Microscopic Evaluation of Ovarian Surface Epithelium Following Treatment with Conjugated Estrogens in a Mouse Model

Background: This study was designed to evaluate the effect of different concentrations of conjugated equine estrogens (CEE) on the ovarian epithelium of female CD1 mice. Methods: Twenty-four female mice at 7 months with irregular estrus cycles were randomly divided into four groups of 6 mice each. Group one was considered as a control group and received a daily dose of 0.5ml of propylene glycol, for three weeks, while those in the treatment groups received a daily dose of 14μg/kg, 28μg/kg and 56μg/kg conjugated equine estrogens, respectively. Results: The results from this study showed a strong correlation between elevated concentrations of CEE and histological changes in ovarian surface epithelium (OSE). They also showed that administration of high-dose estrogen created the conditions for excessive proliferation of OSE which may progress into the development of cysts in the ovaries. Conclusion: This study concluded that high concentrations of CEE may increase the chances of developing epithelial ovarian cancer.


Introduction
Ovarian cancer is one of the most common gynecological malignancies and is ranked third after cervical and uterine cancers worldwide Sung et al., 2020). In the Arab countries, ovarian cancer ranked fourth among top ten diagnosed cancer cases in women from different age groups (Oetashi and Al Kalbani, 2013). Ovarian cancer also has the worst prognosis and the highest mortality rate, and it is three times more lethal than breast cancer (Yoneda et al., 2013;Coburn et al., 2017). Ovarian cancer is a heterogeneous disease and the subtypes are very different from other gynecological cancers such as breast and endometrial cancers with respect to origins, behaviors, mutations, markers, and prognosis and responses to therapy (Mungenast and Theresia, 2014). Epithelial ovarian cancer (EOC) constitutes the majority of malignant ovarian tumors in adult women and is the commonest cause of gynecological cancer-associated death with the highest mortality rate of all cancers in the female reproductive system with a five-year survival Murdoch, 1996). OSE at the site of rupture becomes exposed to stromal microenvironment (Wu et al., 1992(Wu et al., , 1993 and fragments of OSE become invaginated in the ovarian cortical stroma giving rise to epithelial crypts and inclusion cysts (Womg and Leung, 2007;Auersperg, 2013). These structures regarded as preneoplastic and may promote malignant transformation of entrapped OSE by generating a microenvironment enriched with growth factors, cytokines and hormones (Berchuck, 1997;Auersperg, 2013). Higher frequency of inclusion cysts is observed in postmenopausal than in premenopausal women (Kronenberg et al., 2008).
Established risk factors for the development of EOC are age, infertility, and family history, whereas oral contraceptive use, increasing parity, hysterectomy or tubal ligation decreases the risk (Riman et al., 2002;Kazerouni et al., 2006;Hunn and Rodriguez, 2012;Rice et al., 2014;Zhang et al., 2019;Momenimovahed et al., 2019). In addition, well-established associations between reproductive physiognomic and EOC support an involvement of reproductive steroid hormones in the etiology of EOC (Riman et al., 2004;Schock et al., 2014). Furthermore, studies reported that steroid hormones, including estrogens, progesterone and androgens, are involved in ovarian carcinogenesis (Pearce et al., 2009;Koskela-Niska et al., 2013a, Blanco et al., 2017. Menopausal hormone replacement therapy (HRT) is widely used to treat postmenopausal symptoms and reduce bone loss. Findings from epidemiological studies investigating the effects of HRT on carcinogenesis are contradictory. The use of physical or oral contraceptives including HRT have been reported to increases the risk of ovarian cancer (Morch et al., 2009;Tsilidis et al., 2011;Urban et al., 2015;Simin et al., 2017). However, several studies found no relationship between them (Hunn andRodriguez, 2012: Li et al., 2015;Rasmussen et al., 2017) and others studies found a positive association in individual histological subtype (Morch et al., 2012;Yang et al., 2012;Koskela-Niska et al., 2013b).
Conjugated equine estrogens (CEE) are commonly used in postmenopausal HRT, and administration of CEE in animal model reduces ovulation and promotes proliferation of the OSE (Murdoch and Van Kirk, 2002;Laviolette et al., 2014) as well as the emergence of ovarian cysts (Bai et al., 2000;Laviolette et al., 2010). However, the effects of these hormones on the OSE are not fully understood. Estrogens favor neoplastic transformation of the OSE, and their mechanisms of action on the tumorigenic process are complex, involving considerable genetic alterations (Hua et al., 2018). Several studies found a significant estrogen increase in EOC risk Beral, 2007;Rasmussen et al., 2017) but others reported no significant positive associations (Simin et al., 2017;Coughlin et al., 2000). Other studies have suggested there may be a small increased risk of EOC associated with longer-term use of estrogen (Ho, 2003;Wentzensen et al., 2016). Data about the association of HRT with ovarian cancer, therefore, is still controversial (D' Alonzo et al., 2019). The aim of the present study was to evaluate the carcinogenic effects of three different concentrations of CEE on the histological structural changes of ovarian surface epithelium of the adult female mice at cellular level following a twenty-one-day period.

