Male fertility is largely controlled by the hypothalamic-pituitary axis, a careful balance between stimulating and suppressing gene expression and the secretion of hormones. The critical factors for male fertility have in the past been thought to be limited to testosterone and the gonadotropins. Estrogen has only recently been demonstrated to be both a crucial requirement for fertility and a cause of infertility. Reports in the early 1990s demonstrated a decrease in mean sperm counts over the last 50 years. A hypothesis for this observation is the increase of xenoestrogens in the environment that are able to mimic and potential disrupt the natural estrogens involvement in fertility. Although the mechanisms of estrogens involvement are not yet defined, the Sertoli cells are a potential sites of action as they possess receptors for the hormone and are able to locally produce it. Sertoli cells both act to protect and provide for the male germ cells and the developing spermatozoa. Pyrethroids are common synthetic insecticides of which some have previously shown estrogenic activity. Therefore this investigation examined the effects of pyrethoids, whose estrogenicity was confirmed via the yeast assay, on the estrogen receptor expression in mouse Sertoli cells as a model for general effects of estrogenic chemicals on male fertility. The results first confirmed the estrogenicity of some pyrethroids and these pyrethroids when exposed to mouse Sertoli cells effected estrogen receptor mRNA expression however in a different way to the natural ligand 17β-estradiol.

This thesis aims to assess the human health impact of exposure to estrogenic compounds from the diet. A multi-disciplinary approach is taken to address various aspects of this issue. An introduction to xenoestrogens, including international research priorities, wildlife and human health effects, mechanisms of action, structure activity relationships and additivity of estrogenic effects is provided as background information. An assessment of exposure to a range of naturally occurring and synthetic estrogenic compounds found in food is derived in Chapter 2. The assessment combines new and existing data on food concentration, food consumption and serum levels for each xenoestrogen. Exposure is combined with relative estrogenic potency data from published bioassasy data to estimate risk relative to normal circulating levels of estradiol. Assuming additivity of xenoestrogens, for an average New Zealand male and for post-menopausal women, xenoestrogens in the diet contribute an additional 12-90% of estrogenicity above normal circulating levels. For a pre-menopausal female, the contribution from the diet represents in the order of an additional 2%. The level of exposure determined in this thesis would seem to be of pharmacological relevance, especially for men with low levels of estrogen and for post-menopausal women. Bisphenol A (BPA) is an important monomer used in the manufacture of epoxy resins for internal food can linings. A survey of the BPA content of a range of 80 canned foods available to the New Zealand consumer was undertaken and the results used in the exposure and risk assessments. BPA was detected in all foods analysed except soft drinks, at concentrations ranging from <10-29 µg/kg, except for individual samples of tuna, corned beef and coconut cream that were 109, 98 and 191 µg/kg respectively. None, of over 4000 individual exposure scenarios, exceeded the temporary Tolerable Daily Intake (TDI) of 10 µg/kg body weight per day set by the Scientific Committee on Food in 2002. Intestinal microflora influence the bioavailability of the naturally occurring xenoestrogens genistein and daidzein that contribute significantly to total estrogenicity from the diet. The degradation of genistein and daidzein by the faecal microfloral of 5 human subjects was variable and unpredictable between individuals and within an individual. These findings have important implications for the promotion and prescription of soy foods and supplements for disease prevention and health benefits. The "yeast assay" is one of a number of methods available to measure estrogenicity. This assay was established and validated. In utero exposure to estrogenic compounds at critical periods of sexual differentiation and endocrine development may imprint for health effects observed later in life. Placental transfer of estrogenicity, from 17β-estradiol was studied using the human placental perfusion model and the yeast assay. The placenta provides a protective barrier to the transfer of estrogenicity. Experiments with genistein showed that 5-15% placental transfer occurred, suggesting that in utero exposure might be in the order of 10% of maternal exposure. The thesis concludes with consideration of a genomic approach to substantiate, or refute, the mechanistic link between exposure to xenoestrogens and claimed human health effect. Such an approach offers exciting opportunity to clarify the mode of action of the synthetic versus the naturally occurring xenoestrogens, to confirm or dispute additivity of effect that is an important premise of the exposure assessment, to identify key genes involved in the many possible health effects and thence risk to the individual from dietary exposure to xenoestrogens.

