Abstract
Previous evidence suggests a link between the ovarian surface epithelium and cortical inclusion cysts and the origin of epithelial ovarian cancer. The presence of 17β-hydroxysteroid dehydrogenase type I, aromatase and estrogen receptor alpha has previously been demonstrated in these epithelial components of the human ovary. The aim of this study was to provide further evidence on these enzymes involved in estrogen biosynthesis and the estrogen receptor in cortical inclusion cysts and the ovarian surface epithelium in the ovaries of women between the ages of 23 and 80. Our results were obtained via immunohistochemistry of formalin-fixed, paraffin-embedded tissue samples from patients who underwent medically indicated oophorectomy. The observed frequencies of cortical inclusion cysts within the ovarian cortex did not reveal significant variation with patient age. The presence of HSD17B1, aromatase and estrogen receptor alpha was observed in more than 60% of the cortical inclusion cysts, with a higher frequency than that of the ovarian surface epithelium. The highest frequency was observed in cortical inclusion cysts with tubal-like epithelium. The estrogen receptor distribution was determined in the ovary stroma, and 17β-hydroxysteroid dehydrogenase type 1 was also identified in the stromal cells. The expression of estrogen receptor alpha in the cortical inclusion cysts was reduced in patients with cervical squamous cell carcinoma. This evidence supports the involvement of estrogen in the biology of the ovarian surface epithelium and cortical inclusion cysts in the ovary.
Introduction
The most common ovarian cancers are those with an epithelial phenotype and can be classified as serous, endometrioid, mucinous or clear cell carcinomas (1). The origin of epithelial ovarian cancer has been attributed to neoplasia generated in the ovarian surface epithelium (OSE) or ovarian inclusion cysts (CICs). Furthermore, the implantation of malignant cells from in situ carcinoma within the fimbriated region of the uterine tubes has also been documented (2, 3, 4).
The human ovary is lined by the OSE, which is a single layer of epithelial cells and is recognizable from the fifth week of embryonic development within the genital ridge. Invagination of this epithelium leads to the formation of the duct of Müller, the precursor of the uterine tubes, uterus and upper region of the vagina (5). In the adult ovary, OSE invaginations result in epithelial inclusion cysts that are found in patients of reproductive age (6, 7). The epithelium of cortical cysts exhibits two histological variants: one is observed as a single uniform layer of cubic or flat cells and the second is characterized by the presence of columnar and ciliated cells, with some mixed CICs displaying combined histological features (8).
The involvement of estrogen in cancer tissue of the reproductive tract has been reported previously (9, 10). Similarly, estrogen receptor alpha (ERα) has been detected in the OSE and the CIC epithelium and aromatase and 17β-hydroxysteroid dehydrogenase type I (HSD17B1) have been demonstrated in the CIC epithelium and surrounding stroma (11, 12). HSD17B1, the first member of the short-chain dehydrogenase/reductase superfamily, catalyzes the oxidoreduction of hydroxyl/keto groups at the C17 position, which is involved in the activation or inactivation of androgens and estrogens. The main activity of HSD17B1 is the reductive conversion of estrone (E1) to 17β-estradiol (E2) (13, 14). Moreover, aromatase, encoded by the CYP19A1 gene, is required for estrogen biosynthesis from androgen precursors (9), which suggests its involvement in estrogen biosynthesis and supports its estrogenic activity.
Considering the estrogen involvement in the biology of the ovarian epithelial components, the expression of HSD17B1, aromatase and ERα in OSE and CIC variants would be relevant. The OSE and CICs are present in the ovaries of patients of reproductive age and postmenopausal patients and may lead to the development of some of the subtypes of epithelial ovarian cancer (15). This study provides evidence of the expression frequency of HSD17B1, aromatase and ERα in the OSE and the epithelial lining of CICs in ovary samples from patients with medically indicated oophorectomy.
