Genomic alterations in ovarian endometriosis and subsequently diagnosed ovarian carcinoma

STUDY QUESTION

Can the alleged association between ovarian endometriosis and ovarian carcinoma be substantiated by genetic analysis of endometriosis diagnosed prior to the onset of the carcinoma?

SUMMARY ANSWER

The data suggest that ovarian carcinoma does not originate from ovarian endometriosis with a cancer-like genetic profile; however, a common precursor is probable.

WHAT IS KNOWN ALREADY

Endometriosis has been implicated as a precursor of ovarian carcinoma based on epidemiologic studies and the discovery of common driver mutations in synchronous disease at the time of surgery. Endometrioid ovarian carcinoma and clear cell ovarian carcinoma are the most common endometriosis-associated ovarian carcinomas (EAOCs).

STUDY DESIGN, SIZE, DURATION

The pathology biobanks of two university hospitals in Sweden were scrutinized to identify women with surgically removed endometrioma who subsequently developed ovarian carcinoma (1998-2016). Only 45 archival cases with EAOC and previous endometriosis were identified and after a careful pathology review, 25 cases were excluded due to reclassification into non-EAOC (n = 9) or because ovarian endometriosis could not be confirmed (n = 16). Further cases were excluded due to insufficient endometriosis tissue or poor DNA quality in either the endometriosis, carcinoma, or normal tissue (n = 9). Finally 11 cases had satisfactory DNA from all three locations and were eligible for further analysis.

PARTICIPANTS/MATERIALS, SETTING, METHODS

Epithelial cells were collected from formalin-fixed and paraffin-embedded (FFPE) sections by laser capture microdissection (endometrioma n = 11) or macrodissection (carcinoma n = 11) and DNA was extracted. Normal tissue from FFPE sections (n = 5) or blood samples collected at cancer diagnosis (n = 6) were used as the germline controls for each included patient. Whole-exome sequencing was performed (n = 33 samples). Somatic variants (single-nucleotide variants, indels, and copy number alterations) were characterized, and mutational signatures and kataegis were assessed. Microsatellite instability and mismatch repair status were confirmed with PCR and immunohistochemistry, respectively.

MAIN RESULTS AND THE ROLE OF CHANCE

The median age for endometriosis surgery was 42 years, and 54 years for the subsequent ovarian carcinoma diagnosis. The median time between the endometriosis and ovarian carcinoma was 10 (7-30) years. The data showed that all paired samples harbored one or more shared somatic mutations. Non-silent mutations in cancer-associated genes were frequent in endometriosis; however, the same mutations were never observed in subsequent carcinomas. The degree of clonal dominance, demonstrated by variant allele frequency, showed a positive correlation with the time to cancer diagnosis (Spearman's rho 0.853, P < 0.001). Mutations in genes associated with immune escape were the most conserved between paired samples, and regions harboring these genes were frequently affected by copy number alterations in both sample types. Mutational burdens and mutation signatures suggested faulty DNA repair mechanisms in all cases.

LARGE SCALE DATA

Datasets are available in the supplementary tables.

LIMITATIONS, REASONS FOR CAUTION

Even though we located several thousands of surgically removed endometriomas between 1998 and 2016, only 45 paired samples were identified and even fewer, 11 cases, were eligible for sequencing. The observed high level of intra- and inter-heterogeneity in both groups (endometrioma and carcinoma) argues for further studies of the alleged genetic association.

WIDER IMPLICATIONS OF THE FINDINGS

The observation of shared somatic mutations in all paired samples supports a common cellular origin for ovarian endometriosis and ovarian carcinoma. However, contradicting previous conclusions, our data suggest that cancer-associated mutations in endometriosis years prior to the carcinoma were not directly associated with the malignant transformation. Rather, a resilient ovarian endometriosis may delay tumorigenesis. Furthermore, the data indicate that genetic alterations affecting the immune response are early and significant events.

