Endocrine Disrupting Chemicals in the Environment

Thought I would share this.  I wrote it while at university and it was completed in 1997.  This is an area that interested me at the time (and still does).  I have changed some Greek characters for the words (like alpha and beta) as I wasn't sure whether they would copy across well.  I hope there aren't too many errors of formatting here.  In the original, there were a few illustrations, which aren't here yet (if I get the time, I will add them).  It's rather long and you will find all the references at the end.  Enjoy!


1. INTRODUCTION

In recent years we have seen headlines in newspapers and the popular scientific press such as: "Research shows fall in male fertility" (Cookson, 1996), and "The emasculators" (Tyler, 1995). These have been referring to the potential of some anthropogenic chemicals to cause reproductive problems. There has been particular attention paid to the ability of some chemicals to imitate the properties of oestrogen. This is not to say that these chemicals are structurally similar to oestrogen - they are often not, but that they appear to mimic the effects of oestrogen to a greater or lesser extent.

Many chemicals in the environment have been implicated in connection with male reproductive problems (Carlsen et al., 1992), breast cancer (Davis and Bradlow, 1995), and even the feminisation of male embryos (Fry and Toone, 1981). A wide range of endocrine disrupting chemicals have been suggested, including the organochlorine pesticides, DDT and dieldrin, and polychlorinated biphenyls (PCBs) (McKinney and Waller, 1994; Soto et al., 1994; Burlington and Lindeman, 1950). These, along with many others, are widely dispersed in the environment. In addition to these man-made chemicals, there are naturally occurring plant oestrogens. These phytoestrogens appear to be able to act as oestrogens within the bodies of animals (Adams, 1995).

It is not only the media that has shown an interest in this subject. A group of scientists from a diverse range of disciplines, including endocrinology, ecology, medicine, toxicology, and wildlife management met at Wingspread, Racine, Wisconsin toward the end of July 1991 to discuss the current state of knowledge with respect to endocrine disrupting chemicals. From this came the Wingspread statement (Colborn and Clement, 1992). Amongst other things, this statement said that many chemicals have the capability of disrupting the endocrine systems of animals, including humans, and that further research should be conducted, especially into predicting the likely disruptive potential of anthropogenic chemicals (Colborn and Clement, 1992).

This discussion looks primarily at recent research that has been conducted into the effects of endocrine disrupting chemicals. In addition, however, older research will be considered, which will show that this is not a recently identified problem, but that the oestrogenic potential of some man-made chemicals was suggested nearly fifty years ago. In this connection, it has been suggested that there is the possibility that one early identified problem of organochlorine pesticides, namely the thinning of eggshells, may have been due to the oestrogenic nature of these compounds (Peakall, 1967; Bitman et al., 1969; Ratcliffe, 1967; Jefferies, 1967).

2. THE ENDOCRINE SYSTEM

2.1 Introduction

Any discussion dealing with endocrine disrupting chemicals would be incomplete without a description of the endocrine system. However, a complete synopsis of all present knowledge as it relates to hormones and the endocrine system will not be presented. The following is a summary of the processes that occur within the endocrine system as it relates to the present theme.

2.2 The Endocrine System in Outline

Hormones are secreted by a system of organs, tissues and cells that make up the endocrine system. These include the ovaries, the placenta, the testes, and the thyroid gland. Hormones regulate various bodily processes, including reproduction and sexual maturation (Wessells and Hopson, 1988). In general, a particular hormone exerts an effect only on particular target cells. These target cells have receptor molecules, which form a hormone-receptor complex, which, in turn, activate the transcription machinery that carries out the cellular response (Wessells and Hopson, 1988). It is thought that specificity is determined by the characteristics of the target cell, not the characteristics of the hormone (Wessells and Hopson, 1988).

The level of a given hormone in the organism is determined by feedback controls. The primary mechanism being negative feedback control, in which a rise in the concentration above a certain level results in inhibition of secretion, and a low concentration gives rise to increased secretion. In addition to negative feedback loops, there are also positive feedback systems involved with some endocrine processes. These are generally considered to be unstable as, when following this path, the release of a hormone stimulates more of its own production. Because of this, positive feedback systems are relatively rare (Laycock and Wise, 1996). Hormones do not usually have a very long residence time in the body as they are usually rapidly broken down, or made biologically inactive prior to excretion in the urine or faeces. When they form a complex with the receptor, this hormone-receptor complex is rapidly metabolised (Laycock and Wise, 1996; Wessells and Hopson, 1988).

There are two main ways in which hormones work, one for steroid hormones, the other for non-steroid hormones. For the purpose of this discussion, it is the steroid hormones that we are most interested in, as it is these that, it has been suggested, have been most affected by xenobiotics. Oestrogen and testosterone are steroid hormones, and follow the pattern to be described. It is interesting, however, that at least one non-steroid hormone, thyroxine, works inside the cell in a similar manner (Wessells and Hopson, 1988).

2.3 Steroid Hormones

Steroid hormones include the oestrogens and androgens, which are female and male sex hormones respectively. The testes and ovaries, from where the sex hormones are secreted, are influenced by gonadotrophins, which are released from the pituitary gland. The release of the gonadotrophins are regulated by releasing factors. The gonadotrophins are the follicle-stimulating hormone (FSH) and luteinizing hormone (LH). In the female, FSH promotes follicular growth and prepares the follicle for the action of LH. In the male, FSH stimulates testicular growth and plays an important role in the early stages of spermatogenesis (Grodsky, 1979b).

In the female, LH stimulates the final maturation of the follicle, ovulation, development of the corpora lutea and secretion of oestrogen and progesterone. In the male, LH stimulates testosterone production by the testis, which maintains spermatogenesis (Grodsky, 1979b). In female mammals, oestradiol stimulates growth of the reproductive tract organs and the mammary glands. In fish, reptiles and amphibians, oestradiol stimulates the liver to synthesise vitellogenin, a precursor to egg yolk (Nimrod and Benson, 1996).

Steroid hormones are able to diffuse through the cell membrane as they are hydrophobic, they bind to a receptor molecule in the nucleus. It is thought that the hormone-receptor complex then attaches to acceptor sites on chromosomes, which activate specific genes that synthesises mRNAs (messenger-RNAs), which translate the message into proteins. The hormone-receptor complex is then broken down, and the gene becomes inactive again (Wessells and Hopson, 1988).

The circulating levels of steroid hormones varies from 10^-6 to 10^-9 M (Grodsky, 1979c). Hormones are excreted from the body regularly, particularly in the urine. Before excretion, the hormones are modified in the liver. For example, oestradiol is often oxidised to oestrone, which is much less potent than oestradiol. Oestrone, in turn, can then be hydrated to oestriol, which is less potent still (Laycock and Wise, 1996; Hadley, 1992). The molecules are then made more water-soluble by conjugation with an acid, such as sulphuric acid, this makes it easier to excrete them in the urine. Although some hormones are excreted unaltered, the main excretion products are the modified molecules (Cook and Beastall, 1987). This renders the hormone, biologically, relatively inactive (Grodsky, 1979a).

It has been estimated that over 70% of circulating oestrogens are bound to a plasma protein called sex-hormone-binding globulin (SHBG), which will also bind testosterone, about another 25% is bound to plasma albumin (Laycock and Wise, 1996). The hormone that is bound to proteins is in equilibrium with the unbound hormone, and it is thought that the physiological effect of the hormone is exerted by this free portion (Brook and Marshall, 1996).

The system is regulated by positive and negative feedback loops (Grodsky, 1979b, Grodsky, 1979c, Nimrod and Benson, 1996). Hormone levels are not regulated in a simple manner. There may be environmental stimuli, as well as biological influences, affecting an organism. Environmental influences can be seen in certain behavioural effects via the central nervous system (Porterfield, 1997). For example, in the courtship behaviour of animals, and humans, in which the presence of the opposite sex can bring about a response. Physiologically, an increase in concentration of a hormone can bring about a reduction in production in a negative feedback mechanism. Additionally, however, the release of, for example, the LH and FSH can also be subject to feedback control (Grodsky, 1979c; Laycock and Wise, 1996). The hormone, 17 beta-oestradiol is also subject to positive feedback control under certain circumstances (Laycock and Wise, 1996). When the plasma concentration of oestrogen rises sharply, and is maintained at a high level (for example, above 800 pmol per litre) for at least 36 hours, in the absence of raised levels of progesterone, a positive feedback situation occurs, in which the release of LH is stimulated (Laycock and Wise, 1996). This appears to be important in the initiating of ovulation, and is kept in check by other feedback processes (Laycock and Wise, 1996).

The above is a very simplified description of some of the processes that take place in the endocrine system. It is possible that any chemical that interacts with, and disrupts the normal function of the endocrine system, could do so through a number of pathways. The present discussion, however, will not be discussing every possible route of disruption.

3. SOURCES OF ENDOCRINE DISRUPTING CHEMICALS

3.1 Introduction

Potentially, endocrine disrupting chemicals come from a number of sources. In addition to the organochlorine pesticides already referred to in the introduction, which were liberated into the environment deliberately, there are others from natural sources, such as oestrogens from cattle slurry (Tyler, 1995). Others include the breakdown product of alkylphenolethoxylates (APEOs), which are used in detergents, namely nonylphenol (Tyler, 1995), PCBs, which were used as coolants in transformers that have been disposed of without due attention to the fate of their contents (Tyler, 1995), and phthalates, which are used in a wide range of products, including paints and plastics (ENDS Report, 1996a).

3.2 Possible Routes of Contact with Endocrine Disrupting Chemicals

As many of the compounds suspected of disrupting the endocrine system are long lived in the environment, it should not be surprising that they may turn up in landfill sites, which receive rubbish from domestic and industrial sources. Li and Hansen (1996) tested the soil, air and dust of a landfill site in Southern Illinois, USA. They found that these extracts contained polychlorinated biphenyls (PCBs), polychlorinated dibenzofurans (PCDFs), with small amounts of polychlorinated dibenzodioxins (PCDDs).

In order to test the biological effects of these landfill extracts, they were administered to prepubertal female rats, which would show a wider range of biological effects than using cell cultures. A range of doses were administered, which were stated as milligrams of PCB per kilogram of body weight. It was found that higher doses (346 mg PCB/kg for soil; 78 mg PCB/kg and 382 mg PCB/kg for dust; and 175 mg PCB/kg for air) produced statistically significant increases in the uterine weights of the animals (Li and Hansen, 1996). This would indicate that the contents of these landfill extracts were having an oestrogenic effect on the animals.

