Polycystic ovary syndrome (PCOS) is a heterogeneous disorder encompassing reproductive and metabolic phenotypes. Genetic analysis, targeting candidate genes has to date proven unsuccessful in the search for a truly dominant genetic link. Another hypothesis to explain the etiology of PCOS is that of fetal programming in the context of developmental origins of health and disease. Extensive animal studies, validated by human data, support the fetal origins hypothesis of PCOS and highlight that PCOS may arise due to excess androgen exposure in fetal life. Previous reports from our laboratory found metabolic dysfunction in 11 month old prenatally androgenised females (d62-102 of fetal life), which included pancreatic and hepatic alterations. The pancreatic alterations seemed to result from gene expression changes induced in fetal life. Therefore, chapter 3 focuses on the gluconeogenic response in the day 90 fetus following maternal androgenisation from day 62 of gestation. Interestingly hepatic gluconeogenic enzymes, specifically phosphoenolpyruvate caboxykinase (PEPCK) and glucose 6 phosphatase (G6PC), were not altered. However they were decreased in the kidney, in a sex specific manner with PEPCK significantly decreased (P<0.01) and G6PC showing a strong trend toward reduction (P=0.056) in females only. This chapter progresses to explore regulatory pathways involved in gluconeogenic regulation. It seems probable that the female specific increase in circulating testosterone (P<0.001), with increased renal androgen reception (P<0.01), may be accountable for the altered expression of gluconeogenic enzymes in the kidney. Chapter 4 investigates why testosterone concentrations were not increased in the male fetus, after maternal androgenisation, by focusing on the site of testosterone production, the fetal testis. Results demonstrate that the day 90 fetus is capable of responding to prenatal androgenisation by decreasing luteinising hormone (P<0.01) and thus testicular testosterone production, such that there was a global down regulation in steroidogenic enzyme expression, in vivo testosterone production (P<0.001) and Leydig cell morphology was altered (P<0.001). As prenatal androgenisation is administered through the maternal route and placental aromatisation may occur, a novel method whereby the fetus was directly injected was utilised to assess the effects of control oil (C), testosterone (TP) or diethylstilboestrol (DES) on the fetal testis. Unlike DES, direct fetal injection with TP mimics the results found from maternal androgenisation. When the testis are examined at a later date, day 112, ten days after androgen treatment ceases, Leydig cell morphology and steroidogenic gene expression return to control values, although fascinatingly, an overshoot of in vivo testosterone production (P<0.01) was observed. When the maternal androgenisation window is extended to begin at day 30 of fetal life, further changes are noted including increased circulating testosterone (P<0.01), a strong trend toward decreased testis weight (P=0.0519) and altered expression of Sertoli and germ cell specific markers. These studies are followed up by assessing the legacy effect of testosterone on the peripubertal male testis in Chapter 5. At ten weeks of postnatal life, males, exposed to androgens from day 62-102 of fetal life had reduced testis weight (P<0.05). However, functional or cellular alterations were not observed and by 12 weeks of age, when LH had normalised, testicular weight and stimulated testosterone secretion of prenatally TP-treated males was comparable to controls. This highlights the remarkable plasticity of the testis and the unremarkable legacy of altered prenatal androgen exposure. The legacy effect of testosterone on the fetal ovary is examined in Chapter 6. Previous studies from our laboratory found minor functional alterations but no structural alterations in the fetal ovary at day 90 following androgenisation from day 62. However, as this was at a time of a highly androgenic environment we assessed the function and morphology of the ovary ten days after the removal of testosterone at day 112. In marked contrast to the normalisation of the male gonad, we observe structural changes with an increase in recruited follicles from the primordial to primary stage in the testosterone treated group (P<0.01). The chapter continues with an investigation of pathways involved in the altered follicular dynamics that may account for the change in follicular recruitment. Furthermore, the functional changes which were previously noted in the day 90 ovary were also examined in response to direct exogenous steroid treatment including, C, TP, DES and dexamethasone (DEX) and also when the window of maternal androgenisation was extended to begin at day 30. Interesting changes are observed such that the direct fetal injection treatments induce similar changes to each other, regardless of the steroid, whilst maternal androgenisation induces a different response. This highlights the complexity of the pathways involved in female gonadal development.