STUDY FUNDING/COMPETING INTEREST(S)

The present work has been funded by the Sjöberg Foundation (2021-01145 to K.S.; 2022-01-11:4 to A.S.), Swedish state under the agreement between the Swedish government and the county councils, the ALF-agreement (965552 to K.S.; 40615 to I.H.; 965065 to A.S.), Swedish Cancer Society (21-1848 to K.S.; 21-1684 to I.H.; 22-2080 to A.S.), BioCARE-A Strategic Research Area at Lund University (I.H. and S.W.-F.), Mrs Berta Kamprad's Cancer Foundation (FBKS-2019-28, I.H.), Cancer and Allergy Foundation (10381, I.H.), Region Västra Götaland (A.S.), Sweden's Innovation Agency (2020-04141, A.S.), Swedish Research Council (2021-01008, A.S.), Roche in collaboration with the Swedish Society of Gynecological Oncology (S.W.-F.), Assar Gabrielsson Foundation (FB19-86, C.M.), and the Lena Wäpplings Foundation (C.M.). A.S. declares stock ownership and is also a board member in Tulebovaasta, SiMSen Diagnostics, and Iscaff Pharma. A.S. has also received travel support from EMBL, Precision Medicine Forum, SLAS, and bioMCC. The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Collagens in the human ovary and their changes in the perifollicular stroma during ovulation

BACKGROUND

Remodeling of the collagens around the follicle is a major event in ovulation. The aim of the present study was to investigate the distribution of collagen I, III, and IV in the human ovary.

METHODS

Biopsies of the perifollicular stroma were obtained at sterilization during the preovulatory phase (follicle size >14 mm) or at any of three intervals (12-18 h after human chorionic gonadotrophin: early ovulatory phase; >18-24 h: late ovulatory phase; 44-77 h: postovulatory phase) after human chorionic gonadotrophin. Excised dominant follicles and whole ovarian sections were also obtained. Immunohistochemistry using antibodies against collagen I, III, IV, vimentin, and CD 45 was performed.

RESULTS AND CONCLUSIONS

Collagens I and III were distributed in concentric layers in the capsular stroma with bundles of collagens connecting these layers to form a mesh. Collagen I was present in larger quantities in the outer layers and collagen III showed the inverse distribution. In the theca, collagen I was present in the externa and collagen III in the entire layer. The staining intensity of collagens I and III in the perifollicular stroma decreased from the preovulatory stage. Collagen IV was present in the basal lamina separating granulosa and theca cells. This study shows that collagen I and III are abundant in and around the ovulating human follicle with typical patterns of distribution. Collagen IV is present in the basal membrane that separates the granulosa from the theca cells. Taking into account the abundance of collagens in the follicular wall and their specific localization, major site-directed degradation of collagens seems to be necessary for follicular rupture to occur.

Wnt-signalling pathway in ovarian epithelial tumours: increased expression of beta-catenin and GSK3beta

Beta-catenin is involved in both cell–cell adhesion and in transcriptional regulation by the Wingless/Wnt signalling pathway. Alterations of components of this pathway have been suggested to play a central role in tumorigenesis. The present study investigated, by immunohistochemistry and immunoblotting, the protein expression and localisation of β-catenin, adenomatous polyposis coli (APC), glycogen synthase kinase 3β (GSK3β) and lymphocyte enhancer factor-1 (Lef-1) in normal human ovaries and in epithelial ovarian tumours in vivo and in vitro. Immortalised human ovarian surface epithelium and ovarian cancer cell cells (OVCAR-3) expressed β-catenin, APC, GSK3β and Lef-1. Nuclear staining of β-catenin and Lef-1 were demonstrated only in OVCAR-3 cells. There were significant increases of β-catenin and GSK3β, while APC was reduced in ovarian cancer compared to the normal ovary. Beta-catenin and Lef-1 were coimmunoprecipitated in ovarian tumours, but not in the normal ovary. Nuclear localisation of β-catenin or Lef-1 could not be demonstrated in the normal ovary or in the ovarian tumours. The absence of nuclear localisation of β-catenin could be due to an increased binding to the cadherin–α-catenin cell adhesion complex. In fact, we have earlier reported an increased expression of E-cadherin in ovarian adenocarcinomas. In summary, this study demonstrates an increase in the expression of components of the Wingless/Wnt pathway in malignant ovarian tumours. The increase suggests a role for this signalling pathway in cell transformation and in tumour progression. However, it remains to be demonstrated whether it is an increased participation of β-catenin in transcriptional regulation, or in the stabilisation of cellular integrity, or both, that is the crucial event in ovarian tumorigenesis.