Male fertility is largely controlled by the hypothalamic-pituitary axis, a careful balance between stimulating and suppressing gene expression and the secretion of hormones. The critical factors for male fertility have in the past been thought to be limited to testosterone and the gonadotropins. Estrogen has only recently been demonstrated to be both a crucial requirement for fertility and a cause of infertility. Reports in the early 1990s demonstrated a decrease in mean sperm counts over the last 50 years. A hypothesis for this observation is the increase of xenoestrogens in the environment that are able to mimic and potential disrupt the natural estrogens involvement in fertility. Although the mechanisms of estrogens involvement are not yet defined, the Sertoli cells are a potential sites of action as they possess receptors for the hormone and are able to locally produce it. Sertoli cells both act to protect and provide for the male germ cells and the developing spermatozoa. Pyrethroids are common synthetic insecticides of which some have previously shown estrogenic activity. Therefore this investigation examined the effects of pyrethoids, whose estrogenicity was confirmed via the yeast assay, on the estrogen receptor expression in mouse Sertoli cells as a model for general effects of estrogenic chemicals on male fertility. The results first confirmed the estrogenicity of some pyrethroids and these pyrethroids when exposed to mouse Sertoli cells effected estrogen receptor mRNA expression however in a different way to the natural ligand 17β-estradiol.

Xenoestrogens are compounds, foreign from the body, that can enter cells and interact with the estrogen receptors (ER) to produce an estrogenic response. Many additives used in plastics are compounds with estrogenic activity. Some of these additives are known to slowly leach from the plastics. When using plastic containers as lunchboxes for reheating or food storage, these additives can leach from the plastics and end up in the food. In this project, food simulates were cooked in six different thermoplastic containers, made of polypropylene, in an oven at 100 °C for 15 minutes. Three of the thermoplastic containers were lunchboxes marketed to be able to withstand cooking in a microwave. The other three were provisional lunchboxes made from various food storing containers originally made for refrigeration purposes. The estrogenic activity in the different samples was measured using an ER-CALUX in vitro assay. The results were measured in 17β-estradiol equivalent (Bio-EEQ) values in pg/ml. The purpose of this project was to investigate whether or not these plastic containers leach xenoestrogens that can be measured with an ER-CALUX assay, and compare the results with the results from other existing toxicological studies, and also to see if there is a difference in Bio-EEQ levels between the plastic containers made for microwave usage and those made for refrigerated purposes. The results from this project indicate that most of these plastic containers do leach estrogenic compounds that can be detected in the ER-CALUX, even the ones made for microwave usage. Fortunately, compared to other toxicological studies, the Bio-EEQ levels in these food samples cooked in plastic containers are low. However the potential adverse effects in prenatally exposed children cannot be ignored as other studies have shown that very low levels of xenoestrogens are enough to potentially cause a disturbance in the reproductive development and fertility.

Synthetic xenoestrogens have differential estrogenic properties. Research has shown that exposures to xenoestrogens could promote breast cancer by disrupting normal function of the human epidermal growth factor receptor 2 (Her2) gene. Although animal models demonstrated a connection between xenoestrogen exposure and Her2 activity, no study using human cells has systematically examined their carcinogenic potential influencing the Her2 gene expression. Furthermore, breast cancer cells are phenotypically disparate (ER+, Her2+), with some phenotypes (Her2+), leading to more aggressive disease. This study aimed to dosimetrically assess the carcinogenic potential of commonly used xenoestrogens influencing Her2 gene expression, and delineate cellular phenotypes at greater risk of more aggressive disease. The study assessed whether the composition, concentrations, and exposure duration of BPA, EE, NPH, and DDT significantly altered Her2 copy numbers in estrogen and Her2 receptor positive or negative breast cancer lines. Each line was randomly assigned to cases (exposed) and control (unexposed) groups using a randomized block design. Fluorescent in-situ hybridization measured Her2 gene copies. Mann Whitney, Kruskal Wallis, and Incidence Rate Ratios revealed Her2 copy gains in all 4 xenoestrogens and receptor types with persistent exposures. A 44% increase in Her2 was observed in the normal ER and Her2 line, marking a shift in its Her2 status, and a 30-times greater risk was noted in the Her2+ lines. These findings promote positive social change by revealing all 4 xenoestrogens as risk factors for breast cancer. This information can be used by breast cancer advocacy groups, health educators, and steering committees to educate women and formulating policies.