Submitted by Samira Ramos (samira.ramos@unifenas.br) on 2019-02-26T18:07:29Z No. of bitstreams: 1 Tese_Ariane_Borges.pdf: 1192468 bytes, checksum: 14a5f0b78794c95edf17ef4ee983572f (MD5) / Made available in DSpace on 2019-02-26T18:07:29Z (GMT). No. of bitstreams: 1 Tese_Ariane_Borges.pdf: 1192468 bytes, checksum: 14a5f0b78794c95edf17ef4ee983572f (MD5) Previous issue date: 2018-06-29 / Fundação de Amparo à Pesquisa do Estado de Minas Gerais - FAPEMIG / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - CAPES / Variations in lipids and lipoproteins are considered a risk factor for cardiovascular diseases, such as atherosclerosis. This study evaluated the effect of genetic and alimentary dyslipidemia on testicular morphofunctional characteristics and whether its relation with the cardiovascular diseases developed by hyperlipidic diets in genetically modified mice, absent LDL receptor (Ldlr-/-) and wild type mice. In addition to evaluating the replacement of testosterone in the protection of the cardiovascular system of Ldlr-/- mice fed or not with a hyperlipidic diet. In the first study, Ldlr-/- mice were selected, divided into four groups (n = 10): S: animals fed standard diet (Nuvital®) for rodents and without testosterone application; ST: animals fed standard diet (Nuvital®) for rodents and with testosterone application (0.01 ml per week); HL: animals fed a hyperlipid diet (20% total fat, 1.25% cholesterol and 0.5% cholic acid) and without testosterone application; HLT: animals fed a hyperlipid diet (20% total fat, 1.25% cholesterol and 0.5% colic) and with testosterone application (0.01 ml per week). In the analysis of the lipid profile, the mice that received hyperlipidic diets (HL and HLT) presented severe mixed dyslipidemia with increased serum levels of total cholesterol, LDL, VLDL and triglycerides, when compared to the mice of the S and ST groups. However, the HLT group showed an increase in serum HDL levels when compared to the mice in the HL group. The mice from the S and ST groups had increased serum HDL levels in relation to the other groups studied. In the second study, it was observed that the dyslipidemia generated by the LDL receptor deficiency (Ldlr-/-) has a positive relation with the increase in the production of vascular superoxide anions, increased expression of CD40 and FasL in the testis. Genetic deletion of the LDL receptor (Ldlr-/-) in mice associated with a hyperlipidic diet increased both systemic and testicular damage. In conclusion, the metabolic disorders of the lipids generated by the deletion of the LDL gene induced testicular dysfunction, by mechanisms involving oxidative stress, inflammation and apoptosis, impairing spermatogenesis and testicular steroidogenesis. It is also suggested that testosterone may indirectly cause hypertrophy of cardiomyocytes / As variações nos lipídeos e lipoproteínas são consideradas um fator de risco para doenças cardiovasculares, como a aterosclerose. Este estudo avaliou o efeito da dislipidemia genética e alimentar nas características morfofuncionais testiculares e se sua relação com as doenças cardiovasculares desenvolvidas por dietas hiperlipídicas em camundongos geneticamente modificados, ausente do receptor do LDL (Ldlr-/-) e em camundongos selvagens (wild type). Além de avaliar a reposição da testosterona na proteção do sistema cardiovascular de camundongos Ldlr-/- alimentados ou não com dieta hiperlipídica. No primeiro trabalho, foram selecionados camundongos Ldlr-/-, divididos em quatro grupos (n=10): S: animais alimentados com dieta padrão (Nuvital®) para roedores e sem aplicação de testosterona; ST: animais alimentados com dieta padrão (Nuvital®) para roedores e com aplicação de testosterona (0,01ml por semana); HL: animais alimentados com dieta hiperlipídica (20% de gordura total, 1,25% de colesterol e 0,5% de ácido cólico) e sem aplicação de testosterona; HLT: animais alimentados com dieta hiperlipídica (20% de gordura total, 1,25% de colesterol e 0,5% de ácido cólico) e com aplicação de testosterona (0,01ml por semana). Na análise do perfil lipídico, os camundongos que receberam dietas hiperlipídicas (HL e HLT) apresentaram dislipidemia mista severa com aumento dos níveis séricos do colesterol total, LDL, VLDL e triglicérides, quando comparados com os camundongos dos grupos S e ST. Contudo os camundongos dos grupos HLT apresentaram aumento nos níveis séricos de HDL quando comparados com os camundongos do grupo HL. Os camundongos do grupo S e ST apresentaram níveis séricos de HDL aumentados em relação aos demais grupos estudados. No segundo trabalho, observou-se que a dislipidemia gerada pela deficiência do receptor de LDL (Ldlr-/-) tem uma relação positiva com o aumento da produção de ânions de superóxido vascular, aumento da expressão de CD40 e FasL no testículo. A deleção genética do receptor de LDL (Ldlr-/-) em camundongos associados a uma dieta hiperlipídica aumentou o dano sistêmico e testicular. Em conclusão, os distúrbios metabólicos dos lipídeos gerados pela deleção do gene LDL induziram a disfunção testicular, por mecanismos envolvendo estresse oxidativo, inflamação e apoptose, prejudicando a espermatogênese e a esteroidogênese testicular. Sugere-se também que a testosterona possa causar indiretamente a hipertrofia de cardiomiócitos.