Animals
Twenty-four female CD-1 mice of 7 months of age and weighing between 30 and 35g, were used in this study following approval by the Animal Ethics Committee at University of Nizwa. Mice were divided into four groups of six mice per cages and housed in a small animal facility at University of Nizwa on a 12-hour light cycle with access to ad libitum food and water with controlled room temperature (24-25oC) and humidity (30-35%). One week after adaptation, the first set represents control group and received a daily dose of 0.5ml of propylene glycol (vehicle). The second, third and fourth groups received a daily dose of 14μg/kg, 28μg/kg and 56μg/kg conjugated equine estrogens (CEE), respectively. The conjugated equine estrogens were diluted in 0.5 ml of propylene glycol and administered daily in the morning by gavage for 21 days. Gavage were performed with a standard metal tube supplied by the Stem Cells and Regenerative Medicine Department at the Natural and Medical Research Center, University of Nizwa.

Histology
A day after final treatment, mice were decapitated under deep anesthesia of 5% isoflurane for 5 minutes. Both ovaries were dissected using stereomicroscope and examined with the naked eye for gross abnormal changes then immersed in 10% formaldehyde solution for tissue processing. Tissue preparations were accomplished by using the Automated Tissue Processer instrument (ATP 140). In this process tissues were dehydrated by graded concentrations of ethanol 50%, 70%, 90%, 1h each and 3 changes of 100% ethanol for 1, 2 and 2½h, respectively. Tissues were then cleared with 3 changes of xylene for 1, 1, and 2h, respectively, then infiltrated in 3 changes of melted paraffin wax at 60oC for 1h each. Samples were then embedded in pure paraffin wax by using the Embedding Center Machine (AEC 380).
Paraffin blocks were sectioned by Rotary Microtome (AEM 450) to produce thin sections of 5 μm thickness, three slides being therefore prepared for each block. Sections were collected with glass slides and placed on hot plate. Sections were then placed in xylene for 2 periods of 5 minutes each to remove wax, then hydrated in 2 changes of pure ethanol of 5 minutes each, then in 95% ethanol and then in 70% ethanol for 2 minutes each, then in distilled water for 2 minutes. Each section stained with Hematoxylin and Eosin (H and E), dehydrated with 95% ethanol for 3 to 5 dips, 2 changes of 100% ethanol for 2 minutes each, and then cleared with xylene for 5 minutes each. Finally, slides were mounted with DPX and coverslip and screened under digital light microscope. Ten randomly selected non-overlapping fields were investigated by histologists under light microscopy for each slide at different magnifications. and made up of theca layer, granulosa layer with an antrum fluid, zona pellucida, ooplasm and oocyte (Figure1B). The ovarian surface epithelium (OSE) made up of monolayer of squamous epithelium ( Figure 1C).
In group 2, tissues from the ovaries exhibited no change in the morphology after three-week treatment with CEE ( Figure 2A-2C). In group 3, the ovaries exhibited slight thickening in the OSE and increased cellular contents of the theca-interstitial tissue in stroma ( Figure 3A-3C). The cortex and medulla showed normal morphological details ( Figure 3A) and the structure of developing ovarian follicles made up of theca layer, granulosa layer, zona pellucida, ooplasm and oocyte ovarian follicles ( Figure 3B). In stroma, however, an increased number of theca-interstitial cells was observed. The OSE showed abnormal morphology with slight thickening of OSE exhibiting multilayer epithelium ( Figure 3C). In group 4, the ovaries exhibited marked alterations to the OSE morphology after three-week treatment with CEE ( Figure  4A-4C). The ovarian follicles in the cortex and the medulla showed normal morphological details with the exception that the stroma exhibited an increased number of cells in the theca-interstitial tissue ( Figure 4A, 4B and 4C). The OSE appeared thicker than those observed in group 3