In another study, the two phthalates, butyl benzyl phthalate (BBP) and di-n-butyl phthalate (DBP) were found to be oestrogenic in vitro at concentrations of between 10^-6 and 10^-4 M (Jobling et al., 1995). Phthalates are major components of many plastics, in which they are used to give flexibility and softness (Pirie et al., 1996). BBP was found to stimulate the transcriptional activity of the oestrogen receptor at concentrations of 10^-6 to 10^-4 M, and DBP had a similar effect at concentrations of 10^-5 to 10^-4 M (Jobling et al., 1995). This indicates an oestrogenic effect, but does not indicate directly what effect these compounds would have in vivo, as little is known about how they are metabolised within the body, although it is known that they are lipophilic, and have a tendency to accumulate in fatty tissues (Jobling et al., 1995).

Phthalates are known to be components of sewage effluent, and in recent samples tested in Scotland, phthalates were found to be present in concentrations 1 microgram per litre to 1 689 micrograms per litre. The larger figure was from a sample collected at a sewage outfall. Of this figure, less than 2.1 micrograms per litre was BBP, but DBP accounted for 513.2 micrograms per litre (Pirie, 1996).

In addition to the two phthalates above, Jobling et al., (1995) tested a number of chemicals for oestrogenic activity, among which was the antioxidant butylated hydroxyanisole (BHA), which is a common additive in foods. BHA was also found to stimulate transcription at concentrations of 10^-5 to 10^-4 M as with DBP, but to a lesser extent (Jobling et al., 1995). It has been estimated that the mean human intake of BHA averages 0.13 milligrams per kilogram per day (Jobling et al., 1995). BHA was found to have an oestrogenic effect at concentrations of 2-3 parts per million (ppm) (Jobling et al., 1995), which is about 15 to 23 times the mean daily intake. It is possible that bioaccumulation may take place to a small extent, but this is uncertain (Jobling et al., 1995).

Another source is that of lacquer-coated cans. Brotons et al. (1995) found that bisphenol-A leached from the lining of cans had oestrogenic activity. MCF-7 human breast cancer cells, which are known to be responsive to oestrogen were used in the assays (Soto et al., 1992; vom Saal et al., 1995). In addition to control experiments, MCF-7 breast cancer cells were exposed to 17 beta-oestradiol, which occurs naturally, and would be expected to cause the proliferation of the cells. As expected, the 17 beta-oestradiol experiments caused the cells to increase in number, producing an increase of about six times. The control experiments yielded minimal increase in cell number (Brotons et al., 1995). In addition to the solution in the cans, extracts from some of the vegetable contents also showed oestrogenic activity. These included peas, artichoke hearts, and mushrooms. Peas were found to contain the greatest quantity of bisphenol-A at 22.9± 8.8 mg per can, which produced a statistically significant proliferative effect of the MCF-7 cells, and was shown to have about 58% the proliferative effect of 17 beta-oestradiol (Brotons et al., 1995). The relative amount of bisphenol-A in the vegetables varied greatly, and as a consequence the proliferative effect also varied. For example, bisphenol-A was not detectable in peppers or asparagus but was found in mushrooms, although at a much lower concentration than the peas referred to above (Brotons et al., 1995). While not as effective as 17 beta-oestradiol, bisphenol-A was shown to be significantly oestrogenic.

Brotons et al. (1995) considered the possibility that the oestrogenic effects observed were possibly due to the presence of phytoestrogens or organochlorine pesticides, but neither of these were found to be present in either the cans or the vegetables examined. They also filled cans with distilled water, which was then autoclaved and tested for contaminants. Bisphenol-A was detected in the distilled water after this treatment, which would suggest that the effects observed were due to the substance being leached from the can lining.

The conclusions found by Brotons et al. (1995) concerning the oestrogenicity of bisphenol-A are confirmed by Olea et al. (1996). In this study of materials used in dentistry, a sealant based on bisphenol-A, diglycidylether methacrylate (bis-GMA) was found to increase proliferation of MCF-7 cells (Olea et al., 1996). Bisphenol-A was also detected in the saliva of patients treated with this sealant. In proliferation tests, samples of saliva from these patients increased cell number, so indicating oestrogenicity. It is possible that endogenous oestrogens were responsible for these results, so samples were taken prior to treatment, but none of these showed any oestrogenic activity (Olea et al., 1996).

Plants are a more natural source of environmental oestrogens. Many plant species contain phytoestrogens that can cause oestrogenic effects in mammals (Adams, 1995). For example, clover disease is a well known problem, primarily affecting sheep (Adams, 1995). Temporary infertility results from the animals eating too much clover, or other species containing phytoestrogens. One of the most potent of the phytoestrogens is coumestrol, which appears to exert its effect via the oestrogen receptor (Jordan et al., 1985).

Fertility is usually restored after the oestrogenic feed is removed. However, if the intake of an oestrogenic species is over a long period (four or five months), infertility may be permanent (Adams, 1995). Various levels of infertility due to phytoestrogens, including reduced lambing rates have been observed, which can be a problem in countries, such as Australia, where sheep farming forms an important part of the economy (Adams, 1995).

4. EFFECTS ON WILDLIFE

4.1 Introduction

The ability of any polluting substance to exert an effect on living organisms is dependent upon its ability to come into contact with living systems. The most damaging compounds are ineffective at disrupting living processes if they are locked away in some way. This general principle is as true of endocrine disrupting chemicals as it is of any other polluting substance. One way that these chemicals can come into contact with living systems is through pollution incidents.

The population of common seals (Phoca vitulina) in the Western Wadden Sea, The Netherlands has seen its numbers drop between 1950 and 1975, from about 3 000 to about 500 animals. This appears to be due to a reduction in the number of pups being born (Reijnders, 1986). There was circumstantial evidence that this may be due to PCB pollution from the river Rhine, which mainly effects this part of the North Sea. Animals from the Western Wadden Sea had elevated levels of PCBs compared with the relatively unpolluted Northern North Sea (Reijnders, 1986).

In experiments, Reijnders (1986) found that seals fed fish caught in the more polluted Western Wadden Sea had significantly lower reproductive success compared with those fed on fish caught in the north-east Atlantic Ocean. As the number of animals in his experiments were low, it was not possible for Reijnders to statistically test how likely it was for the problem observed to be associated with disruption of the endocrine system. In this section, however, there are other examples in which pollution incidents have, in all probability, produced an effect on the endocrine system.

4.2 Alligators

A decline in the abundance of American alligators (Alligator mississippiensis) had been noted on Lake Apopka, Florida, USA during the 1980s (Guillette Jr. et al., 1994). The abundance of juvenile alligators showed a particularly marked decline. Their abundance went from about thirty per kilometre of shore line in 1980 to about four per kilometre of shore line in 1983, a situation which has continued to the present, with the lowest abundance being attained in 1985 of about two juveniles per kilometre of shore line (Guillette Jr. et al., 1994).

In 1980 there had been a pollution incident from the Tower Chemical Company involving the pesticide dicofol, but also containing about 15% DDT and its metabolites DDD, DDE, and chloro-DDT, as well as sulphuric acid (Guillette Jr. et al., 1994). It might be expected that these chemicals would exert an effect simply because they are poisonous to living organisms. It was found, however, that the resulting changes in the abundance of Lake Apopka's alligators were probably due to the disruption of their endocrine systems.

Guillette Jr. et al. (1994) compared the Lake Apopka alligators with those in the uncontaminated Lake Woodruff. They found that the plasma concentration of 17 beta-oestadiol was much greater in females from Lake Apopka when compared with the control (about 55% greater), and that the plasma concentration of testosterone was much reduced in males from Lake Apopka (by about 73%). Guillette Jr. et al., (1994) also injected juvenile alligators with Luteinizing hormone (LH), and then measured the plasma concentrations of 17 beta-oestradiol and testosterone. LH stimulates secretion of oestrogen and progesterone in females, and testosterone in males (Grodsky, 1979b). It would be expected, therefore, that the plasma concentration of both 17 beta-oestradiol and testosterone would increase in females and males respectively. The plasma concentration of 17 beta-oestradiol, as measured by Guillette Jr. et al. (1994), increased by 7.6% in females from the control lake, and 54.6% in the control males. The equivalent figures for the Lake Apopka animals were: 44.5% increase in females and 227.6% increase in males. Plasma concentrations of testosterone increased slightly in Apopka males and decreased slightly in Apopka females, whereas the reverse was found in the Lake Woodruff animals, where the males experienced a slight decline and females an increase in plasma testosterone concentration (Guillette Jr. et al., 1994).

From the results of Guillette Jr. et al. (1994), it can be seen that the production of sex steroids is significantly altered in the animals from the polluted lake. The precise mechanism for these changes is still under study, but those mechanisms that have been identified include: reduced gonadotrophin-releasing hormone synthesis from the hypothalamus, reduced LH release from the pituitary, and reduced availability of the required precursors (Guillette Jr. et al., 1994). The fact that oestrogen production was stimulated in the animals that Guillette Jr. and his co-workers injected with LH could suggest that the altered hormone levels in the Lake Apopka animals were due to the xenobiotics reducing the production of LH, or gonadotrophin-releasing hormone. This effect would be expected if the endocrine systems of the animals responded to the xenobiotics as if they were oestrogens, which would result in reduced steroid synthesis. The feedback processes of the endocrine system would maintain the steroid levels at standard concentrations, including the xenobiotics (Grodsky, 1979c).

Variations were also found in the genitalia and internal sexual characteristics of the Lake Apopka alligators compared with the control animals. For example, the average penis size of the Lake Apopka males was 24% less than in the control animals (Guillette Jr. et al., 1994; Guillette Jr. et al., 1996). The sex of the animals was determined by external features, which was then confirmed by dissection and examination of the gonads. It was found that four of the Lake Apopka alligators had been assigned to the wrong sex. Two animals that were identified as females due to the lack of a penis, were subsequently identifies as males as they had testes. The other two were identified as males due to the presence of a penis, but after dissection it was found that they possessed ovaries, and that the 'penis like structures' were due to enlarged clitoral development (Guillette Jr. et al., 1994). None of the Lake Woodruff animals showed such abnormalities. Additionally, many of the other Lake Apopka alligators showed minor gonadal abnormalities compared with the control animals.

It is also interesting to note that Guillette Jr. et al. (1996) found that, although there were reductions in the penis size and plasma testosterone in alligators from Lake Apopka compared with the control lake, the differences were particularly marked in those alligators collected near the entry site of the chemical spill. This would appear to add credence to the suggestion that there appears to be a correlation between the chemical spill and the observed differences between Lake Apopka alligators and those from the control lake.

The presence of xenobiotics in the Lake Apopka system compared with Lake Woodruff would appear to suggest that these may be responsible for the abnormalities observed. This would also agree with observations made by others, such as Burlington and Lindeman (1950), who found that DDT suppressed secondary sexual characteristics and significantly reduced testes size in cockerels.