Production of steroids by human ovarian surface epithelial cells in culture: possible role of progesterone as growth inhibitor

OBJECTIVE

The purpose was to investigate whether normal ovarian surface epithelial cells, harvested from premenopausal and postmenopausal women, are capable of steroid production, and to evaluate effects of estradiol and progesterone on growth regulation of such cells.

METHODS

Ovarian surface epithelial cells were obtained by brushing of the ovarian surface of 9 premenopausal and 10 postmenopausal women undergoing surgery for benign gynecological diseases. The conditioned media after culture, with and without addition of FSH and LH, were analyzed for estradiol and progesterone. The proliferative effects of the steroids were analyzed using two different culture models, nonconfluent cells and confluent cells, and two different detection methods, [(3)H]thymidine incorporation and a colorimetric method assaying cell number.

RESULTS

The normal ovarian surface epithelial cells were found to secrete both estradiol and progesterone, a production that was not regulated by FSH or LH. Addition of steroids to the cultured cells did not induce any overall significant growth effects. However, progesterone significantly inhibited the growth of ovarian surface epithelial cells from three of the patients. Enhanced thymidine incorporation was observed in the presence of the progesterone receptor antagonist Org 31710 in the nonconfluent cultures of cells from postmenopausal women, but no effect of an estrogen receptor antagonist was observed.

CONCLUSIONS

The normal ovarian surface epithelium is capable of steroid production, which is also often observed in tissue from ovarian epithelial tumors. Progesterone appeared to be a negative regulator of ovarian surface epithelial growth, while estradiol had no effect.

Higher levels of soluble E-cadherin in cyst fluid from malignant ovarian tumours than in benign cysts

A major diagnostic dilemma in the clinical gynaecological oncology setting is to preoperatively determine whether a complex ovarian mass is benign or malignant. The cell-cell adhesion molecule E-cadherin has previously been localised in biopsies from both benign and malignant epithelial ovarian tumours. In this study, soluble E-cadherin levels was measured with ELISA-technique in peripheral blood, ascites and cystic fluids from patients (n = 33) undergoing surgery for ovarian cystic masses. The levels of soluble E-cadherin were significantly higher in cystic fluid from cystadenocarcinomas (p < 0.001) and borderline tumours (p < 0.05) as compared to cystic fluid from cystadenomas. In ascites fluid and peripheral blood no significant differences were seen. However, ratios of cystic fluid/peripheral blood levels were significantly higher in cystadenocarcinoma (p < 0.001) and borderline tumours (p < 0.05) as compared to benign tumours. In conclusion, measurements of soluble E-cadherin in cystic fluid from patients presenting with complex ovarian masses may be beneficial in increasing the accuracy of preoperative diagnosis.

Diverse effects of FSH and LH on proliferation of human ovarian surface epithelial cells

The purpose of this study was to evaluate the effects of FSH and LH on growth regulation of normal ovarian surface epithelial (OSE) cells harvested from both premenopausal and postmenopausal women. Ovarian surface epithelial cells were obtained through brushing of the ovarian surface during surgery. FSH and LH were added to the OSE cultures and the proliferative effects were analysed using two different culture models, non-confluent and confluent cells, and two different detection methods, [(3)H]thymidine incorporation and a colorimetric cell number assay. FSH lowered the OSE proliferation under non-confluent conditions (10-27%), and the inhibitory effect was most pronounced among cells from postmenopausal women (P: < 0.01). In the confluent model only cells from postmenopausal women showed significantly (P: < 0.05) decreased proliferation. No effects of LH on OSE cells were detected. The unexpected results of an anti-proliferative effect of FSH on OSE, and the absence of effect by LH, do not support the theory that gonadotrophins are directly involved in ovarian carcinogenesis through an enhanced proliferation of OSE cells.

