Low sperm concentration, reduced motility and/or abnormal morphology of sperm are the dominant causes of subfertility in 20 – 25% of couples 181920. Male subfertility is generally expressed as a reduced ability of the female partner to become pregnant.

Problems with ovulation account for subfertility in another 20 – 25% of couples and is thereby a frequent cause of subfertility in women 18. Ovulation problems present themselves as irregular or absent menstrual periods and can be substantiated through measurement of reproductive hormones.

Abnormal cervical mucus may prevent the sperm from reaching the oocyte. Some authors estimate that cervical hostility is a cause of subfertility in 10 – 15% of couples 1819, whereas others deny its mere existence 20.

Tubal damage and/or obstruction, hydrosalpinx, pelvic adhesions, and endometriosis are the main cause of subfertility in 10 – 30% of couples 18. In many instances, these problems originate from infections.

Despite advances in the diagnosis of causes of subfertility, inability to conceive remains unexplained in 25 – 30% of fully investigated couples 181920.

An important factor in female subfertility is age. The risk of subfertility increases from 10% to 30% when women are over 35 years of age. This is of particular importance nowadays, as an increasing number of women delay pregnancy until the age when natural female fertility is in decline, due to a higher number of chromosomal aberrations in the oocytes. Hormonal balance is another important factor in female fertility, in particular regarding the ovarian cycle. Lifestyle factors, including stress, extreme body weight (too low or too high), coffee consumption, diet, and excessive exercise can affect a woman's hormonal balance and subsequent ovulatory pattern. Hormonal imbalance and ovulatory problems are much less often caused by hormonal diseases, such as pituitary gland tumors. But there are indications that endocrine disrupting chemicals, such as PCBs and certain pesticides, can influence the hormonal balance and thus increase the risk of subfertility 21. In this review, we will elaborate on hormonal function disturbances associated with pesticide exposure.

All hormones differ in their chemical structure and have a different route of synthesis with innumerable different steps. If one substance or link is disturbed in the chain of hormone synthesis, the hormone may not be produced or may get different properties. Some pesticides, such as fenarimol, prochloraz, and other imidazole fungicides possess the ability to inhibit estrogen biosynthesis through CYP19 aromatase inhibition in vitro 262728, preventing the conversion of androgens to estrogens. Vinggaard et al. hypothesized that compounds which can inhibit aromatase acitivity in vitro may be able to cause local changes in estrogen and androgen concentrations in vivo 26. Aromatase induction is a physiological mechanism to deactivate xenobiotics, which does not inevitably cause a toxic effect. The pesticides atrazine, simazine, and propazine (2-chloro-triazine herbicides) induce aromatase activity in vitro 29. For p,p-DDE, the induction of aromatase has been demonstrated in vitro and in vivo 30. In addition, the pesticides methomyl, pirimicarb, propamocarb, and iprodion can weakly stimulate aromatase activity 31, whereas heptachlor may act as an inducer of testosterone 16-alpha and 16-beta hydroxylases 32.

Thiram, Sodium N-methyldithiocarbamate (SMD), and other dithiocarbamates are known to suppress the dopamine-beta-hydroxylase activity leading to reduced conversion of dopamine to norepinephrine 333435. This may lead to changes in hypothalamic catecholamine activity involved in generating the proestrus surge in LH, which stimulates the final stages of ovulation. Goldman et al. concluded that SMD is able to block the LH surge and ovulation in rats 33. Ketaconazole inhibits various enzymes which belong to the CYP450-dependent monooxygenases and also inhibits progesterone synthesis 2836.

Interference with hormone storage and/or release is also mentioned in the definition of EDCs as a mechanism of action. Catecholamine hormones (e.g. norepinephrine) are stored in granular vesicles of chromomaffin cells within the adrenal medulla and within presynaptic terminals in the central nervous system. Therefore, they can be released quickly on demand. In contrast, steroid hormones are not stored intracellularly within secretory granules, but are readily synthesized after gonadotropin stimulation of the gonads.