Material and methods
Ovary samples
Ovarian tissue samples (n = 116) were obtained from the pathology service of the Hospital Militar de Especialidades de la Mujer y Neonatología and the pathology department of the Instituto Nacional de Cancerología, both of which are located in México City. Ovary samples were selected from archived specimens of patients who had undergone medically indicated oophorectomy. Ovaries that included the cortex region were selected for immunohistochemistry. Patient age, menopausal status and associated conditions were obtained from the patient’s medical records. The study protocol was reviewed and approved by the ethics committees of both hospitals (310-18 and 019/060/OMI) and the Facultad de Medicina, Universidad Nacional Autónoma de México (FM/DI/114/2022). The patients were informed of and consented to the use of the tissue samples for this study.
Immunohistochemistry
Histological sections (3 μm thick) from formalin-fixed paraffin-embedded ovarian samples were processed as previously described (16). Epitope recovery was performed using citrate buffer (pH 6.0) with heat treatment. Hydrogen peroxide was used to block endogenous peroxidase and nonspecific binding was prevented using equine serum. The primary antibodies used were the HSD17B1 antibody (cat. GTX12312; GeneTex, USA), aromatase antibody H4 (cat. OASA02666; Aviva Systems Biology, USA) and Erα HC20 antibody (cat. sc-543; Santa Cruz Biotechnology, Inc., USA). The secondary antibody was Mach2 anti-rabbit horseradish peroxidase (HRP) (cat. RHRP520; Biocare Medical, USA). Staining was performed using a Metal Enhanced DAB Substrate Kit (cat. 34065; Thermo Fisher Scientific Inc., USA). The nuclei were counterstained with Gill’s hematoxylin.
Microscopy observations of hematoxylin–eosin stained sections were performed to describe the CIC variants and identify regions with the conserved surface epithelium and epithelium in ovarian surface clefts. Detection of positive immunoreactivity for HSD17B1, aromatase and ERα in blinded samples was performed by two independent observers (EP and CM). At least 10% of labeled epithelial cells were required for a sample to be considered positive. HSD17B1 and aromatase were labeled positive when observed in the cytoplasm, while ERα was only considered positive when located in the nucleus.
Statistical analysis
Descriptive values were generated from the database using patient age, associated disease and positive reactions for each of the two enzymes and ERα in their corresponding locations of the epithelial components. Differences between groups and associations were analyzed using the chi-square test or Fisher’s exact test, according to the requirements. Paired frequencies were evaluated by comparing proportions and the Z-value. P-value ≤0.05 was considered significant.
Results
The median age of the study population was 49 years, ranging from 23 to 80 years, and 80% of cases were between the ages of 40 and 60. The frequency of CICs found in the ovaries revealed no association with the patient age: 24/59 (41%) in patients aged ≤49 years vs 26/48 (54%) in patients aged ≥50 years (P > 0.05).
CIC variants could be identified in the ovarian cortex by their epithelial characteristics: i) CICs lined by uniform cuboid or flat-shaped epithelial cells, resembling the OSE; hereafter referred to as CICose (Fig. 1A). ii) Cysts with an epithelium that resembles that of the uterine tube (CICtube), where the following cell types could be recognized: ciliated cells, columnar secretory cells (characterized by cytoplasmic bulbous tips), peg or intercalated cells and migrating immune cells (Fig. 1B, C, D). iii) Cysts with mixed epithelium (CICmix) covered with an OSE-like epithelium and exhibiting a discontinuous region of tuba-like epithelium (Fig. 2A), regions with secretory cells without ciliated cells, single ciliated cells with a cuboid or a flat shape intercalated with OSE-like epithelium (Fig. 1C), and occasionally regions with columnar mucinous cells with a basal nucleus and goblet cells could be observed (Fig. 2B).
Photomicrograph of a cortical inclusion cyst (CIC) in the ovarian cortex. (A, B, C, D) Immunohistochemistry for ERα with a hematoxylin counterstain. (A) CIC lined with the OSE. (B) CIC lined with tubal-like epithelium. (C) Flat epithelium of the OSE-like cyst with isolated ciliated cells (arrows). (D) CIC with tubal-like epithelium, and ERα labeling in the stromal cell nucleus is observed surrounding the cyst (left). The stroma near the surface (right) lacks ERα labeling. Scale bars = 50 μm.
