about the extent of more than additive effects (see issues page), so exposure to a range of chemicals at a low levels has the same effect as exposure to one chemical at a higher level.

MECHANISMS OF ENDOCRINE DISRUPTION

A endocrine disrupting chemical can affect the endocrine system of an organism in a wide variety of ways. Here are some of them, particularly focusing on sex hormone disrupters:

1. BINDING AND ACTIVATING THE OESTROGEN RECEPTOR (therefore acting as oestrogen): By imitating the female hormone 17beta-oestradiol. One complexity of this mode of action is the fact that there are a variety of oestrogen receptors, present in a wide range of tissues.
   It has been found that if several chemicals that can bind and activate the oestrogen receptor are added together, their effects will usually be additive, so the effects of small quantities of a range of oestrogenic chemicals can add together into a much larger effect (Soto et al., 1995). In addition, chemicals such as butylbenzyl phthalate and di-n-butyl phthalate have been shown to add their effects to any natural oestrogen present (Jobling et al., 1995).
   It is also possible that some chemicals will demonstrate more than additive effects, they will demonstrate synergism. It is unlikely that synergism will occur with two chemicals acting through one receptor; the one paper that claimed to show this effect was later withdrawn (McLachlan, 1997). However, synergism is known to occur in other areas of toxicology when the two chemicals are working through different mechanisms, for example tobacco smoking accentuates the toxicity of many chemicals.
2. BINDING BUT NOT ACTIVATING THE OESTROGEN RECEPTOR (therefore acting as an anti-oestrogen)
3. BINDING OTHER RECEPTORS: There are many other receptors involved in the hormonal system, for example the androgen receptors for male hormones. This binding can either activate the receptor, or inactivate it, as happens with anti-androgens like the DDT metabolite p, p'-DDE (see the DDT page).
4. MODIFYING THE METABOLISM OF NATURAL HORMONES: Some chemicals, such as lindane and atrazine, can effect the metabolic pathway of oestradiol, producing more oestrogenic metabolites (see the DDT page). Other chemicals activate enzymes which speed up the metabolism of hormones, so disrupting their natural state. The testes contain specific enzymes to metabolise oestrogens (Toppari et al., 1996). These enzymes break down oestrogen rapidly to a form where they can no longer bind their receptor. However, if this enzyme is affected by a xenoestrogen, this metabolism will be reduced, increasing the exposure of the testes to oestrogen. This could be particularly relevant during foetal development, when there are high levels of oestrogens (Toppari et al., 1996).
5. MODIFYING THE NUMBER OF HORMONE RECEPTORS IN A CELL: Complex mechanisms control the numbers of hormone receptors present in cells. A chemical could reduce or increase the number of receptors, and so affect the extant of response to natural or artificial hormones.
6. MODIFYING THE PRODUCTION OF NATURAL HORMONES: Chemicals can affecting natural hormone production by interfering in other signalling systems, such as other hormone systems like the thyroid system, or the immune and nervous systems.

CRUCIAL WINDOWS OF SENSITIVITY IN MALE DEVELOPMENT

The development of the testis occurs almost entirely during early development in the womb. It is in this period that the Sertoli cells differentiate, and any exposure to oestrogen at this time reduces the number of Sertoli cells produced (Jensen et al., 1995). The Sertoli cells are responsible for producing sperm in later life, and it has been shown that the number of Sertoli cells is directly related to the sperm count, so fewer cells will lead to a lower sperm count (Jensen et al., 1995). It is also believed that abnormal germ cells, formed in early development, are responsible for most testicular cancers in later life (Jensen et al., 1995).

The oestrogen diethylstilbestrol (DES) was given to > 5 million pregnant mothers in the period between the late 1940s and the early 1970s to prevent miscarriage. Its use was stopped after a high incidence of a rare cervical cancer in pubertal girls exposed to DES in the uterus. It was later found that male offspring also had a higher level of reproductive abnormalities, including low sperm counts (Jensen et al., 1995). The case of DES is a clear indication that exposure of the foetus to external oestrogens can result in reproductive problems later in life (Toppari et al, 1996).

For the adult males, direct toxic effects on sperm production by chemicals such as phthalates could also be an issue. However as far as oestrogenic effects go, it is clear that the final 3 months of pregnancy and the first few months of life will be where any exposure of a male to oestrogens is likely to have the greatest effect. The research showing that metabolites of DDT can block the male hormonal system is worrying.