Nonmammalian vertebrates possess some unusual features in their hormonal systems/ when compared to mammals. As a consequence, they can make an important contribution in investigations concerning the fundamental mechanisms operating in endocrinology. Such studies concerning androgens include inter alia their effects on developmental aspects in the brain of birds and related singing behaviour; the role of neural enzymes in reproductive processes in fish; and the relation between androgens and the stages of spermatogenesis in amphibia, The present thesis examines the biochemistry of androgens in the Australian lizard Tiliqua rugosa. The major compounds studied were testosterone and epitestosterone, which are known to be present in high concentrations in the plasma of the male animal. Previous investigations are expanded, particularly in the areas of steroid identification and testicular biosynthesis. In addition, preliminary studies on the metabolism in the brain (and other tissues) and plasma protein binding are reported. The presence of epitestosterone as a major free androgen in the plasma of the male lizard was confirmed. Other steroids were found in the sulphate fraction. Testosterone sulphate was the most rigorously identified compound, while some evidence was also found for the presence of conjugated 5-androstene-3β,17-diols, etiocholanolone and dehydroepiandrosterone (DHA). Epitestosterone does not appear to be extensively conjugated in this animal. Steroids were not found to be conjugated as glucuronides. The identification studies employed a novel method of electrochemical detection of steroids. This technique was investigated and extended in the current thesis. Biosynthetic studies were carried out on androgen interconversions in the testis, in vitro. The major enzyme activities detected were 17α-arid 17β-oxidoreductases (17α-OR and l7β-OR) and 3β-hydroxysteroid dehydrogenase (3β-HSD)/isonerase. No evidence was found for the presence of a steroid-17-epimerase that would directly interconvert testosterone and epitestosterone. The 17-oxidoreductases were found to be dependent on the cofactor NBDFH. Testosterone appears to be formed mainly via the 4-ene pathway, whereas epitestosterone is formed from both the 4- and 5-ene routes. The compound 5-androstene-3β, 17α-diol was found to be an intermediate in the synthesis of epitestosterone from DHA. Temperature was found to significantly affect 17α-OR activity (maximum at 32°C). In contrast,17β-OR activity was independent of this factor in the testis. Androgen metabolism in the testis was found to be regulated by cofactors, temperature and season. The major enzyme activities found in the male brain were 17α- and 17β-OR. 3βHSD/isomerase was not found; however a low activity of 5α-reductase was identified. Aromatase activity was not positively identified, but preliminary results suggest that it may be present at low levels. The 17-oxidoreductases were widespread throughout the brain. The 17α-OR was significantly lower in the forebrain than other brain sections. The 170-OR activity did not vary significantly throughout the organ, although there was a trend for its activity to be higher in the midbrain region (containing the hypothalamus in these sections). The concentration of endogenous steroids in brain tissue was estimated by radioimmunoassay. Epitestosterone was found throughout the organ structure, whereas testosterone was found mainly in the midbrain (containing hypothalamic regions in these sections). Correlations between enzyme activities and steroid concentrations in brain regions suggested that the main function of 17α-OR is to produce epitestosterone, whereas the 17β-OR may catalyse a more reversible reaction in vivo. Temperature was found to significantly affect both 17α- and 17β-OR activities in the brain. In contrast to the testis, the maximum activity of the brain enzymes occurred at 37°C. The level of 17α-OR activity in the male lizard (100 nmol/g tissue/h) is the highest reported for this enzyme in vertebrates. Both activities were found to be quantitatively similar in the whole brain homogenates of male and female animals, and did not vary seasonally when examined in