Citation: Endocrine Connections 14, 4; 10.1530/EC-24-0643
Micrograph of sections of a CIC lined with mixed epithelium stained with hematoxylin and eosin. (A) CIC with flat epithelium displays regions of ciliated cells. (B) Transition from flat to mucinous epithelia is observed in CICmix. Sections of the same cortical inclusion cyst (CIC) lined with tubal-like epithelium (CICtube) show immunoreactivity for: (C) 17β-hydroxysteroid dehydrogenase type 1 (HSD17B1), (D) aromatase and (E) ERα. The labeling for aromatase is lighter than that for HSD17B1 and ERα reactivity. Scale bar = 50 μm.
Citation: Endocrine Connections 14, 4; 10.1530/EC-24-0643
The size of CICs ranged from small cysts (100–300 μm in diameter) close to the ovarian surface to large cysts (≥1.0 mm diameter) found deeper within the cortex. CICmix was found as large cysts. In the evaluated ovarian samples, the frequency of CICose was 51/116 (44%), CICtube was observed in 54/116 (46%) and CICmix was identified in 28/116 (24%) of the cases.
Immunohistochemistry
Immunoreactivity for HSD17B1 in the epithelium was found in 92% of the CIC samples, while aromatase exhibited a positive reaction in 85% of the samples and ERα positivity was observed in 68% of the cases. Compared with the immunoreactivity frequencies in OSE cells (HSD17B1 73%, aromatase 65% and ERα 46%), CIC epithelium exhibited a significantly higher frequency of HSD17B1, aromatase and ERα (Table 1). No difference in the frequencies of expression of HSD17B1, aromatase and ERα was observed in the OSE and CICs related to patient age (data not shown). An association of the expression of ERα and aromatase was observed within the OSE and CICs (P ≤ 0.05). The association of HSD17B1 with ERα is significant considering the location of the OSE and CICs.
Frequency of positive immunoreactivity of 17β-hydroxysteroid dehydrogenase type 1 (HSD17B1), aromatase and ERα in the ovarian surface epithelium (OSE) and cortical inclusion cyst (CIC).
| HSD17B1 | Aromatase | ERα | |
|---|---|---|---|
| OSE | 61/84 (73%) † | 51/79 (65%)* | 38/82 (46%)* |
| CIC | 74/80 (92%) | 64/75 (85%) | 48/71 (68%) |
ERα, estrogen receptor alpha. Values are expressed as number of cases (percentage).
P ≤ 0.01.
P ≤ 0.001 from chi-square test.
Regarding the medical indications for oophorectomy, the expression of HSD17B1, aromatase and ERα in the ovarian cortex was documented along with the associated diagnoses (Table 2). In patients diagnosed with cervical squamous cell carcinoma (CSCC), a positive immunoreaction for ERα was observed in 7/23 (30%) of CICs. Moreover, ERα was also observed in 11/12 CIC samples (92%) from patients with endometrial carcinoma and in 13/14 CIC samples (93%) from patients with leiomyomas. The reduction in the ERα frequency in the CSCC group is significant (P < 0.001).
Frequency of positive immunoreactivity of 17β-hydroxysteroid dehydrogenase type 1 (HSD17B1), aromatase and ERα in cortical inclusion cyst (CIC) and ovarian surface epithelium (OSE) related to medically indicated oophorectomy.
| HSD17B1 | Aromatase | ERα | ||||
|---|---|---|---|---|---|---|
| CIC | OSE | CIC | OSE | CIC | OSE | |
| Squamous cervical cancer | 17/18 (94) | 15/20 (75)a | 13/18 (72) | 11/19 (58) | 7/23 (30)a | 3/16 (19) |
| Endometrial cancer | 13/15 (87) | 4/12 (33)b | 7/10 (70) | 4/11 (36) | 11/12 (92)b | 4/15 (26) |
| Uterine leiomyoma | 18/18 (100) | 17/20 (85)a | 20/21 (95) | 13/19 (68) | 13/14 (93)b | 10/18 (56) |
| Other cancers | 10/11 (91) | 10/12 (83)a | 9/10 (90) | 9/11 (82) | 5/7 (71)b | 5/10 (50) |
| Other pathologies | 9/10 (90) | 6/9 (67)a | 8/9 (89) | 6/9 (67) | 8/9 (89)b | 7/12 (58) |
| P-value | 0.633 | 0.023 | 0.226 | 0.242 | <0.001 | 0.093 |
ERα, estrogen receptor alpha. Values are expressed as number of cases (percentage). Values with different letters indicate statistical significance, Bold indicates statistical significance. P-values are shown within each column of data. Chi-square test and proportions comparison, ‘Z’ value.