Despite the fact that the Lake Apopka alligators appeared to be showing signs of excess oestrogen, it should not be assumed that the features identified were due to the xenobiotics mimicking oestrogen. Another possible explanation is that the chemicals were preventing the endogenous androgens binding to the androgen receptor, thus producing a ratio of oestrogens to androgens that is skewed in favour of the oestrogens (Guillette Jr. et al., 1996). A major breakdown product of DDT is p,p'-DDE, which is lipophilic, and was found to be stored in the fat of the Lake Apopka alligators (Guillette Jr. et al., 1996). Kelce et al. (1995) found that p,p'-DDE did not bind to the oestrogen receptor, but did prevent androgens binding to the androgen receptor in adult male rats.

On the other hand, Vonier et al. (1996) conducted experiments to assess the ability of chemicals to bind to alligator steroid receptors. Using tissue prepared from the oviducts of female alligators, the ability of various chemicals to inhibit [³H]17 beta-oestradiol binding to oestrogen receptors was determined. The chemicals found in the Lake Apopka alligators were found to inhibit 17 beta-oestradiol binding to oestrogen receptors. This shows that the alligator oestrogen receptors are "capable of recognizing environmental chemicals" (Vonier et al., 1996). Combinations of chemicals were found to have an additive or a synergistic effect, depending on the combinations involved. The chemical combinations identified in the Lake Apopka system produced a synergistic effect (Vonier et al., 1996). For example, the DDT isomer o,p'-DDT, was found to inhibit binding by 18%, whereas the seven chemicals identified in Lake Apopka (p,p'-DDE, p,p'-DDD, two types of PCBs, dieldrin, toxaphene and chlordane), reduced binding by about 60% (Vonier et al., 1996). Binding to the testosterone receptor was also assessed. Some chemicals did reduce testosterone binding, but this result was not as marked as the oestrogen receptor binding assays (Vonier et al., 1996). Vonier et al. (1996) suggest that their findings support the hypothesis that there is an association between these chemicals in the environment and the observed reproductive abnormalities observed in American alligators on Lake Apopka.

The sex of alligators, and many other reptiles is determined by the temperature of incubation (Bergeron et al., 1994). It has been found that by placing PCBs on the eggshells of developing red-eared slider turtles (Trachemys scripta), the sex could be altered. If eggs were incubated at a cooler temperature, that should have produced all males (in this case 27.8 °C), one PCB isomer (2',4',6'-trichloro-4-biphenylol) resulted in 100% of the hatchlings being female, the same as that produced by the 17b-oestradiol control (Bergeron et al., 1994). Other isomers had varying levels of effect upon the sex of the hatchlings. PCBs usually occur in the environment as a mixture of isomers, and this study, while not proving the link, does show the potential for organochlorines to cause reproductive problems in animals (Bergeron et al., 1994).

4.3 Fish

Vitellogenin is a protein that is a precursor to the synthesis of yolk in various organisms including fish, reptiles and amphibians. It is normally found in females, but not in males (Wahli et al., 1981). Oestrogen controls the production of vitellogenin in the liver, from where it is transported to the ovaries in the blood (Wahli et al., 1981). It is interesting to note that males of the frog Xenopus laevis can be induced to produce vitellogenin in large quantities if oestrogen is administered to them (Wahli et al., 1981). Other steroids, however, such as testosterone, and progesterone will not induce the synthesis of vitellogenin (Wallace, 1978; cited in Wahli et al., 1981).

In recent experiments, male fish exposed to sewage effluent in rivers responded by producing vitellogenin (Sumpter, 1995). The proximity of the caged fish to the sewage outfall, and the condition of the rivers involved, all affected the outcome of in vivo studies into the effects of sewage effluents on male fish (Sumpter, 1995). In a study in Minnesota, USA, male carp (Cyprinus carpio) collected in the vicinity of a sewage outfall also produced vitellogenin (Folmaret al., 1996). In this study, fish were collected from five rivers in Minnesota. The levels of vitellogenin, testosterone and 17b-oestradiol were determined and compared with control fish, which were collected from the St. Croix River, which has National Wild and Scenic River status, and is considered clean (Folmaret al., 1996). The levels of 17 beta-oestradiol were not significantly different between those fish collected from the sewage effluent channel, and those collected from the St. Croix River. The same was not, however, true of the levels of testosterone. Testosterone levels in fish collected from the effluent channel were significantly less compared with those from the clean river, in one case less than half the value obtained from fish collected from the St. Croix River (Folmar et al., 1996).

The vitellogenin levels were determined for both female and male fish. As expected, all females had detectable levels of vitellogenin, whether collected from the St. Croix River, or the effluent channel. Vitellogenin was not detected in males from the St. Croix River, but many collected from the effluent channel had detectable levels. The average concentration of vitellogenin found in females from the effluent channel was 1 706.7 mg/ml serum, with a range of 35-7 500 micrograms/ml serum. The average concentration of vitellogenin detected in males from the effluent channel was 1 113 micrograms/ml serum, with a range of 0-10 000 micrograms/ml serum (Folmar et al., 1996).

Folmar et al. (1996) suggest that their results indicate that one or more chemicals in the sewage effluent are responsible for producing an effect similar to that of oestrogen. They rule out the likelihood of this being due to endogenous oestrogens because the detected levels of 17 beta-oestradiol were not significantly elevated in those fish collected from the effluent channel.

There are many substances occurring in sewage effluent that are potentially oestrogenic, including the synthetic oestrogen used in the contraceptive pill. An earlier study by Jobling and Sumpter (1993) found that the breakdown products of alkylphenol-polyethoxylates, which are surfactants used in detergents, are weakly oestrogenic to rainbow trout (Oncorhynchus mykiss). The most active compound, was 4-tert-butylphenol, which was found to be 1.6 x 10^-4 as potent as 17 beta-oestradiol (Jobling and Sumpter, 1993). While this may not appear to be very significant, these experiments were in vitro, and as such may not tell the whole story. These compounds are lipophilic, and will tend to accumulate in fatty tissues, in which case the effects in vivo may be greater (Jobling and Sumpter, 1993).

Jobling and Sumpter (1993) also tested the effects of tamoxifen on vitellogenin production. Tamoxifen is known to inhibit the binding of oestradiol to the oestrogen receptor. When administered simultaneously with either oestradiol or 4-nonylphenol there was a significant reduction in vitellogenin production, which would seem to suggest that the oestrogenic effects of these compounds is mediated via the oestrogen receptor (Jobling and Sumpter, 1993).

Vitellogenin production in male fish has also been observed in marine fish. Lyle et al. (1997) studied the flounder (Platichthys flesus) in sites off the coast of Great Britain. Two sites in the Tyne Estuary were sampled, and one in the Solway Firth. The Tyne Estuary sample sites were close to a major sewage works, which serves a large population. The Solway Firth is adjacent to a relatively sparsely populated region (Lyle et al., 1997). The protein, vitellogenin, was detected in fish from all three sites, but was much higher in the Tyne Estuary, particularly at sample sites close to the sewage outfall. Additionally, abnormal testes were found in 30% and 53% of the male fish from the two sites in the Tyne, whereas none were found in fish caught in the Solway Firth (Lyle et al., 1997). Lyle et al. (1997) suggest that their findings indicate that this wild population of Platichthys flesus is "suffering disruption to its reproductive health in areas exposed to sewage effluent". They also express concern about the ability of these populations to maintain themselves in waters receiving sewage effluent.

While we can see from the above that xenobiotics can exert an oestrogenic effect upon fish, this is not the only explanation for some of the effects observed. Preliminary results from recent work that has been conducted into this problem has shown that, at least in some cases, the observed effects are not due to chemicals normally considered to be pollutants. As one would expect, sewage is composed of many chemical species, some natural, some not. The chemicals found to be causing the problem were three hormones found in women, 17b-oestradiol and oestrone, which occur naturally, and ethynyl oestradiol, which is a synthetic oestrogen used in the manufacture of the birth control pill (Kaiser, 1996). For example, at a sample site adjacent to a sewage treatment works in Southend-on-Sea, Essex, a peak of 48 nanograms of both 17 beta-oestradiol and oestrone per litre of water were recorded on one occasion. At a sample site in Naburn, Yorkshire, a peak of 76 nanograms of oestrone per litre of water was detected. These results, however, were exceptions as the majority of samples taken yielded much lower concentrations of oestrogens, with the synthetic oestrogen, ethynyl oestradiol not being detected in many of the samples at all (Brighty, 1996). This Environment Agency report suggests that vitellogenin production in fish could possibly be explained by these findings, but that the more exaggerated problems, such as altered gonads could not (Brighty, 1996).

These chemicals would have been excreted, and then passed through the sewage treatment system before entering a river. Prior to excretion, these substances are usually made biologically relatively inactive by transformation to less potent oestrogens such as oestriol, and then the conjugation of an acid, which makes them more soluble in water (Cook and Beastall, 1987). It would appear that the sewage treatment process renders them biologically active again. Dr. Sumpter of Brunel University suggests that bacteria in the treatment process clip off the added chemical groups (Kaiser, 1996). Sumpter also suggests that the results found do not mean that industrial chemicals do not harm fish, especially if pollutant loadings are heavy, but that there are other areas that should be considered (Kaiser, 1996).

This recent work highlights two basic problems; the dangers of assuming the worst possible scenario and; the possible effects of sewage treatment processes upon chemicals prior to discharge. It is, of course, the intention that sewage effluent should be rendered relatively harmless by the treatment processes, but we can see from the above that this is not necessarily always so.

Another area of concern is that of industrial effluent. Davis and Bortone (1992) report incidences of masculinization of female mosquito fish (Gambusia affinis and G. holbrooki) exposed to kraft pulp mill effluent that is discharged into streams in Florida, USA. The main effect was to cause the female anal fin to elongate into a gonodopodium, which the male fish normally uses to transfer and insert sperm. The chemicals involved appeared to be phytosteroids (steroids of plant origin) that had been transformed by the bacterium Mycobacterium smegmatis, although the specific substance was not identified.

Any effluent released into the environment is likely to have an impact, which will depend upon the nature of the release, the level of dilution, and the interactions that take place between the effluent and the environment. Pulp mill effluent is not known to bioaccumulate, but it is released in large quantities, one American pulp mill, for example, was found to release more than 70 000 litres of effluent per minute (Fox, 1992).