the male. The 17-oxidoreductases were also found in most other tissues in T.rugosa, including epididymis, adrenal, kidney and liver (but not blood). This suggests that the high activities of both 17α-OR and 17β-OR are dominant features of the steroid system in this animal. The formation of 11-oxygenated compounds was found in the adrenal, in addition to the formation of polar metabolites in the kidney and liver (possibly polyhydroxylated and conjugated steroids). A preliminary investigation into the plasma binding of androgens was carried out. The insults suggest that there are several binding sites for testosterone; one with high affinity and low capacity; the other with low affinity and high capacity. Binding experiments were carried out at 32°C. At this temperature, specific binding was greater than at 25 or 37°C. From the results of competition studies it was suggested that epitestosterone (with a K(i)= 3 X 10 (-6)M for testosterone binding) regulates the binding of testosterone (K(i)=10(-7)M) and hence the concentrations of the latter steroid as a free compound in plasma. In general, the study has shown that the biochemistry of androgens in the reptile T.rugosa is largely similar to that found in other vertebrates. The major difference is a greatly increased activity of 17α-OR, which causes a higher concentration of 17α-compounds to be present in the tissues of this lizard. The physiological roles for epitestosterone are not yet clear. However it appears from this study that this steroid regulates testosterone concentrations in several tissues by either steroidogenic or binding mechanisms. Several major influences on this regulation include temperature, availability of cofactors and seasonal effects.

Understanding and studying the normal bone growth and development is causal. Bone and cartilage tissue provide in addition to their mechanical support, they provide a protection for vital organs such as heart, lung and brain. Longitudinal growth is regulated by the activity of chondrocytes in the epiphyseal growth plates of long bones. Many hormones and growth factors are involved in the regulation of this process. Among these, sex steroids are of crucial importance, especially during puberty. In long bones, endochondral bone formation occurs at the growth plate, a region of developing cartilage located between the epiphysis and the metaphysic. The process of endochondral ossification is regulated in part by sex steroid hormones. Androgens stimulate endochondral bone growth and elongation, while estrogen is known to suppress longitudinal bone growth and accelerate growth plate closure. Studies using rat costochondral growth plate chondrocytes as a model show that the effects of 17β-estradiol (E₂) on apoptosis are found in both male and female cells and the same mechanism is involved. In contrast, E₂ causes rapid activation of PKC in female cells but not in male cells. Dihydroxytestosterone (DHT) also has direct effects on growth plate chondrocytes, increasing matrix synthesis including sulfated glycosaminoglycan production, and enhancing cell maturation by increasing alkaline phosphatase enzymatic activity. Short stature and abnormally slow increase in height is one of the main reasons for referral to endocrinologist. Excessive growth and abnormally tall is also a problem, especially because it increase risk for the trunk abnormalities. Furthermore until now a few growth-promoting therapies are available for clinical use. Therefore future therapies for treating the growth disorders are essential. The overall goal of this project is to investigate the sexual-dimorphic effect of the sex steroid hormone in rat growth plate chondrocytes, the cellular signaling pathways mediating these actions, and their physiological role. The information gleaned from this study will provide new information about the role of sex steroid hormones in chondrogenesis and has implications in the development of new therapies for the treatment of bone fracture healing, and growth plate disorders. The central hypothesis was that sex steroid would play an important and sex-specific role in regulating chondrocytes as a main regulator of longitudinal bone growth.