OTHER FACTORS AFFECTING THE ACTIVITY OF ENDOCRINE DISRUPTERS

There are many variables which affect whether a hormone disrupting chemical has a biological effect, including uptake, distribution, nature of action and time of action. These points are discussed in more detail here. For more detail and references, see Toppari et al. (1996) and Kavlock et al. (1996).

UPTAKE AND DISTRIBUTION

First the chemical must enter the body, through ingesting food or drink, being absorbed from cosmetics or inhaled. The chemical will then be distributed through the body, usually by the blood. Several systems exist in the body to detoxify chemicals, notably the liver enzyme systems. These systems remove chemicals by a combination of breaking them down and attaching them to other chemicals, which promotes their excretion, usually through the kidneys and into the urine.

Some chemicals are not removed effectively by these processes, so remain in the body. Those chemicals that are lipophilic (fat soluble) can accumulate and be stored in the fat, notably PCBs and DDT. The fat stores of the body can be mobilised during stressful periods, malnutrition, or in pregnancy, releasing the stored chemicals into the blood stream. WWF have recently published a disturbing review of all the chemicals that have been found in body fat and breast milk - see the issues page for more details.

BIOAVAILABILITY

Natural hormones such as oestrogen have their concentration in the blood modified by sex hormone binding globulin and albumin, which bind the majority of the hormone in the blood, so reducing its availability to bind to receptors and initiate responses (Arnold et al., 1996). Sex hormone binding globulin binds very strongly and specifically to oestradiol, whilst albumin binding is weaker and less specific. Arnold et al. (1996) have tested how these two compounds affect the availability of 17beta-estradiol, diethylstilbestrol, octylphenol and o, p'-DDT, by seeing how their presence affects binding to the human oestrogen receptor (expressed by a yeast). They found that the xenoestrogens bound far less to the albumin and the sex hormone binding globulin, leaving more free chemical available to bind the receptor and initiate an oestrogenic response.

Bioavailability of chemicals to a foetus is affected by the placenta, which is able to prevent the crossing of some chemicals from the mother's blood into the developing child or animal (Manson and Wise, 1991). This only provides a partial barrier and many chemicals can transfer into the foetus, particularly those which are lipophilic (fat soluble).

TIMING

The timing of exposure to an endocrine disrupter can be crucial, as some stages of development are far more sensitive. This is covered in more detail on the complexity page.

LOW DOSE EFFECTS

One area of endocrine disruption which is a subject of huge debate is the issue of low dose effects. What a 'low dose' is is not always clear, but it's generally treated as a dose much lower than would normally be expected to have an effect. This page provides a very brief introduction to the debate. For more information, look at the papers themselves and look in Journals such as Environmental Health Perspectives and Environmental Science and Technology (e.g. Renner, 1998).

SCIENTIFIC BASIS

There are two particularly important aspects to low dose effects:

1. THE INHERENT SENSITIVITY OF THE ENDOCRINE SYSTEM AND DEVELOPMENT: As described on the complexity page, the interacting endocrine and developmental systems are extremely complex and sensitive. Science does not yet properly understand the operation and interaction of these systems, so it is not possible to properly evaluate the impact of chemicals on them. It is quite possible that very low doses of a chemicals could have very significant effects.
2. ADDITIVITY AND SYNERGISMS: The addition of a small quantity of a chemical to the body may have effects because it acts in addition to another chemical which is already there. For example, an oestrogen such as nonylphenol could act in addition to a natural oestrogen already present.
   In synergism the combination of two chemicals results in a more than additive response - for example one chemical might lead to an increase in the number of receptors in a cell, whilst the second chemical binds and activates them.
   Additivity and synergism can lead to no threshold effects, were even the tiniest amount of chemical added could have an effect, because its activity adds to a chemical which is already there.
   A parallel is: You are trying to look over a wall which is higher than you. You have a small brick to stand on - it's still not high enough by a long way. Then you try a block of concrete - you can almost see over, but you're not quite there. You put the brick on top of the concrete block - now you can just see over the wall.
   If the threshhold for activity is equivalent to seeing over the wall, you can see how if the block of concrete - the natural oestrogen - is just too small, you cannot see over, there is no activity. If the brick, an environmental oestrogen, is very small but just big enough, it can add to the block and allow you to see over the wall or activate the system. The turtle experiments below are an example of such a system.