Increased expression of the transcription factors CCAAT-enhancer binding protein-beta (C/EBBeta) and C/EBzeta (CHOP) correlate with invasiveness of human colorectal cancer

Regulation of cell differentiation is most often impaired in malignant tumors and may represent a key mechanism for the progression of the disease. CCAAT-enhancer binding protein (C/EBP) is a family of transcription factors involved in the regulation of embryonic gut development in rodents, which has also been detected in various malignancies, e.g., liposarcomas and breast and ovarian epithelial tumors. We studied the relationship between C/EBP and tumor histology (Duke's invasive stage and pathological grade) in colorectal cancer. Immunoblotting techniques were used on microdissected fresh frozen tumor specimens, and expression of C/EBPalpha, C/EBPbeta and C/EBPzeta (CHOP) was analyzed in addition to that of the cell-cycle regulator p53 and the proliferation marker PCNA. Expression of C/EBPbeta (LAP isoforms) was markedly increased in all tumors compared with normal colon mucosa. Although the inter-patient variability was large, we found that LIP, the isoform of C/EBPbeta known to inhibit transcription, was expressed at higher levels in Duke's stage B tumors compared with Duke's stage A, whereas Duke's C tumors had the lowest LIP expression. A similar relationship was seen for CHOP. The cell-cycle regulator gene p53 was the only factor that clearly correlated with pathological grade: a decrease in p53 expression was demonstrated. Our data suggest that genetic and cellular events involving C/EBPbeta and CHOP are important for tumor invasion and that these events do not appear to be related to the pathological grade of the tumor.

E-cadherin-catenin complex in the rat ovary: cell-specific expression during folliculogenesis and luteal formation

The cadherins and their cytoplasmic counterparts, the catenins, form the adherens junctions, which are of importance for tissue integrity and barrier functions. The development and maturation of the ovarian follicle is characterized by structural changes, which require altered expression or function of the components involved in cell-cell contacts. The present study examined the cell-specific localization and temporal expression of epithelial cadherin (E-cadherin) and alpha- and beta-catenin during follicular development, ovulation and corpus luteum formation in the immature gonadotrophin- and oestrogen-stimulated rat ovary. Immunohistochemistry and immunoblotting demonstrated the expression of E-cadherin in theca and interstitial cells of immature ovaries before and after injection of equine chorionic gonadotrophin (eCG). E-cadherin was not detected in granulosa cells, except in the preantral follicles located to the inner region of the ovary. The content of E-cadherin in theca and interstitial cells decreased after an ovulatory dose of hCG. Granulosa cells of apoptotic follicles did not express E-cadherin. Oestrogen treatment (diethylstilboestrol) of immature rats for up to 3 days did not result in a measurable expression of E-cadherin in granulosa cells. alpha- and beta-catenin were expressed in all ovarian compartments. The concentration of beta-catenin was constant during the follicular phase, whereas the content of alpha-catenin decreased in granulosa cells after treatment with diethylstilboestrol or hCG. The expression of alpha-catenin was also reduced in theca and interstitial cells after hCG. alpha- and beta-catenin were present in most ovarian cells at all stages of folliculogenesis. Therefore, the catenins have the potential to associate with different members of the cadherin family and to participate in the regulation of cytoskeletal structures and intracellular signalling. The restricted expression of E-cadherin in granulosa cells of preantral follicles indicates a role in the recruitment of these follicles to subsequent cycles. The specific decrease of alpha-catenin in granulosa cells and the reduction of both alpha-catenin and E-cadherin in theca cells of ovulatory follicles might reflect some of the molecular changes in cell-cell adhesion associated with ovulation and luteinization.

E-cadherin-catenin complex in the rat ovary: cell-specific expression during folliculogenesis and luteal formation

The cadherins and their cytoplasmic counterparts, the catenins, form the adherens junctions, which are of importance for tissue integrity and barrier functions. The development and maturation of the ovarian follicle is characterized by structural changes, which require altered expression or function of the components involved in cell-cell contacts. The present study examined the cell-specific localization and temporal expression of epithelial cadherin (E-cadherin) and alpha- and beta-catenin during follicular development, ovulation and corpus luteum formation in the immature gonadotrophin- and oestrogen-stimulated rat ovary. Immunohistochemistry and immunoblotting demonstrated the expression of E-cadherin in theca and interstitial cells of immature ovaries before and after injection of equine chorionic gonadotrophin (eCG). E-cadherin was not detected in granulosa cells, except in the preantral follicles located to the inner region of the ovary. The content of E-cadherin in theca and interstitial cells decreased after an ovulatory dose of hCG. Granulosa cells of apoptotic follicles did not express E-cadherin. Oestrogen treatment (diethylstilboestrol) of immature rats for up to 3 days did not result in a measurable expression of E-cadherin in granulosa cells. alpha- and beta-catenin were expressed in all ovarian compartments. The concentration of beta-catenin was constant during the follicular phase, whereas the content of alpha-catenin decreased in granulosa cells after treatment with diethylstilboestrol or hCG. The expression of alpha-catenin was also reduced in theca and interstitial cells after hCG. alpha- and beta-catenin were present in most ovarian cells at all stages of folliculogenesis. Therefore, the catenins have the potential to associate with different members of the cadherin family and to participate in the regulation of cytoskeletal structures and intracellular signalling. The restricted expression of E-cadherin in granulosa cells of preantral follicles indicates a role in the recruitment of these follicles to subsequent cycles. The specific decrease of alpha-catenin in granulosa cells and the reduction of both alpha-catenin and E-cadherin in theca cells of ovulatory follicles might reflect some of the molecular changes in cell-cell adhesion associated with ovulation and luteinization.