4.4 Birds

During the 1950s and 1960s there was an increase in the incidence of broken eggshells in the nests of certain raptors, notable in Britain were peregrine falconFalco peregrinus, sparrowhawk Accipiter nisus and golden eagle Aquila chrysatos (Ratcliffe, 1967). Ratcliffe's 1967 study noted that of 109 peregrine eyries examined between 1904 and 1950 only three egg breakages were found, which represents a breakage rate of about 0.03 broken eggs per eyrie, compared with forty-seven broken eggs in 168 eyries examined between 1951 and 1966, which represents a breakage rate of about 0.3 broken eggs per eyrie. It can be seen that the incidence of broken eggs increased by a factor of ten during the 1951 to 1966 period. Similar results were found for sparrowhawks and golden eagles (Ratcliffe, 1967).

Ratcliffe (1967) used an index of eggshell thinning, which involved taking the mean weight of the eggs and dividing it by the mean index of eggshell size, which was taken to be length times breadth. He found that during the period 1947 to 1967 the index of eggshell thinning for peregrines was about 20% lower than in the period 1900 to 1946. In the case of sparrowhawks, there was a reduction of about 16% during the same time period, and a reduction of about 8.5% was experienced by golden eagles (Ratcliffe, 1967).

This incidence of eggshell thinning and breaking was not, however, geographically uniform. It was noted that eggs examined from the central and eastern Highlands of Scotland did not exhibit the same levels of breakage, or change in the eggshell thinning index. This area was much less intensively farmed, and as a consequence pesticides were used much less. Ratcliffe (1967) also noted that the incidence of egg breakages coincided with the introduction and widespread use of organochlorine, and other, pesticides. This correlation, and the geographical evidence suggested that these may have been responsible.

It is possible that the index of eggshell thinning used above could have indicated density rather than thinning (Ratcliffe, 1967). Bitman et al., (1969), however, found, by direct measurement, that two isomers of DDT, o,p'-DDT and p,p'-DDT caused statistically significant reductions in eggshell thickness of Japanese quail. They found that this was due to a reduction in the calcium content of the shell. The amount of calcium present fell from 2.03% Ca of the fresh weight of the egg, to 1.95% Ca in the case of o,p'-DDT, and 1.96% when p,p'-DDT was administered.

As hormone levels are regulated by feedback loops, particularly negative feedback loops (Laycock and Wise, 1996), it might be expected that elevated body burdens of environmental oestrogens would result in a lowering of the endogenous oestrogen levels. During the time that eggshell thinning was a problem, there were no assaying techniques that were sufficiently sensitive to accurately measure the small concentrations of endogenous oestrogens (A. Dawson, pers. comm.). However, Peakall (1967) found that by administering DDT and Dieldrin to pigeons, the metabolites of two steroid hormones (testosterone and progesterone) increased (table 4.1). Female White King pigeons were used for the progesterone experiments, and male White King pigeons for the testosterone experiments.

	Amount of polar metabolites formed in millimicromoles
	Testosterone	Progesterone
Control	28.7 ± 4.7 (8)	30.1 ± 8.4 (8)
DDT	75.4 ± 18.0 (6)	78.3 ± 8.4 (6)
Dieldrin	111.4 ± 12.7 (6)	90.3 ± 6.1 (6)
DDT + Dieldrin	168.2 ± 9.9 (4)	155.4 ± 17.8 (4)

Table 4.1. Increase of steroid metabolism in pigeons treated with DDT and dieldrin. The figures are (from left to right) mean, standard deviation and number of birds (Peakall, 1967).

As can be seen in table 4.1, above, the administering of DDT increased the metabolites of testosterone by 162.7%, and of progesterone by 160.1%. Dieldrin increased the production of testosterone metabolites by 288.2%, and of progesterone by 200%. The combination of DDT and dieldrin produced increases of 486.1% and 416.3% for testosterone and progesterone respectively (Peakall, 1967). From these results, it would appear that the endogenous levels of these steroid hormones may have decreased.

Risebrough et al. (1968) found that there was a similar increase in oestrogen metabolites when pigeons were treated with DDT, DDE and PCBs (see table 4.2).

Amount of polar metabolites formed in millimicromoles
Control	29.3 ± 6.5
DDE (40mg/kg)	76.2 ± 13.1
DDT (40mg/kg)	93.1 ± 11.2
PCB (20mg/kg)	160.0 ± 10.5

Table 4.2. Increase of oestradiol metabolites in pigeons treated with three organochlorine substances. The figures are means and standard deviations (Risebrough et al., 1968).

As can be seen in table 4.2, DDT, for example, produced an increase in oestradiol metabolites of 217.7%. The presence of metabolites does not, however, demonstrate with certainty that the endogenous levels of steroids have declined, but the indirect evidence does indicate that this may be so. Endogenous oestrogen is important for calcium absorption (Sturkie, 1954), and it may be that the problem of eggshell thinning may have been one of disrupted endocrine operation.

Bar and Hurwitz (1979) found that by administering gonadal hormones (17 beta-oestradiol and testosterone) to male chicks the rate of calcium absorption increased. This increase in calcium absorption was found to be independent of dietary calcium intake (Bar and Hurwitz, 1979). In another experiment, it was found that duodenal calcium absorption increased in a dose responsive way when 17 beta-oestradiol was injected into male chicks (Sommerville et al., 1989). This dose-responsive result spanned doses of 0 to 100 mg oestrogen per day (100 micrograms = 0.7 micrograms per gram body weight). At doses in excess of 100 micrograms oestrogen per day the level of calcium absorption fell (Sommerville et al., 1989).

We can see from the above that calcium absorption appears to be dependent upon endogenous oestrogen levels, and that there is indirect evidence linking xenobiotics with reduced steroid concentrations. It is possible, therefore, that eggshell thinning, and reduced levels of eggshell calcium that were linked to organochlorine pesticides were due to reduced levels of endogenous oestrogens, brought about by the oestrogenic properties of these compounds.

As far back as 1950 Burlington and Lindeman found that DDT administered to White Leghorn cockerel males resulted in the development of secondary sex characteristics being reduced. For example, the combs of treated birds were significantly decreased in size. After 52 days of treatment, the average comb size of the control group was measured at 11.7 cm, compared with a measurement of 7.56 cm for the treated group (Burlington and Lindeman, 1950). It was thought that this may have been due to the testes being reduced in size. It was found that the mean weight of the testes from treated birds was only 18% that of the control group - 1.05 g compared with 5.63 g for the control birds (Burlington and Lindeman, 1950). Burlington and Lindeman suggested in their report that their findings may indicate that DDT exerts an "estrogen-like action".

However, other explanations have been suggested for the phenomenon of eggshell thinning. For example, administering DDT to mallard ducks (Anas platyrhynchos) was found to reduce the activity level of calcium adenosine triphosphatase (Ca-ATPase), which is thought to be responsible for calcium transport (Kolaja and Hinton, 1977). During the experiments reported by Kolaja and Hinton (1977), eggshell thickness was reduced, as was the total calcium content of the shell. They also noted the accumulation of DDT in the fatty tissues of the birds. The controls (those that received no DDT with their food) had a DDT content in their abdominal fat of 176 ppm, compared with 21 590 ppm DDT in the fat of the ducks fed 50 ppm DDT with their feed for six months (Kolaja and Hinton, 1977). From this we can see that a reduction in the activity of this enzyme could be the primary cause of reduced calcium levels (Cooper and Kavlock, 1997). It is also possible that there is not a simple explanation, but that some combination of the above, and possibly other, processes are involved.

There has been shown to be a significant correlation between p,p'-DDT and a delay in ovulation in Bengalese finches Lonchura striata (Jefferies, 1967), which could have important implications for the successful breeding of birds exposed to this substance. DDT has been shown to have other detrimental effects on reproduction. For example, DDT injected into the eggs of western gulls Larus occidentalis resulted in the feminization of males, and caused the growth of both the right and left oviducts in females (the development of the right oviduct is abnormal) (Fry and Toone, 1981). It was found that the DDT isomer that had the greatest oestrogenic effect was o,p'-DDT. At low doses primordial germ cells (PGC) were localised to form a thickened, ovary-like cortex in the males. Also in the males and at higher doses, testicular changes occurred, including the development of a left oviduct and shell gland (Fry and Toone, 1981). It is possible that these changes were responsible for the female-female pairings and skewed sex ratios that were observed in Southern California (Fry and Toone, 1981; Hunt and Hunt, 1977).

Additionally, birds exposed to organochlorine pollution have been shown to be poor parents. For example, Forster's terns (Sterna forsteri) living at Green Bay, Lake Michigan, USA were found to have low reproductive success (Kubiak et al., 1989). These birds were exposed to a number of compounds, including 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) and PCBs, and were shown to have a much lower rate of successful hatchings compared with an uncontaminated control area (Kubiak et al., 1989). It was found that the rate of successful hatchings of eggs from the contaminated site was greatly improved when they were incubated by birds living in the uncontaminated area, whereas eggs from the uncontaminated area that were incubated by the birds from the contaminated area showed a marked reduction in successful hatchings (Kubiak et al., 1989).

We can see from the above that anthropogenic chemicals, particularly organochlorines that have been released into the environment (whether deliberately of accidentally), have produced effects upon birds that could be described as disruption of the endocrine system. It will also be noted that these effects have occurred over at least the last fifty years, and are not a recent phenomenon.

5. EFFECTS ON HUMANS

5.1 Introduction

Whilst there has been much concern expressed with respect to the effects of environmental chemicals on wildlife, there has been even greater concern expressed as to the effects these may have on humans. People do not, generally, live insulated from the wider environment, so there is at least the potential for harm to be done to human populations. In this section the evidence for, or against, environmental endocrine disrupting chemicals exerting an effect on humans will be discussed. The areas of primary concern are male reproductive health (level of sperm production and abnormalities of the reproductive organs) and some types of breast cancer in women.

5.2 Male Reproductive Health

In recent years concern has been expressed about the quantity and quality of sperm, as well as other aspects of male reproductive health. Carlsen et al. (1992) reviewed the literature between 1938 and 1990. They excluded publications that included men from infertile couples, men with genital abnormalities, those with a bias toward men with high or low sperm count, and where computer assisted counting had been used. It was found that there appeared to have been a small decline in seminal volume per ejaculate between 1940 and 1990, from a mean of 3.40 ml to 2.75 ml. There was also a significant decrease in mean sperm concentration between 1940 and 1990 from 113 x 10⁶/ml to 66 x 10⁶/ml (Carlsen et al., 1992).

The method used by Carlsen et al. (1992) was essentially a statistical technique, and the period covered was some fifty years to 1990, so there is the possibility that there could be statistical errors, or that like may not be being compared with like when different studies are compared. Bromwich et al. (1994) assert that the decline in sperm concentration observed by Carlsen et al. (1992) may have been due to a change in the reference value. In the 1940s the reference value for a "normal" sperm count was 60 x 10⁶/ml, which was subsequently changed to the present value of 20 x 10⁶/ml (Carlsen et al., 1992).