SOME OF THE RESEARCH EVIDENCE

• TURTLES: In the red-eared slider turtle gonadal sex is determined by the temperatue eggs are incubated at, with higher temperatures leading to the production of endogenous estrogen and the development of females. Sheehan et al. (1999) incubated eggs at a temperature which normally generates a minority of females, then added a single dose of 17beta-oestradiol to the shells. The lowest oestradiol dose tested, 400 pg/egg (40 ng/kg) sex reversed 14.4% of the animals. A dose response curve suggested no threshold dose for oestradiol addition. This is a perfect illustration of additivity leading to no threshold.
• BISPHENOL A IN MICE: In contrast to the clear and undisputed findings of the turtle research, the situation with low dose effects of bisphenol a is a subject of huge and continuing scientific debate. Research by a group led by Fred vom Saal has found that bisphenol a and methoxychlor exposure of pregnant mice resulted in increases in prostate weight in their offspring (vom Saal et al., 1998; Welshons et al., 1999). However, industry-funded studies have failed to repeat these results (Ashby, 1999; Cagen et al, 1999). In the latest twist, vom Saal's team has shown low dose effects on female offspring, a delay in oestrus (see bisphenol a page).
• VINCLOZOLIN IN RATS: The fungicide vinclozolin is a well know antiandrogen, listed on the pesticides page. Research by L. Earl Gray's team (Gray et al., 1999) has established that alterations in sexual differentiation of developing male rat occur with low doses of vinclozolin given to their dams during preganancy and just after birth. Alterations including retention of nipples occurred at the lowest dose tested, 3.125 mg/kg/day.
  The authors conclude: "In summary, studies of vinclozolin that did not include endpoints sensitive to antiandrogenic effects and were not designed to detect transgenerational efffects have reported LOEL [lowest observed effect levels] for vinclozolin of 100 mg/kg/day or higher. In contrast, in the current study we found that vinclozolin treatment induced reproductive malformations below 100 mg/kg/day, and ano-genital distance was shortened, areolas were induced, and other endpoints were affected by treatment with vincolzolin at 3.125 mg/kg/day, our lowest dosage level"

SOME PROBLEMS WITH LOW DOSE RESEARCH

• REPRODUCIBILITY: As low dose research is often looking for subtle, but potentially significant, changes, there have been major problems with reproducibility. The dispute over bisphenol a is one example, another is the research that was done by Richard Sharpe's team at Edinburgh University, which initially showed that exposure of rats in the womb to octylphenol (OP) or butylbenzyl phthalate (BBP) led to reductions in testicular weight (Sharpe et al., 1995). The BBP experiments were repeated by others, and the OP experiments by Sharpe et al. - none of the repeated experiments showed effects. The authors explained the situation in a letter to Environmental Health Perspectives (Sharpe and Turner, 1998). The reason for the lack of reproducibility of the original results is unknown.
  
  One suggested problem obtaining reproducible results is influences of the conditions in which lab animals are kept. For example, John Ashby of AstraZeneca's central toxicology lab has said that the number of animals in each cage and whether sexes are mixed in the cages may make a major difference (C&I, 1999).
• OESTROGEN RESISTANT ANIMALS: Spearow et al. (1999) examined how responsive different strains of juvenile mice were to 17beta-estradiol (E2), the main female hormone. They found a more than 16-fold variability in response between different strains, with low doses of E2 eliminating spermatid maturation in some strains, whilst 16 times as much E2 had no effect on the widely used CD-1 strain. This strain is widely used in research, and has been bred for large litter sizes. This research suggests that CD-1 mice may give a very misleading view of hormone disrupting effects

BISPHENOL A

• USES: Bisphenol A is used in the production of epoxy resins and polycarbonate plastics. These plastics are used in many food and drink packaging applications, whilst the resins are commonly used as lacquers to coat metal products such as food cans, bottle tops, and water supply pipes (ENDS, 1995).

Some polymers used in dental treatment contain Bisphenol-A. (fig. 2)

• HORMONE DISRUPTING EFFECTS: Bisphenol A was first shown to be oestrogenic in 1938, using ovariectomized rats (Dodds and Lawson, 1938)

More recently, it was found to be oestrogenic in the MCF-7 human breast cancer cell culture assay in 1993 (Krishnan et al., 1993). The hormonal effects could be measured at concentrations as low as 2-5 ppb (2-5 µg/l).

Bisphenol a can also act as an antiandrogen, blocking the action of dihydrotestosterone in a yeast screen containing a human androgen receptor (Sohoni and Sumpter, 1998). In this screen bisphenol a was approximately as potent as flutamide, a well known anti-androgenic chemical.

Liquor containing bisphenol-A obtained from tinned vegetables has been found to be oestrogenic to human breast cancer cells (see below).

Bisphenol a produces identical effects to those produced by oestradiol on rat uterus and vagina; the vagina was particularly sensitive to the chemical (Steinmetz, 1998).