Androgens (such as testosterone) have been shown to increase lean body mass and reduce fat body mass in men through activation of androgen receptors (AR). While this suggests a potential clinical use for androgens, attempts at utilization of this class of hormones as a therapeutic are limited by side effects due to indiscriminate AR activation in various tissues. Thus, a greater understanding of the tissues and cells involved in promoting these changes would be beneficial. Here we show that selective overexpression of AR in muscle cells of transgenic (HSA-AR) rodents both increases lean muscle mass and significantly reduces fat mass in males. Similar effects can be induced in HSA-AR females treated with testosterone. Metabolic analyses of HSA-AR males show that these animals demonstrate increased O2 consumption and hypermetabolism. Thus, targeted activation of AR in muscle regulates body composition and metabolism, suggesting a novel target for drug development.

Androgens (such as testosterone) have been shown to increase lean body mass and reduce fat body mass in men through activation of androgen receptors (AR). While this suggests a potential clinical use for androgens, attempts at utilization of this class of hormones as a therapeutic are limited by side effects due to indiscriminate AR activation in various tissues. Thus, a greater understanding of the tissues and cells involved in promoting these changes would be beneficial. Here we show that selective overexpression of AR in muscle cells of transgenic (HSA-AR) rodents both increases lean muscle mass and significantly reduces fat mass in males. Similar effects can be induced in HSA-AR females treated with testosterone. Metabolic analyses of HSA-AR males show that these animals demonstrate increased O2 consumption and hypermetabolism. Thus, targeted activation of AR in muscle regulates body composition and metabolism, suggesting a novel target for drug development.

A number of widely used industrial chemicals have been shown to possess endocrine-disrupting properties. In this thesis, a series of in vitro tests, and an in vivo reproductive performance test with fathead minnows, were used to clarify the extent of estrogenic activity exhibited by the phthalate esters - a class of compound hitherto referred to as 'estrogenic'. Using a recombinant yeast estrogen screen, I demonstrated that a small number of commercially available phthalates showed extremely weak estrogenic activity. The most potently estrogenic phthalate of those tested was BBP, which was approximately one million-fold less potent than 17B-estradiol. The phthalates which were estrogenic in the yeast screen were also mitogenic on estrogen-responsive human breast cancer cells (MCF-7 and ZR-75 cell lines). The most prolifically used phthalate, DEHP, was not estrogenic in any of these assays. The small number of metabolites of phthalate metabolites tested (including MBuP, MBzP, MEHP and MnOP) were also not estrogenic in the recombinant yeast assay. The ability of BBP (as the most potently estrogenic phthalate in vitro) to induce a vitellogenic response (an indicator of estrogen exposure) in fathead minnows (Pimephales promelas) exposed via the water was assessed. No induction of vitellogenin was observed, indicating that 100 Jlg BBP/L (a concentration higher than would normally be found in the environment) is not estrogenic to this species of fish under the conditions employed for this experiment. In the same study, fecundity of breeding pairs of fathead minnows was assessed; exposure to BBP was not found to affect reproductive performance in these fish. A possible alternative mechanism of action of the way in which the phthalates induce frequently reported reproductive disorders was observed. Some of the phthalates, and, notably, some of their metabolites, were demonstrated to act as anti-androgens in a recombinant yeast androgen assay. 4-Nonylphenol is another industrial chemical which is used in large volumes, and due to the nature of its use (mainly in detergent formulations), is discharged into water systems via sewage effluents. This chemical has been shown to be estrogenic to fish at the concentrations at which it has been detected in the environment. 4-Nonylphenol was tested for its ability to affect plasma and pituitary gonadotropin levels in female recrudescing rainbow trout (Oncorhynchus mykiss). Plasma and pituitary levels of FSH were suppressed in fish exposed to 10 and 100 Jlg 4-NP/L. In addition, FSH gene expression was reduced in these fish, and also in the fish exposed to 1 Jlg 4-NP/L. Pituitary LH content and gene expression of this hormone were suppressed in the fish exposed to 100-, and 10- and 100 Jlg 4-NP/L respectively. Gonadal development in vertebrates is regulated by FSH. Ovarian development ceased in the fish exposed to 100 JAg 4-NP/L, possibly as a result of the suppression of FSH synthesis and/or release in these fish.