Vitellogenin in Pimephales promelas was used to assess estrogenicity of a local municipal effluent. Vitellogenin induction in male P. promelas increased in frequency and magnitude with increased exposure duration and was greater ("=0.05) than controls after 2 and 3 weeks of exposure. The level of vitellogenesis induced by effluent exposure was high compared to similar studies. A spring season evaluation followed. Biomarkers in P. promelas were used to assess the efficacy of a treatment wetland to remove toxicity and estrogenicity in final treated wastewater effluent. Comparisons were made with an effluent dominated stream and laboratory controls. Vitellogenin, GSIs (gonado-somatic indices), HSIs (hepato-somatic indices) and secondary sexual characteristics were biomarkers used in P. promelas models to assess aqueous estrogenicity. Biological indicators used to assess general fish health included hematocrit and condition factors. The estrogenic nature of the effluent was screened, concurrent with fish exposure, with GC/MS analysis for target estrogenic compounds including: 17-b estradiol, estrone, ethynylestradiol, Bisphenol A, nonylphenolic compounds, phthalates, and DDT. Plasma vitellogenin measured in P. promelas was significantly elevated (p < 0.0001) at the inflow site of the wetland and stream sites. GSIs for these exposures were less (a=0.001) at the wetland inflow site. At wetland sites closest to the inflow, secondary sexual charateristics, tubercle numbers and fat pad thickness, were less (a=0.0001). Hematocrit and condition factors were less (a=0.001) at sites closer to the wetland inflow. Seasonal variation was examined by repeating the effluent characterization in summer. Additionally, summer testing included exposure to an effluent dilution series. Fish condition heavily influenced interpretation of the results. Pre-acclimation exposure to spawning stresses may have altered many of the biological markers measured. Results are discussed relative to fish health and pre-exposure environment. Toxicity assessed with P. promelas biomarkers was compared with Ceriodaphnia dubia and Vibrio fischeri toxicty tests on this effluent. Biomarkers of fish health were somewhat less sensitive than C. dubia test endpoints, but more sensitive than V. fisheri.

Les polluants environnementaux, en particulier les perturbateurs endocriniens (PE), agissent à très faibles doses sur des cibles multiples. Les effets rapportés portent en majorité sur les organes de reproduction. Très peu d’études ont porté sur le squelette alors que le cartilage et l’os sont sous un puissant contrôle hormonal, depuis le stade fœtal où le système hormonal se met en place jusqu’au vieillissement, en passant par la naissance (hormones thyroïdiennes, hormone de croissance), la puberté et la ménopause chez la femme (stéroïdes sexuels). L’objectif de ce travail est d’étudier les effets de polluants anti-androgènes (vinclozoline, V et métabolite actif M2) ou xenestrogènes (génistéine, G; bisphénol A, BPA), in vivo sur le développement du squelette du rat Wistar et in vitro sur les marqueurs de différenciation chondrogéniques. Les effets in vivo ont été étudiés à des doses inférieures aux “No Observed Adverse Effect Levels ” (NOAEL) fixés par les instances européennes (EFSA) et internationales (US EPA). Des rattes ont été exposées à V, G seuls, combinés (GV) et/ ou associés au BPA (BGV), et ce de la conception des petits jusqu’à leur sevrage (J30) ou leur sacrifice (J30, J110). Les effets ont été recherchés sur des petits de mères et portées différentes, quatre pour chaque traitement, âge et sexe. Les effets in vitro du métabolite M2 de la Vinclozoline, associé ou non avec G et BPA, ont été étudiés sur l’expression de marqueurs chondrogéniques en utilisant : 1) un modèle murin de cellules souches inductibles vers la voie chondrogénique (C1) pour les effets sur la différenciation chondrogénique précoce et 2) des chondrocytes de souriceaux nouveau-nés, différenciés en culture primaire ou dédifférenciés (passages répétés). Comparaison avec les effets du bFGF, facteur de dédifférenciation chondrogénique. Résultats : In vivo, l’exposition à V, seule ou associée à G ou au BPA induit chez les rattes F1 exposées, une cannelure de la queue, discrète mais perceptible à la palpation en regard de chaque articulation intervertébrale. Les xénestrogènes tendent à réduire cet effet de V. Les rats et les animaux F2 ne sont pas atteints. L’examen par micro CT-scan montre une augmentation significative de la largeur des apophyses transverses (ITA) des vertèbres, et une diminution de la hauteur des corps vertébraux chez les rattes F1 exposées en regard des contrôles. Ces modifications anatomiques rappellent certaines pathologies génétiques des collagènes (dysplasies épiphysaires) chez l’homme Elles sont absentes chez les rats F1 et les animaux F2. Elles sont en