Immunohistochemistry in CIC variants
Positive HSD17B1, aromatase and ERα immunoreactivity was evaluated separately in CICose and CICtube. In CICmix, reactivity was classified based on the corresponding types: Regions with the columnar epithelium without ciliated cells or with ciliated cells alone were assigned to CICtube.
The frequency of positive reactions in CICose epithelium was 37/47 (79%) for HSD17B1, 22/31 (71%) for aromatase and 23/51 (45%) for ERα. On the other hand, CICtube presented with positive reactions for HSD17B1 in 43/44 (98%) cases, aromatase in 38/39 (97%) cases and 37/53 (70%) cases of ERα. The immunoreactivity frequency of HSD17B1, aromatase and ERα was significantly increased in CICtube compared with that of CICose (Table 3).
Frequency of positive immunoreactivity of 17β-hydroxysteroid dehydrogenase type 1 (HSD17B1), aromatase and ERα in variation of cortical inclusion cyst epithelium with OSE-like (CICose) or epithelium of the uterine tube-like (CICtube).
| HSD17B1 | Aromatase | ERα | |
|---|---|---|---|
| CICose | 37/47 (79%)* | 22/31 (71%)* | 23/51 (45%)* |
| CICtube | 43/44 (98%) | 38/39 (97%) | 37/53 (70%) |
OSE, ovarian surface epithelium; ERα, estrogen receptor alpha. Values are expressed as number of cases (percentage).
P ≤ 0.01. Chi-square test.
There was a positive ERα immunoreaction in the ovarian cortex stroma surrounding the cyst (Fig. 1B, C, D). In CICmix with tubal-like epithelium regions, the positivity of the stroma was increased compared with the OSE-like region of the same cyst. Moreover, the delimited regions of the stroma with positive ERα cell nuclei were observed as regions of CICtube and frequently associated with signs of epithelial hyperactivity (Fig. 1D). These stromal regions resembled the uterine tuba in their nuclear density and intercellular fiber distribution. HSD17B1 immunoreactivity was also observed in the stroma and occasionally surrounded the CICs.
Discussion
This study contributes to the characterization of CICs in the human ovary associated with the presence of steroid hormone-metabolizing enzymes in the ovarian cortex epithelial compartments of perimenopausal and menopausal women. The expression of HSD17B1 is relevant because of its role in the conversion of estrone to 17β-estradiol, and aromatase is required for estrogen biosynthesis.
The presence of variations in CICs within the ovarian cortex has been described previously (6, 7, 8); however, the frequency of the presence of the studied enzymes and estrogen receptors in the different CIC types has not been investigated. This study provides evidence that the expression of enzymes and estrogen receptors in the CICose epithelium is similar to that of the OSE, while CICtube exhibits an increased frequency of enzymes related to 17β-estradiol biosynthesis and the presence of ERα in the epithelium. Moreover, the presence of HSD17B1, aromatase and ERα in the regions of tubal-like epithelium and the surrounding stroma indicates similarities between the CICtube and the fallopian tubes, which is in alignment with previous descriptions of similar histology in both epithelia (6, 7). The presence of aromatase and HSD17B1 supports the possibility of 17β-estradiol biosynthesis in the epithelium and the associated ERα expression in the CICtube suggests estrogenic activity. The importance of this estrogenic microenvironment in the growth of the epithelium and papilla formation in the cyst and its eventual carcinogenesis has been previously proposed and warrants further investigation (9).