The expression of CCAAT/enhancer binding protein (C/EBP) in the human ovary in vivo: specific increase in C/EBPbeta during epithelial tumour progression

The CCAAT/enhancer binding protein (C/EBP) family of transcription factors is involved in metabolism and differentiation of cells, especially in rodent liver cells and adipocytes. Their roles in vivo and in particular during pathophysiological conditions in humans are largely unknown. We have investigated the presence of C/EBPalpha, -beta, -delta and -zeta in normal ovaries and in epithelial ovarian tumours of different stages. Immunohistochemical experiments demonstrated that C/EBPalpha and C/EBPbeta were preferentially expressed in epithelial/tumour cells irrespective of stage or grade of the tumour. C/EBPbeta was located in the nuclei of the cells, in contrast to C/EBPalpha, which was present only in the cytoplasm of these cells. The nuclear localization of C/EBPbeta indicates an active role of this transcription factor in tumour cells, whereas the cytoplasmic distribution suggests a more passive function of C/EBPalpha. C/EBPdelta and -zeta demonstrated a more diverse distribution with predominant localization to epithelial cells, but stromal distribution was also noted. The intracellular distribution was confined to both the nucleus and the cytoplasm for C/EBPdelta and -zeta. Western blotting demonstrated that C/EBPalpha, -beta, -delta and -zeta were present in a majority of the samples. The amount of C/EBPbeta increased markedly with malignancy, i.e. with degree of dedifferentiation, while the other members of the C/EBP family displayed a more constant expression level. These results demonstrate an association between the expression of members of the C/EBP family and the formation of epithelial ovarian tumours, with C/EBPbeta as a potential marker for these tumours. As C/EBPbeta is known to be expressed during proliferation of cells in vitro, it may participate in the proliferative process of ovarian epithelial tumour cells in vivo and play a central role in tumour progression.

The chemotactic cytokine interleukin-8--a cyst fluid marker for malignant epithelial ovarian cancer?

Due to the difficulties in separating malignant and benign ovarian cysts by transvaginal ultrasound and other techniques, there is a need for biochemical markers in serum or cyst fluids. In the present study we have evaluated the levels of the chemokine interleukin-8 (IL-8) in ovarian cysts. IL-8 is known to be expressed in the normal ovary and to influence proliferation and angiogenesis of several nonovarian types of tumors. Cyst fluids from benign (n = 15) and malignant (n = 13) ovarian tumors were analyzed. The levels of IL-8 were found to be significantly (13-fold) higher in cyst fluids from malignant tumors (18.1 +/- 7.5 ng/ml; mean +/- SE) compared to benign cysts (1.3 +/- 0.7 ng/ml). The plasma levels of IL-8 were considerably lower (2.9 and 0.3% of levels in benign and malignant cyst fluids, respectively) than in cyst fluids. No difference in the plasma levels of patients with benign or malignant tumor could be detected. In contrast, the levels of CA 125 were significantly higher in plasma of patients with malignant disease with the inverse relation in cyst fluids. In conclusion, the levels of IL-8 are markedly elevated in cyst fluid from malignant tumors compared to benign. This specific increase indicates a role for this cytokine in ovarian tumor biology.