Background: Androgens are important in prostate cancer development but how circulating levels of androgens affect risk of prostate cancer of different aggressiveness is not clear. Being childless has been associated with a lower risk of prostate cancer, but it is not clear if this association is causal or a result of residual confounding. Fathering of dizygotic twins, a marker of high fertility, has not been studied in relation to risk of prostate cancer. Another marker of life-long hormonal exposure is height, which has been associated with increased risk of cancer and cancer death. However, the association to separate cancer sites has not been consistent. The aims of this thesis were to study hormonal factors (paper I), and proxies of hormonal factors (paper II and III), and risk of prostate cancer; as well as height and risk of cancer and cancer death by separate sites (paper IV). Methods: Study designs were i) case-control studies, nested within the Västerbotten Intervention Project (paper I), and in Prostate Cancer database Sweden 2.0 (PCBaSe 2.0) (paper II and III), and ii) cohort study, in the Metabolic Syndrome and Cancer project (Me-Can) (paper IV). Results, prostate cancer: In paper I, increasing levels of serum androgens were not associated with risk of prostate cancer overall or in tumor risk categories. In paper II, childless men had a lower risk of prostate cancer, overall and in all risk categories, compared to fathers, an association which was in part explained by differences in marital status and educational level.  In paper III, fathers of dizygotic twins did not have an increased risk of prostate cancer, either overall or in risk categories, when compared to fathers of singletons. Results, cancer overall: In paper IV, height was associated with an increased risk of cancer and cancer death overall in both women and men. The strongest association for cancer was to malignant melanoma in both women and men, and for cancer death to post-menopausal breast cancer in women and renal cell carcinoma in men. Conclusions: These studies indicate that hormonal factors, when studied as serum levels or when studied using proxies of fertility, do not have a major impact on the risk of prostate cancer. The association between height and an increased risk of cancer appears robust for total cancer and cancer death, as well as for several separate cancer sites.

Epidemiologiniais tyrimais nustatytos androgenų sąsajos su įvairiais metaboliniais veiksniais bei širdies ir kraujagyslių ligų rizika paskatino susidomėjimą naujais vyrų reprodukcinės sveikatos aspektais. Iki šiol Lietuvoje nebuvo nustatytos normalios vyrų androgenų koncen¬tracijos ribos, nebuvo atlikta reprodukcinės sveikatos, antropometrinių ir metabolinių rodiklių sąsajų analizės tyrimų. Pasaulyje iki šiol trūksta nuoseklios informacijos apie metabolinių ir antropometrinių rodiklių ryšius su lytiniais hormonais, ypač homogeniškose tiriamųjų imtyse. Trūksta informacijos apie androgenų sąsajas su pažintinėmis funkcijomis, emocine būkle ir gyvenimo kokybės vertinimu. Daugelis tyrimų, kuriais vertintas pakaitinio gydymo testosteronu efektyvumas – trumpalaikiai, juose dau¬giausiai dėmesio skirta struktūriniams, o ne funkciniams rodikliams. Šiuo tyrimu siekta aprašyti jaunų bei vidutinio amžiaus vyrų reprodukcinės sveikatos sąsajas su antro¬pometrijos, metabolinės bei psichologinės būklės rodikliais. Tai iki šiol didžiausia savo apimtimi Lietuvos vyrų androgenų sąsajų analizė. Šiuo darbu siekiama sistemingai atsakyti į iškeltus aktualius klau¬simus, nustatyti normalios ir optimalios androgenų koncentracijos ribas, homogeniškose amžiumi tiriamųjų grupėse įvertinti šių hormonų koncen¬tracijos grupių sąsajas su kūno sudėties ir medžiagų apykaitos rodikliais bei nustatyti, kokius sergančiųjų androgenų nepakankamumu funkcinius pokyčius sąlygoja ilgalaikis optimalus pakaitinis... [toliau žr. visą tekstą] / Associations between androgens and various metabolic factors, as well as cardiovascular morbidity shown in recent epidemiological studies sparked interest in new aspects of male reproductive health. Reference values for normal androgen levels in Lithuanian population have not yet been established and no relationship studies on reproductive health and anthropometric, as well as metabolic parameters have been performed. Unequivocal information on associations among the aforementioned factors is lacking globally, especially those performed in homogenous populations. More information on androgen associations with cognitive function, emotional state and evaluation of quality of life is required. Most of the studies on effectiveness of testosterone replacement therapy were of short duration and mostly focused on structural, rather than functional parameters. The purpose of the study was to describe associations between reproductive health and anthropometric, metabolic, as well as psychological parameters in young and middle-aged men. This is the largest androgen association analysis in Lithuanian population. This work seeks for systematic responses to questions of interest: to establish reference values for normal and optimal androgen levels, to evaluate their relationship with anthropometric and metabolic factors in populations that are homogenous by age, and to assess the functional changes that long-term optimal testosterone replacement therapy may induce in patients with... [to full text]