CICmix displays both OSE- and tubal-like epithelia and it contains isolated ciliated and secretory cells within the cyst wall regions and, occasionally, columnar mucinous cells as well. These observations in a particular cyst support the hypothesis of epithelial metaplasia within cortical cysts. In addition, the presence of delimited regions of stromal cells expressing nuclear ERα, which resembles the fallopian tube stroma, needs to be included in the metaplasia hypothesis. On the other hand, the hypothesis suggests the potential incorporation of tubal cells into the ovarian cortex to explain the origin of CICs (i.e., CICtube), and this invasion is not limited to epithelial cells. It is more likely that fragments of tubal tissue are implanted into the ovarian cortex because of the adhesion of fimbriae of uterine tubes onto the ovarian surface. Moreover, in superficial cysts near the OSE, ERα expression in the nucleus of adjacent stromal cells is observed within the deeper region of the cyst and not close to the surface (Fig. 1D). A similar observation on variations in stroma morphology has been previously reported to be associated with regions with epithelial dysplasia (17). Our observations of ERα distribution in the stroma corroborate regional variations and suggest the involvement of estrogens in changes at the epithelial level of CICs.
The observed changes in the ERα expression within ovarian CIC epithelia with the associated pathology resulted in a reduction in the frequency of ERα expression in patients with CSCC compared with those with endometrial cancer and uterine leiomyomas. There are variations in estrogen involvement and estrogen receptor expression in these pathologies (18, 19). The reduced ERα expression observed in cervical cancer (20, 21) could be associated with human papillomavirus infection (22), and the simultaneous reduction present in the CIC epithelium requires further investigation to elucidate a common regulatory mechanism in both reproductive tissues. Moreover, the present observations need to be confirmed in a cohort with a large number of patients.
We conclude that CIC variants were observed in the ovarian cortex without changes related to the age of the patients included in the study. The presence of HSD17B1, aromatase and ERα was observed in more than 60% of the CIC epithelium, with a higher frequency than that observed in the OSE. The highest immunoreactivity frequency was found in CICtube. HSD17B1 and ERα were also observed in stromal cells, with a distinct ERα distribution pattern. The ERα expression in CICs was reduced in patients with CSCC.
Declaration of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the work reported.
Funding
The present work was supported by a grant from DGAPA-PAPIIT, IN223823, to CM and a grant from DGAPA-PAPIIT, IN208822, to EP.
Acknowledgments
We are deeply grateful to Mrs Angélica Caballero and MC María José Gomora for their technical assistance.
References
- 1↑
Prat J . Ovarian carcinomas: five distinct diseases with different origins, genetic alterations, and clinicopathological features. Virchows Arch 2012 460 237–249. (https://doi.org/10.1007/s00428-012-1203-5)
- 2↑
Auersperg N . The origin of ovarian carcinomas: a unifying hypothesis. Int J Gynecol Pathol 2011 30 12–21. (https://doi.org/10.1097/PGP.0b013e3181f45f3e)
- 3↑
Kurman RJ & Shih IM . The dualistic model of ovarian carcinogenesis: revisited, revised, and expanded. Am J Pathol 2016 186 733–747. (https://doi.org/10.1016/j.ajpath.2015.11.011)
- 4↑
Lim D & Oliva E . Precursors and pathogenesis of ovarian carcinoma. Pathology 2013 45 229–242. (https://doi.org/10.1097/PAT.0b013e32835f2264)
- 5↑
Cunha GR , Kurita T , Cao M , et al. Tissue interactions and estrogenic response during human female fetal reproductive tract development. Differentiation 2018 101 39–45. (https://doi.org/10.1016/j.diff.2018.04.002)