Other studies have not found reductions in sperm counts in recent years. For example, sperm samples from 1 283 men that were deposited with sperm banks before vasectomy from 1970 to 1994 were examined in the USA (Fisch et al., 1996). They found that over the twenty-five year period, sperm concentration increased slightly, and that there was no change in either sperm motility or semen volume (Fisch et al., 1996). However, Fisch et al. (1996) did find that there were variations between the three sperm banks, and that sperm motility and semen volume decreased with increasing age at the time of collection. These results are supported by the findings of Paulsen et al. (1996). In their study of 510 men over the twenty-one year period 1972 to 1993, they found that there had been no decrease in sperm concentration, semen volume, total number of sperm per ejaculate and percent normal sperm in the Seattle area involved.

Carlsen et al. (1992) looked at 61 studies before they came to their conclusion that sperm quality had declined, but as Fisch and Goluboff (1996) point out, these studies varied considerably. They involved from as few as 7 to as many as 4 435 men. Fisch and Goluboff (1996) excluded those studies that had only a small number of subjects in them. They were left with 20 studies that had at least 100 men. They found that their analysis showed no decline in sperm quality or quantity. They did, however, find significant regional variation. For example, New York had high average sperm counts, whereas third world countries had low average sperm counts. Earlier studies were centred in areas like New York, whereas later studies also include third world countries, which may explain the apparent decline in sperm quantity over recent years in some studies (Fisch and Goluboff, 1996).

It is clear from the foregoing that there is not a consensus of opinion with respect to the levels of sperm production. However, a recent Finnish study came to the conclusion that between 1981 and 1991 there had been a decrease in normal spermatogenesis (Pajarinen et al., 1997). These workers examined the results of two studies, that took place in 1981 and 1991, of men that had been subject to post-mortems. They found that in 1981, 56% of the subjects had normal spermatogenesis, whereas in 1991 this had fallen to 27%, which was considered to be statistically significant (Pajarinen et al., 1997). It was also found that there was a statistically significant reduction in testicular weight of 5.8% between the 1981 and 1991 studies (Pajarinen et al., 1997).

Pajarinen et al. (1997) suggest that their results may help to explain the decline in sperm counts observed in some other studies, such as those referred to above. The presence, or absence, of xenobiotics was not tested for in these studies, so it is not possible to link chemicals with the results obtained. However, other risk factors were considered, such as smoking and obesity. Although there was a slightly larger proportion of men that were over-weight and/or smoked in the 1991 study, there was not found to be a significant correlation between these, or any other, risk factors that were considered (Pajarinen et al., 1997). In spite of these apparent problems, the mean sperm concentration of Finnish men was found to be nearly double that of men world wide (Souminen and Vierula, 1993; cited in Joffe, 1996).

It has been suggested that if there was any decline in sperm concentration, this may be due to environmental factors, such as environmental oestrogens (Carlsen et al., 1992). In this regard, it is interesting to note that semen samples from members of the Danish Organic Farmers Association were found to have a high sperm density when compared with men from other professions (Abell et al., 1994). Organic farmers grow produce without the aid of pesticides or chemical fertilisers, and among the members of this study, the average intake of organic dairy products was at least 50% (Abell et al., 1994). This could, of course, be merely coincidental, or it could indicate a connection between chemicals and sperm count.

In a study conducted by Sharpe et al. (1995) it was found that male rats exposed to 4-octylphenol (OP) and butyl benzyl phthalate (BBP) experienced a reduction in daily sperm production and a reduction in mean testicular weight. The study examined the effects of these chemicals when taken by the females throughout the reproductive period (prior to, during, and after pregnancy), and when taken by the young only after birth. The chemicals were fed to the animals in their water. The males were therefore, either exposed continually during pregnancy, or after birth only.

It was found that only minor reductions in testicular size occurred with the males that were exposed to OP and BBP after birth only. On the other hand, males that were born from the females that had been exposed to these chemicals during the whole reproductive period had significantly reduced testicular weights. The mean daily sperm production was reduced in those males exposed throughout the reproductive cycle by 10-21% (Sharpe et al., 1995). Sharpe et al. (1995) point out that their data does not prove that there is a link between environmental oestrogens and falling sperm counts, but that it is possible for these chemicals to have such an effect.

Whether, or not, there has been a decline in sperm counts over the past fifty years, there has not been a decline in fertility rates, that is the ability of men to father children. For example, over the past thirty years the infertility rates have remained fairly constant at 8-11% in the USA (Hall 1996). Fertility, however, is not uniform, as noted in a study by Joffe (1996), who found that Finland had a statistically greater fertility rate than Great Britain. As Fisch and Goluboff (1996) point out, the level of sperm production is also generally lower in developing countries. If this were to have an effect on fertility we would expect the rate of population increase to be reduced in these countries. The reality, however, is just the opposite. Whereas the rate of population increase in developed nations has been declining (from 1.27% average annual increase in 1960 to 0.64% average annual increase by 1985), the rate of population increase in third world countries rose from 2.13% average annual increase in 1960 to 2.53% average annual increase by 1970. After 1970 the rate of increase fell to 2.01% by 1985, which is still much higher than in the developed countries (United Nations Environment Programme, 1989). As can be seen from this data, the level of increase was generally much greater in the developing world than the developed, which is the opposite to what would be expected if fertility had been affected by reduced sperm counts. The decline in the rate of population increase in the developing countries after 1970 may have been due to improved governmental policies, and education to deal with high population levels, including improved knowledge and availability of contraception, rather than any endocrine disrupting influence. While it may be that organochlorine pollution has affected sperm production in the developing nations, it is also possible that sperm production has been affected by lower levels of nutrition in these countries. Additionally, it is interesting to note that while people living in developing nations have, on average, higher concentration of DDT in their tissues than those in the developed world, people in the developed world have, on average, higher concentrations of PCBs, and other industrial chemicals in their tissues than people in the developing nations (Thomas and Colborn, 1992).

Other problems in which environmental oestrogens have been implicated are testicular cancer, cryptorchidism (testicular maldescent) and hypospadias (urethral abnormalities). Testicular cancer has increased by varying amounts in different countries since 1960. For example, it has increased by about 50% in Finland between 1960 and 1987, and by 138% in the south western UK during the same period (Giwercman, 1995). Hypospadias has shown similar variable increase. It increased by 26% in Sweden between 1970 and 1990, and by 250% in Norway during the same period (Giwercman, 1995).

It is known that some males born to women who were exposed to the drug diethylstilbestrol (DES), which is a synthetic oestrogen, during pregnancy suffered from these reproductive abnormalities to varying degrees (Sharpe and Skakkebaek, 1993). If it is possible for DES to exert an oestrogenic effect, it is possible that environmental pollutants that have been shown to have oestrogenic capabilities in the laboratory could also produce a similar result.

In a study conducted in the Spanish province of Granada there appeared to be a correlation between the incidence of surgery for cryptorchidism (orchidopexy) and the extent of pesticide use (Garcia-Rodriguez et al., 1996). Admissions to the University of Granada Hospital were monitored, and the level of orchidopexy operations were estimated from 1980 to 1991 in all areas served by the hospital. It appeared that the incidence of orchidopexy was greatest amongst males from those areas that had the greatest pesticide usage (Garcia-Rodriguez et al., 1996). However, Garcia-Rodriguez et al. (1996) suggest that caution should be applied to their interpretations as it is possible that selection bias may have been introduced by including only patients having surgery at the only public hospital covering the entire study area. They also suggest that there may also have been problems associated with the classification of the level of pesticide use.

In spite of the possible areas of bias in the above study by Garcia-Rodriguez et al. (1996), it still shows that there is the possibility of damage being done to the male reproductive system by environmental chemicals. In many areas of the developed world, organochlorine pesticides are no longer widely used. However, it is interesting to note that in the area studied by Garcia-Rodriguez et al. (1996), organochlorine pesticides, including lindane and p,p'-DDT, are still being used.

While it may appear that testicular abnormalities, and reduced sperm counts are the result of oestrogen-like chemicals, this is not the only possible way that problems can arise. Male sexual differentiation is, in part, mediated by the steroid hormones testosterone and dihydrotestosterone. In the absence of these androgens a genetic male will not develop properly (Wessells and Hopson, 1988). An androgen antagonist, that is a substance that prevents androgens binding to the androgen receptor could be expected to produce a feminizing of the male.

The chemical vinclozolin is used as a fungicide on several fruits, vegetables, ornamental plants, and vines, and has been found to be an androgen antagonist (Gray et al., 1994). By administering vinclozolin to pregnant female rats during the period of sexual differentiation, Gray et al. (1994) found that the male offspring had genital abnormalities, including hypospadias. When they were first born, all the male rats were incorrectly identified as females, and they were unable to inseminate the females, despite being able to mount in the normal manner (Gray et al., 1994).

The doses administered, 100 milligrams per kilogram per day and 200 milligrams per kilogram per day, are quite large, but they did not prove toxic to the animals, and the females experienced only minor physiological changes (Gray et al., 1994). As this chemical is used on food crops there is the potential for it to be ingested by humans. However, for an effect to be experienced it is likely that a great quantity of treated fruit would have to be eaten during the period of sexual differentiation for there to be cause for concern.

5.3 Female Reproductive Health

It has long been known that breast cancer is often associated with lifetime exposure to oestrogens, which may be increased by the early start of menstruation, entering the menopause late and having never breast-fed a child (Davis and Bradlow, 1995).

DDT and other organochlorines are lipophilic, it is likely, therefore, that any exposure to these chemicals would result in the accumulation of them in the fatty tissues of the breast (Kang et al., 1996). It has been shown that many xenobiotics proliferate oestrogen-sensitive breast cancer cells (Soto et al., 1994; Soto et al., 1992). For example, both dieldrin and DDT were found to be oestrogenic, causing oestrogen-sensitive breast cancer cells to proliferate, but at a much greater concentration than was required for 17 beta-oestradiol. 17 beta-Oestradiol required 10 pM to cause significant proliferation, whereas the xenobiotics required 10 micromoles (Soto et al., 1994).

While it may appear that the doses required are much greater than endogenous oestrogen, there is evidence to suggest that many xenobiotics can work in a synergistic manner. A mixture of endosulfan and dieldrin, for example, has been found, by some workers, to be 160 to 1600 times more oestrogenically potent than either chemical alone (Arnold et al., 1996a). There is also indirect evidence linking xenobiotics with breast cancer. In a study conducted by Dewailly et al. (1994) in the Quebec city region of Canada, it was found that there was a correlation between xenobiotic loading and oestrogen receptor-positive breast cancer (ER-positive). There was a statistically significantly higher concentration of the DDT metabolite DDE, which occurred at a mean of 2132.2 micrograms per kilogram in the ER-positive breast cancer patients, compared with a mean of 765.3 micrograms per kilogram for the control group (Dewailly et al., 1994).