The formamidine pesticides chlordimeform and amitraz have been reported to block norepinephrine binding to the alpha 2-andrenoreceptors 37. Norepinephrine is critical for the preovulatory increase in the pulsatile release of GnRH and the subsequent ovulatory surge of LH 38. Thiram suppresses the proestrus surge of LH and delays ovulation in the female rat 39. Disruption in the timing of the LH surge could alter the viability and the quality of the oocyte 40 and a potential conceptus by pre-ovulatory over-ripeness ovopathy (PrOO) 41. Inhibition of progesterone secretion and poor conception occurred after malathion exposure at the onset of estrus in cattle 42.

For the most part, steroid hormones in the bloodstream do not float around freely, but are bound to carrier proteins, such as SHBG and albumin. Because only free hormones can be biologically active, increases or decreases in the concentration of SHBG will have a major impact on the available and active steroid hormone concentrations in blood. Estrogens are known to increase the synthesis of SHBG in the liver and thus increase the SHBG concentration in plasma, whereas androgens decrease these concentrations 2543. Substances that mimic these natural hormones may cause similar changes, but no specific papers dealing with effects of pesticides on SHBG levels have been found.

In contrast, reports are known about the influence of pesticides on clearance of steroid hormones, mostly occurring in the liver. The clearance rate is different for each hormone and is influenced by compounds that alter liver enzyme activity involved in hormone clearance. Many pesticides induce the liver enzymes monooxygenase and UDP-glucuronosyltransferase 44, resulting in increased clearance of the pesticide itself for detoxification purposes, but also of testosterone 4546. For instance, DDT analogs are potent inducers of hepatic microsomal monooxygenase activity in vivo 47, which degrades endogenous androgens, resulting in suppressed androgen receptor mediated activity. These effects have also been suggested for endosulfan and mirex 48. Similarly, treatment with lindane has been reported to increase the clearance of estrogens 49.

This mechanism of endocrine disruption is much discussed in the literature. Hormones travel from their point of release in the bloodstream to particular tissues where they convey their messages. For the message to be interpreted, hormones bind to receptors. Hormone and receptor have a precise fit, so that only a specific type of hormone can bind to a specific receptor (see Figure 2) 23. A number of environmental agents may alter this process by mimicking the natural hormone (agonists) or by inhibiting receptor binding (antagonists). The latter mechanism is based on complete or partial blocking of the specific receptor. Regarding the estrogen receptor, this mechanism only applies when the endocrine disruptor concentration is high, because the affinity of endocrine disruptors for the estrogen receptor is usually many times lower than that of 17-beta-estradiol. Three different mechanisms are elaborated below: 1. binding and activating the estrogen receptor; 2. binding without activating the estrogen receptor; and 3. binding other receptors.

When an endocrine disruptor or one of its metabolites bind and activate the estrogen receptor, the endocrine disruptor will imitate the hormone 17β-estradiol. The substance thus acts as an agonist and is called estrogenic. Possible effects may be decreased production of GnRH by the hypothalamus (negative feedback system) and of LH and FSH by the pituitary gland. As a result, the levels of LH and FSH will drop and finally lead to a lack of estradiol. Under normal circumstances, the hypothalamus will then be triggered to produce more GnRH, but this will be prevented by the endocrine disruptor. As a result, the hormonal cycle may be disrupted.

Several pesticides or their metabolites have been reported to possess estrogenicity in vivo, e.g. methoxychlor 50515253, kepone 54, and DDT 555657 and/or in vitro, e.g. o,p'-DDT and p,p'-DDT 57, endosulfan 315859, toxaphene 585960, dieldrin 31596061, fenarimol 3162, triadimefon 62, triadimenol 62, aldrin 55, endrin 55, methiocarb 31, pentachrophenol 63, nonylphenol 63, alachlor 64, fenvalerate 65, chlordecone 66, and sumithrin 6566. Some pesticides are weakly estrogenic, but may act additively in combination of two or more pesticides. When mixed together they may induce estrogenic responses at concentrations lower than those required when each compound is administered alone 60.