Exposure of developing male mice in the womb has been shown to enlarge their prostate glands (Nagel et al, 1997). This research has however be disputed by two controversial chemical-industry backed studies, see the low dose page for more details.

It has been claimed that bisphenol acts in the same way as female hormones in the area of the developing rat brain which regulates fertility and sexual behaviour (Anon, 1998).

Administration of bisphenol a to male rats just after birth led to slight changes in the structure of the efferent ducts, though these changes seemed to only be transient (Fisher et al., 1999).

New research shows that female mice exposed in the womb to low doses of bisphenol a (2.4 micro-g per kg per day to the mother) had a significantly reduced delay between vaginal opening and first vaginal oestrus (Howdeshell et al., 1999). This research also used information on litter position in the womb, which showed that those females positioned between female litter mates were most affected by bisphenol a, those between males were least affected. Research is suggesting that girls are reaching puberty earlier, see human health page for details

HUMAN EXPOSURE

• CANS: The liquid in some cans of tinned vegetables have been found to contain both bisphenol A, and the related chemical dimethyl bisphenol-A. The highest levels of bisphenol A were found in cans of peas, with an average of 23 µg per can. Other liquors containing bisphenol A were from cans of artichokes, beans, mixed vegetables, corn and mushrooms. There was no detectable bisphenol A in cans of palm hearts, asparagus, peppers and tomatoes (Brotons et al., 1995).
  All liquors which contained bisphenol A were oestrogenic to a human breast cancer cell assay. Those liquors without detectable bisphenol A were not oestrogenic. The liquor from the most contaminated vegetables, the peas, produced 58% of the oestrogenic response generated by oestradiol (Brotons et al., 1995).
  
  The vegetables themselves were not tested for oestrogenicity or Bisphenol A content , though since they probably contain more fat than their liquor, they are likely to contain at least as much Bisphenol A as the liquor.
  
  This research also included an examination of cans of other, more fatty, products, including condensed milk, pork and beans and concentrated milk-based infant formula. Unfortunately, the products themselves were not analysed. Instead, the cans were emptied, cleaned, then filled with distilled water and autoclaved at 125 °C for 30 minutes, then the water was analysed. Some of these water samples, including those from condensed milk cans, were found to contain bisphenol A and were oestrogenic. All canned foods are autoclaved after canning; the fact that bisphenol A is leached into water during autoclaving in these experiments suggests that any product packed in similar cans will contain bisphenol A. It is also likely that substantially more bisphenol A will leach into fatty products.
  
  NB: Which canned products do or do not contain bisphenol A cannot be determined from this study, since it will depend on the particular brand of product tested. The cans were purchased in Spain and the USA, but came from a variety of countries.
  
  US Food and Drug administration research has found that bisphenol a leaches from infant formula cans into infant formulae (Biles et al, 1997). The levels of bisphenol found in the formula varied between 0.1 ppb and 13.2 ppb, from one to six times lower than the Brotons study, above.
  
  EC rules limit migration of bisphenol A into food to 3 mg/kg, compared with a maximum liquor concentration in this research of 80 µg/kg (ENDS, 1995). The EC limit does not allow for oestrogenic toxicity.
  
  In 2001 the UK Food Standards Agency published a study on the leaching of bisphenol a from food cans.
• POLYCARBONATE BOTTLES: Many transparent 'plastic' bottles are made from polycarbonate, usually a polymer of bisphenol a. Several studies have been done on leaching of bisphenol a from these bottles. The US campaign group the National Environmental Trust has done their own research on this issues and also have background information on their site.
• DENTAL EXPOSURE: Some (but not all) dental resins contain bisphenol A. Olea et al. (1996) found that a sealant containing bisphenol A diglycidylether methacrylate (bis-GMA) was oestrogenic to MCF7 breast cancer cells. Samples of the saliva from 11 patients taken 1 hour after dental treatment contained bisphenol a and bis-GMA. There is some dispute about the deatils of this research (Ashby, 1997; Imai, 1999;Olea, 1999).