partie transitoires car présentes à J30 (effets sur ITA et longueur) quand seul l’effet sur l’ITA perdure à J110. L’examen histologique des cartilages de croissance des corps vertébraux montre un déséquilibre entre les zones de prolifération et d’hypertrophie qui évoque une modification de la maturation du cartilage de type estrogénique. Ces effets sont ici aussi transitoires et majoritairement observés chez les rattes. L’effet plus prononcé du BPA lisse toutes les autres activités. Cela suggère que les PE pourraient moduler la différenciation du cartilage de croissance. C’est ce qui a été étudié in vitro. In vitro. Le premier objectif était d’évaluer les effets des PE sur la dynamique d’induction chondrogénique (cellules C1). Nous montrons que l’addition de M2 seul ou avec G ou BPA modifie le processus de maturation du collagène2 sans effet sur les autres marqueurs (SOX9, Agrécane, Col10). M2 prolonge l’expression de COL2A immature et retarde son remplacement par l’isoforme COL2B. Le second objectif était d’étudier les effets des PE sur la régulation de l’expression de COL2A au cours du processus de dé-différenciation des chondrocytes in vitro. L’expression de COL2A augmente avec le degré de dédifférenciation cellulaire (passages successifs) et double en présence de M2, G et BPA. Cet effet dépend des récepteurs aux estrogènes (ER) et des voies p38-MAPK. (...) / Environmental pollutants, particularly Endocrine Disruptors (ED), show effects on multiple targets at very low doses. Mostly known effects target reproductive organs. Very few studies are conducted on skeleton, although cartilage and bone are under potent hormonal control, from fetal stage, where hormonal system takes place, until aging, through birth stage (thyroid hormones, growth hormones), puberty and menopause for women (steroid hormones). The aim of this work is to study effects of anti-androgenic pollutants (vinclozolin, V, and its active metabolite M2) or xenoestrogens (genistein, G; bisphenol A, BPA), in vivo on Wistar rat skeletal development and in vitro on chondrogenic differenciation markers.In vivo effects were studied at doses below the “No Observed Adverse Effect Levels” (NOAEL) established by European and American agencies (EFSA and US EPA respectively). Female Wistar rats were exposed to V and G alone, in combination (GV) and/or associated to BPA (BGV), from pups conception until weaning (d30) or sacrifice (d30, d110). Effects were investigated on offsprings from different mothers and litters, on four animals by treatment, age and gender. In vitro effects of M2 metabolite of Vinclozolin, combined or not to G and BPA, were studied on chondrogenic markers expression using : 1) inducible murine stem-cells model towards chondrogenesis (C1) to sudy effects on early chondrogenic differentiation and 2) post-natal mouse differentiated chondrocytes, in primary culture or dedifferenciated chondrocytes by successive passages. Comparison with bFGF, a dedifferentiation factor for chondrocytes.Results : In vivo, exposure to V, alone or combined to G or BPA, induce discrete but palpable annealing in F1 treated female rat tails, in front of each intervertebral articulation. Xenoestrogens tend to decrease V effect. Male rats and F2 offsprings were not affected. Micro-CT Scan analysis shows significative increase of vertebrae inter transverse apophyses (ITA) distance, and decrease of vertebral body height in F1 female rats comparing to control animals. Anatomical modifications recall human collagen genetic diseases (epiphyseal dysplasias). They are absent in F1 male rats and F2 offsprings. Furthemore they are partly transient, ITA and height effects being present at d30 whereas ITA effect alone remains until d110. Histological analysis of vertebral body growth plate shows unbalance between proliferative and hypertrophic zones, which evokes estrogenic acceleration of cartilage maturation. Those effects are still transient and mainly observed in female rats. BPA activity is dominant above G and V effects. This result suggests ED can modulate growth plate cartilage differentiation, which was studied in vitro. In vitro : First, we aimed to evaluate eventual role of ED on the dynamic of the chondrogenic differentiation process. We show that M2 addition, alone or in combination with G or BPA, modifies collagen 2 maturation process without any effect on other markers (SOX9, Agrecan, COL10). M2 addition extends immature isoform COL2A expression and delays its replacement by mature isoform COL2B. Second, we studied the effects of ED on the regulation of COLA expression through the dedifferentiating process of chondrocytes in vitro. COL2A expression increases with cell dedifferenciation degree (successive passages) and double with M2, G and BPA. No other chondrogenic marker was modified. This effect depends on estrogen receptors (ER) and p38-MAPK pathway. (...)