If an endocrine disruptor or its metabolites bind, but do not activate the estrogen receptor, the substance acts as an antagonist and inactivates the estrogen receptor, preventing estradiol to bind. As a result, the hypothalamus stimulates production of GnRH and the pituitary gland will produce more LH and FSH. The concentration of estradiol will also increase, but because the feedback mechanism is disrupted, GnRH will not decrease. Eventually, the estrogen receptors may become less sensitive through prolonged exposure to high concentrations of estrogen and/or endocrine disruptors.

The results of a study by Cooper et al. indicate that lindane may effectively block the response of estrogen-dependent tissues and that this apparent anti-estrogenic effect is responsible for the disturbances observed in the neuroendocrine control of ovarian function in rats 67. Other studies also suggest that lindane is anti-estrogenic and is able to disrupt the estrus cycle 6869. Atrazine, simazine, and diaminochlorotriazine expressed anti-estrogenic activity in uteri of female rats without expressing intrinsic estrogenic activity, but the precise mechanism is not known 7071.

The fungicide vinclozolin and two of its metabolites bind the androgen receptor and act as androgen receptor antagonists in vitro and in vivo 72737475. Procymidone and DDT also express anti-androgenic activity 7476. A paper by Kelce et al. presents consistent evidence that DDT and DDE compete with androgens for their receptors 66. In addition, several studies suggest that the pesticides methoxychlor 74, linuron 7477, fenitrothion, and biphenol act as androgen antagonists in vitro and/or in vivo 74787980. The presence of a potent androgen antagonist in a sufficient internal dose may create an overall estrogenic effect. The pyrethroid insecticides fenvalerate and d-trans allethrin seem to antagonize the action of progesterone 65.

Some endocrine disruptors act through one of the above mechanisms, while others may cause their effects in several different ways. For instance, DDT was found to be estrogenic, but can also bind to the androgen receptor 63. The pesticides dieldrin, endosulfan, methiocarb, and fenarimol are known as both estrogen agonists and androgen antagonist 31. Prochloraz acts as estrogen and androgen antagonist 31. Endosulfan and alachlor have been found to bind to the estrogen receptor as well as the progesterone receptor 81. Kelce et al. showed that chlordecone binds rather efficiently to estrogen receptors while it may also bind to androgen receptors at higher concentrations 66. Although the potencies of these pesticides to act as hormone agonist or antagonists are low compared to the natural ligands, the integrated response in the organism might be amplified by the ability of pesticides to act via several mechanisms and by frequent simultaneous exposure to different pesticides 31.

If an agonist binds to its receptor, a cascade of events is initiated for the appropriate cellular response necessary for signal transduction across the membrane or, in case of nuclear receptors, the initiation of or alteration in DNA-transcription and protein synthesis 25. Lindane has been demonstrated to decrease phosphatidylinositol turnover in the membrane and to reduce protein kinase-C activation 82. Steroid hormone receptor activation may also be modified by indirect mechanisms such as a downregulation, which is seen after TCDD exposure 8384.

Pesticides like chlorophenols, chlorophenoxy acids, organochlorines, and quinones have been shown to alter thyroid gland function and to reduce circulating thyroid hormone levels 8586. Reduction in thyroid hormone levels can compromise the catalytic activity of hepatic cytochrome P450 monooxygenases, resulting in an altered hepatic androgen metabolism 87.

The central nervous system (CNS) is very important in the integration of hormonal and behavioral activity. Disturbances in these finely tuned mechanisms can severely impair normal adaptive behavior and reproduction. Since many pesticides are known to be neurotoxic, it is conceivable that these chemicals can disrupt the coordinating activity of the CNS by disrupting brain cell functions 25. Also, pesticides can alter the hypothalamic and pituitary function and thus secretion of GnRH, LH, and FSH in a more direct manner by modifying the feedback of endogenous hormones. For example, it has been demonstrated that low-dose exposure to o,p-DDT and methoxychlor can result in diminished hypothalamic and pituitary function in rodents 8889. Finally, it is postulated that any environmental compound mimicking or antagonizing steroid hormone action could presumably alter the glycosylation of LH and FSH, thereby reducing their biological activity 90.