Results
Light microscopic observation of ovary sections showed significant difference among groups 1, 2, 3 and 4. Control group (1) showed normal ovarian morphology with uniform surface epithelium, ovarian follicles, and connective tissue (stroma) as shown in Figure 1A-1C. Group 4 showed significant changes in the ovarian morphology with thickening of surface epithelium as shown in Figure 4A-4C. The mice in the control group sections showed normal ovarian tissue made up of cortex and medulla ( Figure 1A-1C). The cortex consists of surface epithelium, tunica albuginea, ovarian follicles, and stroma. The medulla consists of blood vessels, lymphatic vessels, and nerve fibers in stroma ( Figure  1A). The ovarian follicles are in different stages of follicular development consisting of primordial, primary, and secondary follicles ( Figure 1B Figure 1A, showing the ovarian cortex and medulla. The cortex is made up of OSE (SE), ovarian follicles (OF), and stroma (ST). Figure 1B, the cortex consists of primordial follicle (PRO), primary follicle (PFO), and secondary follicle (SFO). Both primary and secondary follicles structures include theca layer of endocrine cells (TH), granulosa layer of cuboidal epithelium (G), ooplasm (P) and oocyte (O). SFO with antrum fluid (A). Figure 1C, the OSE made up of monolayer of squamous epithelium (SE). Secondary follicle (SFO), stroma (ST). Mag. 100X, 400X and 600X, for A, B, and C, respectively. with increased multilayered squamous epithelium clearly evident ( Figures 4B and 4C).