In a study conducted by Wolff et al. (1993), a statistically significant correlation was found between the levels of DDE and the incidence of breast cancer. In this study, the serum levels of DDE and PCBs were determined in blood collected from women who subsequently developed cancer. They were matched with a control group, consisting of women with similar risk factors, such as family history of breast cancer, lifetime lactation record and age at first full-term pregnancy. The analysis was conducted in matched pairs. Only the DDE analysis produced a statistically significant result, with breast cancer sufferers having about 35% more DDE than controls (Wolff et al., 1993). The levels of PCBs in the breast cancer patients were about 15% higher than the controls, but this was not statistically significant (Wolff et al., 1993).

Wolff et al. (1993) suggest that a possible explanation for their results may be the length of lactation. It may be that by lactating for longer the xenobiotic residue is reduced. In this study there was found to be a 'strong effect' exerted by lactation on the association between DDE and breast cancer (Wolff et al., 1993).

It has been observed that most tumour promoting chemicals inhibit the gap junctional intercellular communication (GJIC) in cells, that is the passage of ions and small molecules between adjacent cells, which it is thought allows the release of cancer initiating cells, which then go on to promote tumour growth (Kang et al., 1996). Kang et al. (1996) tested DDT, dieldrin, toxaphene, various PCB metabolites, and metabolites of polybrominated biphenyls (PBBs) on normal human breast cancer epithelial cells. If they are shown to inhibit GJIC, then it is likely that they have the ability to promote breast cancer, but not necessarily cause it (Kang et al., 1996).

In this study, DDT, dieldrin, toxaphene, the PCB metabolite 2,2',4,4',5,5'-HCB and the PBB metabolite 2,2',4,4',5,5'-HBB were found to inhibit GJIC in normal human breast epithelial cells (Kang et al., 1996). The effect was dose-dependent and reversible, which indicates that the doses applied (6-100 mM for DDT, and 3-50 micromoles for dieldrin, for example) were not cytotoxic, that is toxic to the cell contents (Kang et al., 1996).

In the study by Kang et al. (1996) the DDT isomers involved were not specified, but these results suggest that the associations found by Wolff et al. (1993) and Dewailly et al. (1994) between DDE residue loading and breast cancer incidence may be valid.

The endometrium, which is the lining of the womb, is subject to a condition known as endometriosis, in which the growth of endometrial cells occurs outside the uterus (Rier et al., 1995). Traditionally, it has been treated by limiting the action of endogenous oestrogen (Rier et al., 1995), which suggests that the presence of oestrogen, or a substance that mimics its effects, may promote the incidence of endometriosis. Rier et al. (1995) exposed rhesus monkeys (Macaca mulata) to 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD, also known as dioxin) for about four years. Ten years after treatment stopped, they found that 43% of the animals exposed to five parts per trillion (ppt) dioxin, and 71% of those exposed to 25 ppt dioxin had developed moderate to severe endometriosis. None of the control animals (those that did not receive dioxin) developed the condition at this degree of severity, although some did suffer from endometriosis at a lower severity (Rier et al., 1995).

At first glance this may appear to indicate an oestrogenic effect on the part of dioxin. However, dioxin has been shown to have anti-oestrogenic properties, including the inhibition of mammary tumours in rodents, and the inhibition of oestradiol-induced proliferation of human breast cancer cells (Safe and Krishnan, 1995). Safe and Krishnan (1995) state that 'the mechanisms of interaction between TCDD- and oestradiol-induced signalling pathways are complex', and Rier et al. (1995) suggest that while incidence of endometriosis may be associated with dioxin exposure, this may not be a simple endocrine system association, and that the immune system may also be involved.

5.4 Is There a Problem?

Although some evidence has been presented above for human reproductive health being adversely affected by environmental chemicals disrupting the hormone systems of people of both sexes, a simple cause and effect should not be assumed. In the view of some scientists the case for endocrine disruption has been "overstated" (A. Dawson, pers. comm.). The fact something has been shown to be possible does not mean that it will happen in vivo in the environment. For example, the study by Sharpe et al. (1995) demonstrated a link between certain chemicals and reduced sperm production in the laboratory rat. From the studies on humans detailed above, however, we can see that a much more complicated picture results, and it is not possible to state with certainty that there is a correlation between exposure to xenobiotics and reproductive problems.

It would appear that many xenobiotics are, at best, weak oestrogens, which do not have the level of effect that endogenous oestrogens do (see table 5.1).

Compound	Concentration	Effect relative to oestradiol (%)
Oestradiol	30 picomoles	100
p,p'-DDT	30 micromoles	0.0001
o,p'-DDD	30 micromoles	0.0001
Coumestrol	3 micromoles	0.001

Table 5.1. Relative oestrogenic potential of selected substances (Soto et al., 1992).

We can see from table 5.1 that the DDT metabolite shown is considerably less oestrogenic than oestradiol. The phytoestrogen coumestrol is ten times more oestrogenic than DDT or DDD, but still considerably less so than oestradiol. Safe (1995) argues that it is not plausible to attribute any connection between industrial chemicals and increased incidence of breast cancer or male reproductive disorders.

Thomas and Colborn (1992) on the other hand are of the view that the evidence strongly suggests a link between xenobiotics and reproductive health. Many of the chemicals involved, such as PCBs, PBBs and organochlorine pesticides are frequently present in human tissue, and are chemically stable, with slow rates of degradation, are soluble in lipids, with a tendency to bioaccumulate, and biomagnify along food webs (Thomas and Colborn, 1992). Along with these considerations, there is evidence to suggest that these chemicals may work in a synergistic manner (Arnold et al., 1996a). This, however, is by no means certain as other workers have failed to reproduce the results of Arnold and his co-workers. Both Ashby et al. (1997) and Ramamoorthy et al. (1997) have disputed claims of synergy between environmental oestrogens. Ramamoorthy et al. (1997) are of the opinion that these weakly oestrogenic chemicals are, at most, additive in combination. In this regard, McLachlan et al. (1997) suggest that synergy can be shown to occur, and that it is differences in the assaying techniques that are causing confusion. This illustrates the complex nature of the mechanisms involved, and the difficulty of assessing the true nature of the problem.

It may be that a cause and effect cannot always be proved, and as Safe (1995) says, these chemicals are often at best only weak oestrogens. However, there is not likely to be a simple answer to the question of what effect, if any, environmental chemicals may have on the reproductive health of humans. For example, Safe (1995) dismisses p,p'-DDE as not being oestrogenic, and thus not worth considering further. Kelce et al. (1995), however, found that p,p'-DDE is a potent androgen receptor antagonist, and it has been suggested that this may bring about an oestrogenic-like action by changing the endogenous oestrogen/androgen ratios (Guillette Jr. et al., 1996).

For any chemical, or chemical combination, to induce a biological response, it has to be at a concentration that is capable of inducing a specific response in the target cell, or cells, (Arnold et al., 1996b). It has already been noted that environmental oestrogens are only weak oestrogens. Some of the in vivo responses noted previously, however, appear to suggest that some chemicals are capable of exerting a greater oestrogenic effect than data, such as that in table 5.1, would appear to indicate. Much of the data that has been obtained to determine the relative effects of environmental chemicals has been obtained in the laboratory in vitro. Much of this research has looked at the relative binding affinities (RBAs) of these compounds compared with 17 beta-oestradiol (Nimrod and Benson, 1996), that is the ability of a substance to bind with the hormone receptor. This may be misleading as there are other factors that should be considered. Some have already been mentioned, such as bioaccumulation and slow rates of degradation. Another, biological, aspect to consider is that of a chemical's ability to bind to the sex hormone binding globulin (SHBG).

In section 2.3 it was said that most circulating oestrogens are bound to either SHBG or plasma albumin, with only a very small amount (about 5%) being unbound. It is thought that it is this unbound fraction that induces the biological effect (Laycock and Wise, 1996; Brook and Marshall, 1996). It has been found that DES, for example, has a much lower affinity for the SHBG than oestradiol, so that, at equivalent concentrations in the blood, the concentration of DES will be greater than oestradiol at the target cell (Arnold et al., 1996b).

Arnold et al. (1996b) found that the xenoestrogens o,p'-DDT and octyl phenol had a lower affinity for SHBG and albumin than oestradiol. They suggest that their findings could indicate that certain environmental oestrogens may exert greater oestrogenicity than had been predicted previously (Arnold et al., 1996b).

It is clear from the foregoing that the interactions are complex, and that there can be dangers involved when attempting to extrapolate findings in the laboratory to the natural environment. It is, however, difficult to dismiss the data presented, and it is probable that there is a problem associated with endocrine disrupting chemicals, but that men are not about to be rendered sterile, nor are all women going to develop breast cancer from coming into contact with environmental chemicals.

6. CONCLUSIONS

The effects of the xenobiotics, and other chemicals, discussed in this project are, for the most part, sub-lethal. Even in the most extreme cases, such as the alligators living in Lake Apopka (Guillette Jr. et al., 1994; Guillette Jr. et al., 1996) that were affected by a chemical spill, death did not result directly. The effects noted were often centred around the ability to reproduce. Of course, this could result in reduced population levels in the same way that poisoning the animals could.

The evidence for reduced sperm quantity and quality appears the most tenuous, as discussed above, while the evidence for an association between xenobiotics and some types of breast cancer appears a little more certain. However, as already said, a simple cause and effect should not be assumed (Safe, 1995) as there may be reasons, other than disruption of the endocrine system, for the apparent correlations observed.

The research findings that have been outlined in this project have prompted concerns as to the likely future environmental impacts of the implicated chemicals. For example, the Worldwide Fund for Nature (WWF) has called for a 'ban on oestrogenic pesticides' (ENDS, 1996b), and Sweden has been pushing for a phase-out of 'oestrogens in pesticides' (ENDS, 1996c).

There is a reluctance on the part of many companies to accept that there is the potential for a problem, or that there is any urgency in making changes to chemical effluents. For example, one Manchester based manufacturer, AKCROS, which releases large amounts of alkylphenol ethoxylates (APEs), which have been reported to be oestrogenic (Nimrod and Benson, 1996), and other chemicals, such as butyl phthalate, into the local sewage works, has made a request to the Environment Agency, asking that it can continue to make discharges that are 125 times greater than the Agency originally specified (ENDS, 1997).