- 6↑
Scully RE . Pathology of ovarian cancer precursors. J Cell Biochem Suppl 1995 59 208–218. (https://doi.org/10.1002/jcb.240590928)
- 7↑
Auersperg N , Wong AS , Choi KC , et al. Ovarian surface epithelium: biology, endocrinology, and pathology. Endocr Rev 2001 22 255–288. (https://doi.org/10.1210/edrv.22.2.0422)
- 8↑
Banet N & Kurman RJ . Two types of ovarian cortical inclusion cysts: proposed origin and possible role in ovarian serous carcinogenesis. Int J Gynecol Pathol 2015 34 3–8. (https://doi.org/10.1097/PGP.0000000000000120)
- 9↑
Cunat S , Rabenoelina F , Daurès JP , et al. Aromatase expression in ovarian epithelial cancers. J Steroid Biochem Mol Biol 2005 93 15–24. (https://doi.org/10.1016/j.jsbmb.2004.10.021)
- 10↑
Kamal A , Tempest N , Parkes C , et al. Hormones and endometrial carcinogenesis. Horm Mol Biol Clin Investig 2016 25 129–148. (https://doi.org/10.1515/hmbci-2016-0005)
- 11↑
Piek JM , Verheijen RH , Menko FH , et al. Expression of differentiation and proliferation related proteins in epithelium of prophylactically removed ovaries from women with a hereditary female adnexal cancer predisposition. Histopathology 2003 43 26–32. (https://doi.org/10.1046/j.1365-2559.2003.01654.x)
- 12↑
Brodowska A , Brodowski J , Laszczyńska M , et al. Immunoexpression of aromatase cytochrome P450 and 17β-hydroxysteroid dehydrogenase in women’s ovaries after menopause. J Ovarian Res 2014 7 52. (https://doi.org/10.1186/1757-2215-7-52)
- 13↑
Labrie F , Luu-The V , Lin SX , et al. The key role of 17β-hydroxysteroid dehydrogenases in sex steroid biology. Steroids 1997 62 148–158. (https://doi.org/10.1016/s0039-128x(96)00174-2)
- 14↑
Lukacik P , Kavanagh KL & Oppermann U . Structure and function of human 17β-hydroxysteroid dehydrogenases. Mol Cell Endocrinol 2006 248 61–71. (https://doi.org/10.1016/j.mce.2005.12.007)
- 15↑
Auersperg N . Ovarian surface epithelium as a source of ovarian cancers: unwarranted speculation or evidence-based hypothesis? Gynecol Oncol 2013 130 246–251. (https://doi.org/10.1016/j.ygyno.2013.03.021)
- 16↑
Calvillo-Robledo A , Pedernera E , Morales-Vásquez F , et al. Simultaneous expression of steroid sulfatase and androgen receptor reduced overall survival of patients with epithelial ovarian tumors. J Ovarian Res 2021 14 98. (https://doi.org/10.1186/s13048-021-00840-x)
- 17↑
Scully RE . Early de novo ovarian cancer and cancer developing in benign ovarian lesions. Int J Gynaecol Obstet 1995 49 (Supplement) S9–S15. (https://doi.org/10.1016/0020-7292(95)02404-z)
- 18↑
Englund K , Blanck A , Gustavsson I , et al. Sex steroid receptors in human myometrium and fibroids: changes during the menstrual cycle and gonadotropin-releasing hormone treatment. J Clin Endocrinol Metab 1998 83 4092–4096. (https://doi.org/10.1210/jcem.83.11.5287)
- 19↑
Wang C , Tran DA , Fu MZ , et al. Estrogen receptor, progesterone receptor, and HER2 receptor markers in endometrial cancer. J Cancer 2020 11 1693–1701. (https://doi.org/10.7150/jca.41943)
- 20↑
Nonogaki H , Fujii S , Konishi I , et al. Estrogen receptor localization in normal and neoplastic epithelium of the uterine cervix. Cancer 1990 66 2620–2627. (https://doi.org/10.1002/1097-0142(19901215)66:12<2620:aid-cncr2820661226>3.0.co;2-s)
- 21↑
Hong MK , Wang JH , Su CC , et al. Expression of estrogen and progesterone receptor in tumor stroma predicts favorable prognosis of cervical squamous cell carcinoma. Int J Gynecol Cancer 2017 27 1247–1255. (https://doi.org/10.1097/IGC.0000000000001004)
- 22↑
Bekkers RL , van der Avoort IA , Melchers WJ , et al. Down regulation of estrogen receptor expression is an early event in human papillomavirus infected cervical dysplasia. Eur J Gynaecol Oncol 2005 26 376–382.