Discussion
Estrogen replacement therapy was initially prescribed for the treatment of menopausal symptoms in the early 1940s. However, due to increased incidents of endometrial cancer in women in the 1970s, estrogen hormonal therapy declined and as a result progestin was introduced to HRT formulations which led to improved management of symptomatic menopause and prevention of chronic diseases (McNeil, 2017). Reports of increased risks of breast cancer and cardiac disease in the early 2000 led to closure of the estrogen and progestin therapy after 5 years of follow-up (Rossouw et al., 2002). Nevertheless, estrogen remain the most effective treatment prescribed to date for the management of menopausal symptoms following a thorough evaluation of the risks and benefits (Hill et al., 2016). Significant association between ovarian cancer and HRT remains controversial largely due to insufficient data. Observational studies reported association of estrogen use vary substantially by histologic subtype and appeared mainly associated with non-serous cancers . However, a recent metaanalysis of studies by histologic subtypes reported menopausal HRT may increase the risk of serous and endometrioid tumors (Liu et al., 2019).
Estrogen is a steroid hormone produced primarily in the ovaries in females and to a lesser extent, in the testes in males and plays an important role in reproductive development, brain functions, cardiovascular remodeling and bone homeostasis. However, high concentrations of estrogen also promote ovarian, mammary and endometrial tumorigenesis (Hua et al., 2018). Estrogen and gonadotropins are known to control the ovary and are strong regulators of surface epithelium proliferation. Studies have reported that gonadotropins and estrogens stimulate surface epithelium cell proliferation (Bai et al., 2000;Hess et al., 1999;Song et al., 2020) through hormone receptors that are present on the surface of cells of the surface epithelium (Kuroda et al., 2001;Saddick, 2014). Evidence showed that estrogen receptors alpha and beta are present in 60-100% of ovarian tumors resected from patients (Lindgren et al., 2004;De Stefano et al., 2011). Other studies have found an epithelial ovarian cancer cell line, PE04, containing high levels of cytosolic estrogen receptor-like binding (Nash et al., 1989).  Figure 3A, showing the cortex consists of normal ovarian follicles (OF) and abnormal OSE (SE). The cortex appeared denser than those observed in groups 1 and 2. Figure Figure 4A showing the cortex consists of normal ovarian follicles (OF) and abnormal OSE (SE). The cortex appeared denser than those observed in other groups. 2Figure 4B, showing OSE (SE), primordial follicle (PRF) and primary follicle (PFO). Stroma (ST). Marked thickening (arrow) observed in the OSE. Figure 4C The monolayer of OSE is a heterogeneous population with multiple morphologies observable in primary cultures and animal models (Okamura and Katabuchi, 2003). Ovarian surface epithelial (OSE) cells exposed to elevated levels of estradiol develop a subpopulation that exhibit loss of cell polarity, contact inhibition, and forms multilayered foci of dysplastic cells with increased susceptibility to transformation (Vuong et al., 2018). Defects in cell polarity are often reported in human cancer (Fish and Molitoris, 1994;Bissell and Radisky, 2001) and loss of contact inhibition usually results in outgrowth of cells from the epithelial monolayer (mcclatchey and Yap, 2012). In this study, the three-week treatment with CEE administrated in adult female mice have demonstrated that estrogen induced OSE dysplasia as a direct result of prolonged exposure to estrogen. Tissue samples obtained from the ovaries of different mice in group three treated with CEE showed slight thickening of the ovarian surface epithelium (OSE) exhibiting multilayered squamous epithelium. Samples from group four treated with CEE, however, showed significant thickening of OSE in the ovaries. OSE exhibited an increase in the multilayers of squamous epithelium. Similar findings also reported in other studies (Perniconi et al., 2008;Vuong et al., 2017). Epithelial cells exhibit contact behavior for proper formation of the epithelial monolayer by inhibiting cell proliferation upon contact with neighboring cells. Studies have reported that loss of contact inhibition result in outgrowth of cells from the epithelial monolayer (Mcclatchey and Yap, 2012) similar to the multilayered OSE observed in our study.
The CEE treated groups in the present study also exhibited an increase in the number of theca-interstitial cells in the stroma of the ovaries of all samples. The theca layer of ovarian follicle is a connective tissue surrounding the granulosa cells and responsible for the production of androgens under the influence of gonadotropins stimulation (Fowler et al., 2011;Allen et al., 2016;Richards et al., 2017;Babu et al., 2000;Li et al., 2007). The theca-derived androgens are then converted into estradiol by granulosa cells (Fortune and Armstrong, 1978). which in turn may increase OSE proliferation. The entrapment of OSE cells in inclusion cysts increases the chance of OSE neoplastic transformation, as a consequence of stromal androgen-rich milieu (Rao and Slotman, 1991;Risch, 1998). Consistent with this androgen theory, women with polycystic ovary syndrome (PCOS) have a higher risk of developing ovarian cancer, as a result of anovulation and higher levels of circulating androgen. Recent studies reported that 70% of PCOS cases have elevated testosterone levels (Rosenfield and Ehrmann, 2016). Although the androgen theory may explain the effect of HRT on the OSE, it is conflicting and yet to be proven, and the derivation and roles of cells within the theca are less well defined (Richards et al., 2017;Mizushima and Miyamoto, 2019).
The role of hormonal therapy in the development of ovarian cancer remains unclear, and the mechanism underlying the association of menopausal HRT with ovarian cancer have not yet been fully established. A recent study of heterogeneity demonstrated possible mechanisms whereby activation of various cancer-associated pathways may increase OSE dysplasia as a result of prolonged exposure to estrogen (Vuong et al., 2018). This could be attributed to the fact that exogenously administered estrogens in postmenopausal women may be severalfold higher than endogenous estrogens . Other studies have suggested that high levels of gonadotropins during menopause promote ovarian cancer (Stadel, 1975;Feng et al., 2017). These findings suggest that menopausal HRT may have a role in decreasing the risk of cancer by reducing gonadotropins level. However, these benefits might be overcome by estrogen-induced ovarian cell proliferation (Vuong et a., 2018;Gramer and Welch, 1983). Estrogen stimulates the proliferation of OSE cells whereas progesterone promotes apoptosis of ovarian cells. The weaker risk effect of HRT than estrogen alone may be because progesterone counteracts the proliferative effect of estrogen on ovarian cells (Lau et al., 1999;Syed et al., 2001;Mukerjee et al., 2005). Interactions between genetic, hormonal and environmental factors may impact ovarian cancer risk in an individual patient, but an understanding of the interactions and the order of risk factors is still incomplete.
In conclusion, hormonal replacement therapy may increase the risk of human ovarian cancer as exogenous estradiol accelerates the onset of ovarian cancer in mouse models. Administration of a high-dose of conjugated equine estrogen in our experimental model (CD1 mice) resulted in cellular proliferation and morphological dysplasia of ovarian superficial epithelium. This work supports earlier studies which have shown that estradiol increases epithelial cell susceptibility to tumor initiation. Further studies are needed to better understand the consequences of prolonged exposure to exogenous steroid hormones for disease prevention and to mitigate the risk associated with hormone therapy. Mashmuli, I. Jamshidi-Adgani, and S. Vakilian performed the experiment, data observation and acquisition. I, Amri, I. Kadim S.Khalaf, A. Amri, S, Al-Yaqoobi and K. Al-Riyami analyzed and interpreted the data. A. Amri, I. Kadim A. Abdulziz wrote the paper. All authors reviewed the manuscript, read and approved the final manuscript. The authors declare that they have no conflict of interest.

Funding information
This work was supported by a grant from the Research Council, Sultanate of Oman.

Ethical approval
The use of the animals was approved by the Animal Ethics Committee at University of Nizwa.