The growth and survival of normal breast epithelial cells and breast cancer cells is promoted by estrogens. In contrast, androgens inhibit the proliferation of normal and malignant breast epithelial cells. While this effect of androgens on breast cells appears to be androgen receptor (AR) dependent, the precise mechanism of inhibition and its functional significance are unknown. The aims of this thesis were to investigate the effect of androgen signalling on growth of normal and malignant breast epithelial cells, and to assess the interactions between androgen and estrogen signalling in the breast. To investigate the role of androgen signalling in the growth and development of the normal mammary gland, female mice were treated with either the native androgen 5α- dihydrotestosterone (DHT) or the antiandrogen, flutamide. Analysis of the mammary glands at the end of the treatment period demonstrated that DHT reduced ductal branching and mammary epithelial cell proliferation when treatment commenced mid-puberty. Conversely, flutamide treatment that commenced post-puberty significantly increased ductal branching and proliferation of mammary epithelial cells. This data demonstrates that androgen signalling inhibits proliferation in the normal mammary gland, and may therefore oppose to the growth stimulatory effects of estrogen signalling to regulate breast growth and development. The antiproliferative effects of androgens on breast epithelial cells may be due in part to direct AR-mediated activation of androgen regulated genes, or alternatively, androgens could act indirectly through AR to inhibit estrogen receptor alpha (ERα) activity. Expression of fulllength AR or a truncated, constitutively active AR (AR-T707) significantly inhibited the activity of ectopically expressed ERα in MDA-MB-231 breast cancer cells (ERα- and ARnegative), in a dose-dependent manner. The functional consequences of inhibition of estrogen signalling by overexpressing AR were investigated in the T-47D breast cancer cell line (ERα- and AR-positive). Expression of AR-T707 in T-47D cells resulted in inhibition of both basal and estradiol-induced cell proliferation and a marked reduction in the steady-state protein levels of the estrogen regulated gene, PR. The final chapter investigated the mechanism by which AR inhibits ERα activity. A coimmunoprecipitation assay demonstrated an interaction between ectopically expressed AR and ERα in COS-1 cells, but not endogenous AR and ERα in a breast cancer cell line. To delineate the regions of AR required for inhibition of ERα signalling, various functional domains of the AR were mutated or deleted. Reporter gene assays showed that the inhibitory effects of AR were abrogated by deletion or mutation of the DNA binding domain (DBD). Furthermore, overexpression of the AR-DBD alone was sufficient to inhibit ERα activity. Consistent with a requirement for the DBD of AR to inhibit ERα activity, mobility shift assays demonstrated binding of AR to the Xenopus vitellogenin A2 consensus estrogen response element (cERE); however AR/ERα heterodimers were not detected on a cERE. Consistent with these findings, molecular modelling demonstrated that it is feasible for the DBD of AR to bind to a cERE and that it is unlikely that AR/ERα heterodimers could bind. Chromatin immunoprecipitation demonstrated recruitment of AR to the promoters of endogenous estrogen regulated genes. The findings suggest that the inhibitory effect of AR on ERα activity may occur either via formation of non-functional AR/ERα heterodimers that are unable to bind to EREs, or AR homodimers competing effectively for binding to EREs, in ERα target genes. The results in this thesis demonstrate an inhibitory effect of androgen signalling on growth of normal and malignant breast epithelial cells. Additionally, the inhibition of breast epithelial cell proliferation by androgen signalling can be attributed, at least in part, to inhibition of ERα activity. These studies have provided insight into androgen action in the breast, and support a model whereby androgens balance the stimulatory effects of estrogen signalling in normal and malignant breast epithelial cells. / Thesis (Ph.D.) -- University of Adelaide, School of Medicine, 2008