This company asked that it be allowed a new time scale in which to comply with its integrated pollution control (IPC) authorisation. On the other hand, major wool scouring companies have had no problems phasing out their discharges of APEs to comply with future IPC requirements. They found that the cost was less than expected, and that there was no deterioration in product quality (ENDS, 1997). The above may illustrate different attitudes to the same problem, or it may be that the circumstances between the two industries are different.

In the USA, laws have been modified to take account of endocrine disrupting chemicals. Amendments were made on the third of August 1996 to the Federal Insecticide, Fungicide, and Rodenticide Act, the Federal Food, Drug, and Cosmetic Act, and the Safe Drinking Water Act to take account of substances that may have endocrine effects. These laws require that a screening programme be developed within three years, and that the public health should be protected (Centre for the Study of Environmental Endocrine Effects web site).

For a screening programme to be set up, it is necessary to be able to assess the ability of a chemical to disrupt the endocrine system. It is perhaps ironic that one of the problems that has been outlined in section 4.3, namely the production of vitellogenin in male fish, has been suggested as a way of assessing the oestrogenic activity of xenobiotics, particularly in populations of reptiles and amphibians (Palmer and Palmer, 1995). This, however, would only look at the problem from one viewpoint, it could be difficult to quantify the results, and it would tell little about the effects that xenobiotics may have on other classes of animals, including humans. It is for this reason that Shelby et al. (1996) suggest using three assaying techniques. These would provide information on the ability of a chemical; to bind to the receptor, to promote transcription, and how the chemical reacts in vivo. Klotz et al. (1996) also suggest a combination of assays, but all in vitro. These workers have suggested using the Yeast Oestrogen Screen (YES), in which a strain of yeast has been genetically modified to express the human oestrogen receptor. The result of the assay is determined by the activity of beta-galactosidase, the greater the activity level is increased, the greater is the oestrogenic potency of a tested substance (Klotz et al., 1996). Additionally, Klotz and her co-workers have suggested using the proliferative effect of oestrogen-like substances on MCF-7 breast cancer cells, and a competitive binding assay, using human oestrogen receptors.

As has already been suggested, there may be problems associated with relying on laboratory based assays as they do not necessarily reflect what will happen in a living organism. It is clear that, as Safe (1995) says, the interactions of chemicals in the environment, and in the organism, are complex, and that more research is needed if the environmental effects of endocrine disrupting chemicals are to be fully understood.

REFERENCES

Abell, A., Ernst, E. & Bonde, J.P., (1994), High sperm density among members of organic farmers' association, Lancet, 343, 1498.

Adams, N.R., (1995), Detection of the effects of phytoestrogens on sheep and cattle, J. Anim. Sci., 73, 1509-1515.

Arnold, S.F., Klotz, D.M., Collins, B.M., Vonier, P.M., Guillette, L.J. & McLachlan, J.A., (1996a), Synergistic activation of estrogen receptor with combinations of environmental chemicals, Science, 272, 1489-1491.

Arnold, S.F., Robinson, M.K., Notides, A.C., Guillette Jr. & McLachlan, J.A., (1996b), A yeast estrogen screen for examining the relative exposure of cells to natural and xenoestrogens, Environ. Health Perspect., 104, 544-548.

Ashby, J., Lefevre, P.A., Odum, J., Harris, C.A., Routledge, E.J. & Sumpter, J.P., (1997), Synergy between synthetic oestrogens?, Nature, 385, 494.

Bar, A. and Hurwitz, S., (1979), The interaction between dietary calcium and gonadal hormones in their effect on plasma calcium, bone, 25-hydroxycholecalciferol-1-hydroxylase, and duodenal calcium-binding protein, measured by a radio-immunoassay in chicks, Endocrinology, 104, 1455-1460.

Bergeron, J.M., Crews, D. & McLachlan, J.A., (1994), PCBs as environmental estrogens: Turtle sex determination as a biomarker of environmental contamination, Environ. Health Perspect., 102, 780-781.

Bitman, J., Cecil, H.C., Harris, S.J. & Fries, G.F., (1969), DDT induces a decrease in eggshell calcium, Nature, 224, 44-46.

Brighty, G., Project Manager, (1996), The identification and assessment of oestrogenic substances in sewage treatment work effluents, Research and Development Technical Summary P38, Environment Agency, Bristol.

Bromwich, P., Cohen, J., Stewart, I. & Walker, A., (1994), Decline in sperm counts: an artefact of changed reference range of "normal"?, BMJ, 309, 19-22.

Brook, C. and Marshall, N., (1996), Essential Endocrinology, third edition, Blackwell Science, Oxford, London, Cambridge, Massachusetts, USA, Carlton Victoria, Australia.

Brotons, J.A., Olea-Serrano, M.F., Villalobos, M., Pedraza, V. & Olea, N. (1995), Xenoestrogens released from lacquer coatings in food cans, Environ. Health Perspect., 103, 608-612.

Burlington, H. and Lindeman, V.F., (1950), Effect of DDT on testes and secondary sex characters of white leghorn cockerels, Proc. Soc. Exp. Biol. Med.,74(1), 48-51.

Carlsen, E., Giwercman, A., Keiding, N. & Skakkebaek, N.E., (1992), Evidence for decreasing quality of semen during past 50 years, BMJ, 305, 609-613.

Centre for the Study of Environmental Endocrine Effects, 11 Dupont Circle, Washington, DC 20036.

Colborn, T. and Clement, C. (1992), Statement from the work session on chemically-induced alterations in sexual development: the wildlife/human connection, in Chemically-Induced Alterations in Sexual Development: The Wildlife/Human Connection, eds Colborn, T. and Clement, C., volume XXI of Advances in Modern Toxicology, series ed Mehlman, M.A., Princeton Scientific Publishing Co., Inc., New Jersey, pp 1-8.

Cook, B. and Beastall, G.H., (1987), Measurement of steroid hormone concentrations in blood, urine and tissues, in Steroid Hormones, eds: Green, B. and Leake, R.E., IRL Press, Oxford and Washington, pp 1-65.

Cookson, C., (1996), Research shows fall in male fertility, Financial Times, 23rd. February 1996.

Cooper, R.L. and Kavlock, R.J., (1997), Endocrine disruptors and reproductive development: a weight-of-evidence overview, J. Endocrinol., 152, 159-166.

Davis, W.P. and Bortone, S.A., (1992), Effects of kraft mill effluent on the sexuality of fishes: an environmental early warning, in Chemically-Induced Alterations in Sexual Development: The Wildlife/Human Connection, eds Colborn, T. and Clement, C., volume XXI of Advances in Modern Toxicology, series ed Mehlman, M.A., Princeton Scientific Publishing Co., Inc., New Jersey, pp 113-127.

Davis, D.L. and Bradlow, H.L., (1995), Can environmental estrogens cause breast cancer? Sci. Am., October 1995, 144-149.

Dewailly, E., Dodin, S., Verreault, R., Ayotte, P., Sauve, L., Morin, J. & Brisson, J., (1994), High organochlorine body burden in women with estrogen receptor-positive breast cancer, J. Natl. Cancer Inst., 86(3), 232-233.

ENDS Report, (1996a), Phthalates shock for food industry, 255, 5-6.

ENDS Report, (1996b), WWF calls for ban on "oestrogenic" pesticides, 259, 13-14.

ENDS Report, (1996c), Sweden pushes for phase-out of oestrogen in pesticides, 260, 40.

ENDS Report, (1997), Wool scourers phase out APEs; and Chemical company stalls on oestrogen discharges, 266: 9-10.

Fisch, H. and Goluboff, E.T., (1996), Geographic variations in sperm counts: a potential cause of bias in studies of semen quality, Fertil. Steril., 65, 1044-1046.

Fisch, H., Goluboff, E.T., Olson, J.H., Feldshuh, J., Broder, S.J. & Barad, D.H., (1996), Semen analysis in 1 283 men from the United States over a 25-year period: no decline in quality, Fertil. Steril., 65, 1009-1014.

Folmar, L.C., Denslow, N.D., Rao, V., Chow, M., Crain, D.A., Enblom, J., Marcino, J. & Guillette, L.J., (1996), Vitellogenin induction and reduced serum testosterone concentrations in feral male carp (Cyprinus carpio) captured near a major metropolitan sewage treatment plant, Environ. Health Perspect., 104, 1096-1101.

Fox, G.A., (1992), Epidemiological and pathobiological evidence of contaminant-induced alterations in sexual development in free-living wildlife, inChemically-Induced Alterations in Sexual Development: The Wildlife/Human Connection, eds Colborn, T. and Clement, C., volume XXI of Advances in Modern Toxicology, series ed Mehlman, M.A., Princeton Scientific Publishing Co., Inc., New Jersey, pp 147-158.

Fry, D.M. and Toone, C.K., (1981), DDT-induced feminization of gull embryos, Science, 213, 922-924.

Garcia-Rodriguez, J., Garcia-Martin, M., Nogueras-Ocana, M., Luna-del-Castillo, J. de D., Garcia, M.E., Olea, N. & Lardelli-Claret, P., (1996), Exposure to pesticides and cryptorchidism: Geographical evidence of a possible association, Environ. Health Perspect., 104, 1090-1095.

Giwercman, A. (1995), Declining semen quality and increasing incidence of abnormalities in male reproductive organs - fact or fiction?, Hum. Reprod. Suppl., 10, 158-164.

Gray Jr., L.E., Ostby, J.S. & Kelce, W.R., (1994), Developmental effects of an environmental antiandrogen: The fungicide vinclozolin alters sex differentiation of the male rat, Toxicol. Appl. Pharmacol., 129, 46-52.

Grodsky, G.M., (1979a), Chemistry & function of the hormones: I. thyroid, pancreas, adrenal, & gastrointestinal tract, in Review of Physiological Chemistry, 17th. edition, Harper, H.A., Rodwell, V.W., Mayes, P.A., Lange Medical Publications, California, pp 511-555.

Grodsky, G.M., (1979b), Chemistry & function of the hormones: II. pituitary & hypothalamus, in Review of Physiological Chemistry, 17th. edition, Harper, H.A., Rodwell, V.W., Mayes, P.A., Lange Medical Publications, California, pp 556-568.

Grodsky, G.M., (1979c), General characteristics of hormones, in Review of Physiological Chemistry, 17th. edition, Harper, H.A., Rodwell, V.W., Mayes, P.A., Lange Medical Publications, California, 505-510.

Guillette Jr., L.J., Gross, T.S., Masson, G.R., Matter, J.M., Percival, H.F. & Woodward, A.R., (1994), Developmental abnormalities of the gonad and abnormal sex hormone concentrations in juvenile alligators from contaminated and control lakes in Florida, Environ. Health Perspect., 102, 680-688.