Hormonal balance, i.e. a proper level of sexual hormones, is important to preserve female reproduction and maintain fertility. This balance can be disturbed by changing levels of estrogen or progesterone. Estrogen levels may be decreased by several pesticides. Treatment of rats with the insecticide heptachlor suppressed estradiol concentrations in blood and reduced the production of estradiol by ovarian cells of treated rats 9293. Lindane, atrazine, and simazine also cause a decrease in circulating estradiol levels in rats 9495. In monkeys, ovulatory levels of estradiol were reduced after high doses of hexachlorobenzene 96, which also induced anovulatory cycles and suppression of circulating levels of estradiol 97, and a dose dependent suppression of serum progesterone concentrations during the luteal phase 98. Progesterone levels may be decreased by exposure to methoxychlor as well, especially during the estrus phase of the estrus cycle in rats 99100. During early pregnancy, progesterone concentrations decreased after treatment with DDT in rabbits 101.

The female ovarian cycle is the result of a balanced cooperation between several organs and is determined by a complex interaction of hormones. Ovarian cycle irregularities include disturbances in the ovarian cycle (e.g. longer cycle, persistent estrus) and ovulation problems (deferred ovulation or anovulation).

Two studies examined the effects of pesticide exposure on the menstrual cycle. Both found associations between serum levels of DDT or a metabolite of DDT and short cycles 130 and undefined 'menstrual disturbances' 131. A recent study observed that women who currently used pesticides experienced longer menstrual cycles and increased odds of missed periods compared with women who never used pesticides 10. In addition, women who used probably hormonally active pesticides had a 60–100% increased odds of experiencing long cycles, missed periods, and intermenstrual bleeding compared with women who had never used pesticides.

In a study in the USA, infertile women were observed to be three times more likely to ever having been exposed to pesticides 14 and nine times more likely to ever having worked in agriculture 13. Another study found no correlations between infertility and self-reported overall pesticide exposure, working in the agricultural sector, or living on a farm during the two years before the diagnosis of infertility or the last pregnancy 132. However, an association was present when exposure to herbicides only was considered.

Three studies examined the effects of pesticide exposure on the time it took couples to become pregnant [time-to-pregnancy (TTP)], which is affected by disturbances in the whole chain from gametogenesis to embryonic survival 133134. No consistent pattern of associations was observed by Curtis et al. in Canada, but some specific pesticides were tentatively associated with prolonged TTP 135. Abell et al. and Idrove et al. found significantly prolonged TTPs related to high levels of pesticide exposure among female worker in flower greenhouses 1112.

A number of studies reported that among women occupationally exposed to pesticides and/or working in the agricultural sector the risks of spontaneous abortion 136137138139 and stillbirth 140141142143144 seemed to be significantly increased. In addition, two reviews concluded that there are numerous indications that exposure to pesticides may contribute to spontaneous abortion and/or stillbirth 145, but it is unclear whether this should be considered as an endocrine disrupting effect 146.

For birth defects, an overall association with agricultural work was observed in a large cohort study 147. Studies focusing on specific birth defects found associations between agricultural work and orofacial clefts 138, hypospadias 148, total anomalous venous return 149, spina bifida 148150, and limb reduction defects 148151, although the relation with limb reduction defects was contradicted by one other study 152. In a well-conducted Finnish study of women in agricultural occupations, the investigators found that exposure to pesticides during the first trimester of pregnancy nearly doubled the risk of cleft lips and palates in offspring 153. A slightly increased risk for central nervous system defects was also observed. Again, a cause-effect relation between these defects and exposure to endocrine disrupting pesticides could not be established.