References

Jensen, T.K., Toppari, J., Keiding, N., Skakkebaek, N.E. 1995. Do environmental estrogens contribute to the decline in male reproductive health?. Clinical Chemistry 41: 1896-1901.
Toppari, J., Larsen, J. C., Christiansen, P., Giwercman, A., Grandjean, P., Guillette, L. J., Jégou, B., Jensen, T. K., Jouannet, P., Keiding, N., Leffers, H., McLachlan, J. A., Meyer, O., Müller, J., Rajpert-De Meyts, E., Scheike, T., Sharpe, R., Sumpter, J. and Skakkebaek, N. E. 1996. Male reproductive health and environmental xenoestrogens. Environ. Health Persp. 104 Suppl. 4: 741-803.
Anthony, D. C. and Graham, D. G. 1991. Toxic responses of the nervous system. In: Casarett and Doull's Toxicology: The Basic Science of Poisons. pp. 407-429. Amdur, M. O., Doull, J. and Klaassen, C. D. Eds., Pergamon Press, New York.
Colborn, T. 1996. Statement from the work session on the chemically-induced alterations in the developing immune system: The wildlife/human connection. Environ. Health Persp. 104 Suppl. 4: 807-808.
Colborn, T., Dumanoski, D. and Myers, J. P. 1996. Our Stolen Future. Penguin, New York.
Dean, J. H. and Murray, M. J. 1991. Toxic responses of the immune system. In: Casarett and Doull's Toxicology: The Basic Science of Poisons. pp. 282-333. Amdur, M. O., Doull, J. and Klaassen, C. D. Eds., Pergamon Press, New York
Kavlock, R. J., Daston, G. P., DeRosa, C., Fenner-Crisp, P., Gray, L. E., Kaattari, S., Lucier, G., Luster, M., Mac, M. J., Maczka, C., Miller, R., Moore, J., Rolland, R., Scott, G., Sheehan, D. M., Sinks, T. and Tilson, H. A. 1996. Research needs for the risk assessment of health and environmental effects of endocrine disruptors: A report of the U.S. EPA-sponsored workshop. Environ. Health Persp. 104 Suppl. 4: 714-740.
Petersen, D. N., Tkalcevic, G. T., Koza-Taylor, P. H., Turi, T. G. and Brown, T. A. 1998. Identification of estrogen receptor b2, a functional variant of estrogen receptor b expressed in normal rat tissues. Endocrinology 139: 1082-1092.
Ruehlmann, D. O., Steinert, J. R., Valverde, M. A., Jacob, R. and Mann, G. E. 1998. Environmental estrogenic pollutants induce acute vascular relaxation by inhibiting L-type CA2+ channels in smooth muscle cells. FASEB Journal 12: 613-619.
Sharpe, R. M. 1997. Do males rely on female hormones? Nature 390: 447-448.
Arnold, S. F., Robinson, M. K., Notides, A. C., Guillette Jr, L. J. and McLachlan, J. A. 1996. A yeast estrogen screen for examining the relative exposure of cells to natural and xenoestrogens. Environ. Health Persp. 104: 544-548.
Jobling, S., Reynolds, T., White, R., Parker, M. G. and Sumpter, J. P. 1995. A variety of environmentally persistent chemicals, including some phthalate plasticizers, are weakly estrogenic. Environ. Health Persp. 103: 582-587.
Kavlock, R. J., Daston, G. P., DeRosa, C., Fenner-Crisp, P., Gray, L. E., Kaattari, S., Lucier, G., Luster, M., Mac, M. J., Maczka, C., Miller, R., Moore, J., Rolland, R., Scott, G., Sheehan, D. M., Sinks, T. and Tilson, H. A. 1996. Research needs for the risk assessment of health and environmental effects of endocrine disruptors: A report of the U.S. EPA-sponsored workshop. Environ. Health Persp. 104 Suppl. 4: 714-740.
McLachlan, J. A. 1997. Synergistic effects of environmental estrogens: Report withdrawn. Science 277: 462-463.
Manson, J. M. and Wise, D. L. 1991. Teratogens. In: Casarett and Doull's Toxicology: The Basic Science of Poisons. pp. 226-254. Amdur, M. O., Doull, J. and Klaassen, C. D. Eds., Pergamon Press, New York.
Soto, A. M., Sonnenschein, C., Chung, K. L., Fernandez, M. F., Olea, N. and Serrano, F. O. 1995. The E-SCREEN assay as a tool to identify estrogens: An update on estrogenic environmental pollutants. Environ. Health Persp. 103: 113-122.
Toppari, J., Larsen, J. C., Christiansen, P., Giwercman, A., Grandjean, P., Guillette, L. J., Jégou, B., Jensen, T. K., Jouannet, P., Keiding, N., Leffers, H., McLachlan, J. A., Meyer, O., Müller, J., Rajpert-De Meyts, E., Scheike, T., Sharpe, R., Sumpter, J. and Skakkebaek, N. E. 1996. Male reproductive health and environmental xenoestrogens. Environ. Health Persp. 104 Suppl. 4: 741-803.