Guillette Jr., L.J., Pickford, D.B., Crain, D.A., Rooney, A.A. & Percival, H.F., (1996), Reduction in penis size and plasma testosterone concentrations in juvenile alligators living in a contaminated environment, Gen. Comp. Endocrinol., 101, 32-42.

Hadley, M.E., (1992), Endocrinology, third edition, Prentice-Hall International, Inc., New Jersey, USA.

Hall, J., (1996), The saga of the vanishing sperm, Biologist, 43(5), 223-226.

Hunt, G.L. and Hunt, M.W., (1977), Female-female pairing in western gulls (Larus occidentalis) in Southern California, Science, 196, 1466-1467.

Jefferies, D.J., (1967), The delay in ovulation produced by p,p'-DDT and its possible significance in the field, Ibis, 109, 266-272.

Jobling, S., Reynolds, T., White, R., Parker, M.G. & Sumpter, J.P., (1995), A variety of environmentally persistent chemicals, including some phthalate plasticizers, are weakly estrogenic, Environ. Health Perspect., 103, 582-587.

Jobling, S. and Sumpter, J.P., (1993), Detergent components in sewage effluent are weakly oestrogenic to fish: An in vitro study using rainbow trout (Oncorhynchus mykiss) hepatocytes, Aquatic Toxicology, 27, 361-372.

Joffe, M., (1996), Decreased fertility in Britain compared with Finland, Lancet, 347, 1519-1522.

Jordan, V.C., Mittal, S., Gosden, B., Koch, R. & Lieberman, M.E., (1985), Structure-activity relationships of estrogens, Environ. Health Perspect., 61, 97-110.

Kaiser, J., (1996), Endocrine disrupters, scientists angle for answers, Science, 274, 1837-1838.

Kang, K.S., Wilson, M.R., Hayashi, T., Chang, C.C. & Trosko, J.E., (1996), Inhibition of gap junctional intercellular communication in normal human breast epithelial cells after treatment with pesticides, PCBs, and PBBs, alone or in mixtures, Environ. Health Perspect., 104, 192-200.

Kelce, W.R., Stone, C.R., Laws, S.C., Gray, L.E., Kemppainen, J.A. & Wilson, E.M., (1995), Persistent DDT metabolite p,p'-DDE is a potent androgen receptor antagonist, Nature, 375, 581-585.

Klotz, D.M., Beckman, B.S., Hill, S.M., McLachlan, J.A., Walters, M.R. & Arnold, S.F., (1996), Identification of environmental chemicals with estrogenic activity using a combination of in vitro assays, Environ. Health Perspect., 104, 1084-1089.

Kolaja, G.J. and Hinton, D.E., (1977), Effects of DDT on eggshell quality and calcium adenosine triphosphatase, J. Toxicol. Environ. Health, 3, 699-704.

Kubiak, T.J., Harris, H.J., Smith, L.M., Schwartz, T.R., Stalling, D.L., Trick, J.A., Sileo, L., Docherty, D.E. & Erdman, T.C., (1989), Microcontaminants and reproductive impairment of the Forster's tern on Green Bay, Lake Michigan - 1983, Arch. Environ. Contam. Toxicol., 18, 706-727.

Laycock, J. and Wise, P., (1996), Essential Endocrinology, third edition, Oxford University Press, Oxford, New York, Tokyo.

Li, M.H. and Hansen, L.G., (1996), Enzyme induction and acute endocrine effects in prepubertal female rats receiving environmental PCB/PCDF/PCDD mixtures, Environ. Health Perspect., 104, 712-722.

Lye, C.M., Frid, C.L.J., Gill, M.E. & McCormick, D., (1997), Abnormalities in the reproductive health of flounder Platichthys flesus exposed to effluent from a sewage treatment works, Marine Pollution Bulletin, 1, 34-41.

McKinney, J.D. and Waller, C.L., (1994), Polychlorinated biphenyls as hormonally active structural analogues, Environ. Health Perspect., 102, 290-297.

McLachlan, J.A., Arnold, S.F., Klotz, D.M., Collins, B.M., Vonier, P.M. & Guillette, L.J., (1997), Science, 275, 405-406.

Nimrod, A.C. and Benson, W.H., (1996), Environmental estrogenic effects of alkylphenol ethoxylates, Crit. Rev. Toxicol., 26(3), 335-364.

Olea, N., Pulgar, R., Perez, P., Olea-Serrano, F., Rivas, A., Novillo-Fertrell, A., Pedraza, V., Soto, A.M. & Sonnenschein, C., (1996), Estrogenicity of resin-based composites and sealants used in dentistry, Environ. Health Perspect., 104, 298-305.

Pajarinen, J., Laippala, P, Penttila, A. & Karhunen, P.J., (1997), Incidence of disorders of spermatogenesis in middle-aged Finnish men, 1981-91: two necropsy series, BMJ, 314, 13-18.

Palmer, B.D. and Palmer, S.K., (1995), Vitellogenin induction by xenobiotic estrogens in the red-eared turtle and African clawed frog, Environ. Health Perspect., Ab., 103 (Suppl 4), 19-25.

Paulsen, C.A., Berman, N.G. & Wang, C., (1996), Data from men in greater Seattle area reveals no downward trend in semen quality: further evidence that deterioration of semen quality is not geographically uniform, Fertil. Steril., 65, 1015-1020.

Peakall, D.B., (1967), Pesticide-induced enzyme breakdown of steroids in birds, Nature, 216, 505-506.

Pirie, D., Steven, L., McGregory, S. & Best, G., (1996), Survey of Hormone Disrupting Chemicals, Scottish Environment Protection Agency, Stirling.

Porterfield, S.P., (1997), Endocrine Physiology, Mosby-Year Book, Inc., St. Louis, Missouri.

Ramamoorthy, K., Wang, F., Chen, I-C., Safe, S., McDonnell, D.P., Gaido, K.W., Bocchinfuso, W.P. & Korach, K.S., (1997), Science, 275, 405.

Ratcliffe, D.A., (1967), Decrease in eggshell weight in certain birds of prey, Nature, 215, 208-210.

Reijnders, P.J.H., (1986), Reproductive failure in common seals feeding on fish from polluted coastal waters, Nature, 324, 456-457.

Rier, S.E., Martin, D.C., Bowman, R.E. & Becker, J.L., (1995), Immunoresponsiveness in endometriosis: implications of estrogenic toxicants, Ab., Environ. Health Perspect., 103 (suppl. 7), 151-156.

Risebrough, R.W., Rieche, P., Peakall, D.B., Herman, S.G. & Kirvan, M.N., (1968), Polychlorinated biphenyls in the global ecosystem, Nature, 220, 1098-1102.

Safe, S.H., (1995), Environmental and dietary estrogens and human health: is there a problem?, Environ. Health Perspect., 103, 346-351.

Safe, S. and Krishnan, V., (1995), Chlorinated hydrocarbons: estrogens and antiestrogens, Toxicol. Lett., 82/83, 731-736.

Sharpe, R.M., Fisher, J.S., Millar, M.M., Jobling, S. & Sumpter, J.P., (1995), Gestational and lactational exposure of rats to xenoestrogens results in reduced testicular size and sperm production, Environ. Health Perspect., 103, 1136-1143.

Sharpe, R.M. and Skakkebaek, N.E., (1993), Are oestrogens involved in falling sperm counts and disorders of the male reproductive tract?, Lancet, 341, 1392-1395.

Shelby, M.D., Newbold, R.R., Tully, D.B., Chae, K. & Davis, V.L., (1996), Environ. Health Perspect., 104, 1296-1300.

Sommerville, B.A., Scanes, C.G., Swaminathan, R., Care, A.D., Harvey, S. & Chadwick, A., (1989), Effect of estrogen on calcium homeostasis and pituitary hormones in the growing chick, Gen. Comp. Endocrinol., 76, 261-266.

Soto, A.M., Chung, K.L. & Sonnenschein, C., (1994), The pesticides Endosulfan, Toxaphene and Dieldrin have estrogenic effects on human estrogen-sensitive cells, Environ. Health Perspect., 102, 380-383.

Soto, A.M., Lin, T.M., Justicia, H., Silvia, R.M. & Sonnenschein, C., (1992), An "in culture" bioassay to assess the estrogenicity of xenobiotics (E-screen), in Chemically-Induced Alterations in Sexual Development: The Wildlife/Human Connection, eds Colborn, T. and Clement, C., volume XXI of Advances in Modern Toxicology, series ed Mehlman, M.A., Princeton Scientific Publishing Co., Inc., New Jersey, pp 295-309.

Souminen, J. and Vierula, M., (1993), Semen quality of Finnish men, BMJ, 306, 1579.

Sturkie, P.D., (1954), Avian Physiology, Comstock Publishing Associates, New York.

Sumpter, J.P., (1995), Feminized responses in fish to environmental estrogens, Toxicol. Lett., 82/83, 737-742.

Thomas, K.B. and Colborn, T. (1992), Organochlorine endocrine disruptors in human tissue, in Chemically-Induced Alterations in Sexual Development: The Wildlife/Human Connection, eds Colborn, T. and Clement, C., volume XXI of Advances in Modern Toxicology, series ed Mehlman, M.A., Princeton Scientific Publishing Co., Inc., New Jersey, pp 365-394.

Tyler, C. (1995), The emasculators, BBC Wildlife, 13(6), June 1995, 74-76.

United Nations Environment Programme, (1989), Environmental Data Report, second edition 1989/1990, Blackwell Reference, Oxford and Cambridge, Massachusetts.

vom Saal, F.S., Nagel, S.C., Palanza, P., Boechler, M., Parmigiani, S. & Welshons, W.V., (1995), Estrogenic pesticides: binding relative to estradiol in MCF-7 cells and effects of exposure during fetal life on subsequent territorial behaviour in male mice, Toxicol. Lett., 77, 343-350.

Vonier, P.M., Crain, D.A., McLachlan, J.A., Guillette, Jr., L.J. & Arnold, S.F., (1996), Interaction of environmental chemicals with the estrogen and progesterone receptors from the oviduct of the American alligator, Environ. Health Perspect., 104, 1318-1322.

Wahli, W., Dawid, I.B., Ryffel, G.U. & Weber, R., (1981), Vitellogenesis and the vitellogenin gene family, Science, 212, 298-304.

Wessells, N.K. and Hopson, J.L., (1988), Biology, Random House, Inc., New York, pp 900-927.

Wolff, M.S., Toniolo, P.G., Lee, E.W., Rivera, M. & Dubin, N., (1993), Blood levels of organochlorine residues and risk of breast cancer, J. Natl. Cancer Inst., 85(8), 648-652.

BIBLIOGRAPHY

Colborn, T., Myers, J.P. & Dumanoski, D. (1996), Our Stolen Future, Little, Brown and Company, Boston, New York, Toronto, London.