The transcription factor C/EBP-beta and its role in ovarian function; evidence for direct involvement in the ovulatory process

Gonadotropins are responsible for maturation of the ovarian follicle and the oocyte. Ovulation is the ultimate step in this process and involves disintegration of the follicular wall and subsequent release of an oocyte into the oviduct. These events are triggered by a surge of luteinizing hormone (LH). Genes expressed in the ovary, that respond to LH, are likely to be involved in the biochemical pathways that regulate ovulation. The transcription factor C/EBP-beta is induced promptly in the ovary, as a response to an ovulatory dose of gonadotropins. We used an ex vivo perfusion system to demonstrate that a specific reduction in ovarian C/EBP-beta expression inhibits ovulation. In such ovaries the oocytes appeared to be entrapped within the follicle. We have found a correlation between the expression level of the activating isoform of C/EBP-beta and the number of oocytes ovulated in response to gonadotropins. Since a reduction in C/EBP-beta expression does not affect the level of the ovulatory mediator prostaglandin endoperoxide synthase-2 (PGS-2), these findings support the view of C/EBP-beta as an important factor in the ovulatory process and highlight a C/EBP-beta-dependent and PGS-2-independent pathway that takes part in regulation of ovulation.

E-cadherin expression in human epithelial ovarian cancer and normal ovary

The ovarian surface epithelium (OSE) is the origin of the majority of human ovarian cancers. These adenocarcinomas are characterized by initial local growth followed by spreading into the peritoneal cavity at later stages of tumor progression. The cell-adhesion molecule E-cadherin (E-cad) plays an important role in maintaining tissue integrity. Disappearance or impaired function of E-cad have often been associated with tumor formation and invasion in vivo and in vitro. The cell-specific expression of E-cad was investigated in normal human ovaries (n = 12), in benign (n = 5) and borderline (n = 4) ovarian epithelial tumors and in adenocarcinomas of different stages and histological grades (n = 18), by immunohistochemistry and immunoblotting. An ovarian cancer cell line (NIH-OVCAR3) was used as a reference. The epithelial origin of the cells was confirmed with cytokeratin (AE1/AE3) staining. In normal ovaries, the expression of E-cad was limited to inclusion cysts or deep clefts lined with OSE, whereas no staining of the OSE could be demonstrated at the surface of the ovary. In contrast, benign and borderline tumors uniformly expressed E-cad. This was observed in malignant tumors of all stages despite their degree of differentiation. E-cad was also present in metastasis from such tumors. The cell-specific expression of E-cad in inclusion cysts of normal ovaries and in epithelial layers of borderline tumors indicates a role for E-cad in the early events of the progression to a malignant phenotype. E-cad was not downregulated in later stages of ovarian cancer progression.

The expression of c-myc during follicular growth and luteal formation in the rat ovary in vivo

The processes of folliculogenesis and formation of corpora lutea involve proliferation and differentiation of the follicular cells. The expression of several oncogenes is associated with the proliferative phase in many cell types. The present study examined the expression and hormonal regulation of the c-myc proto-oncogene during follicular development and the luteal phase of pseudopregnancy. Follicular development was initiated by pregnant mare's serum gonadotropin (PMSG) in immature rats followed two days later by the injection of human chorionic gonadotropin (hCG) to induce ovulation and luteal formation. Ovaries were collected at different time points and the content and distribution of c-myc mRNA/protein were examined. C-myc increased rapidly after the administration of both PMSG and hCG, but the effect of PMSG was less pronounced. The increase after PMSG was transient and localized primarily to the granulosa cells of developing follicles. The ovulatory dose of hCG resulted in a rapid and substantial increase of c-myc mRNA and protein with maximal levels at 1 h and 2-4 respectively. At this stage, the c-myc protein was localized to the follicular cells, the surface epithelium and, to some extent, to the interstitial tissue. There was a subsequent decrease prior to ovulation. The luteal phase was characterized by decreasing levels of c-myc with increasing luteal age. In order to examine the involvement of specific hormones in the regulation of c-myc, hypophysectomized, immature rats were injected sequentially with estradiol (E2) and follicle-stimulating hormone (FSH). Hypophysectomy resulted in a decrease of c-myc compared with intact animals. The administration of E2 resulted in an increase of c-myc mRNA and protein. The subsequent treatment with FSH did not result in a further increase and the levels remained at the same level as with E2 only. However, an ovulatory dose of hCG to E2 and FSH primed animals resulted in an additional increase of c-myc mRNA and protein. The levels after E2 and FSH were considerably lower compared with those of untreated ovaries of intact, immature animals, suggesting the involvement of other endocrine and paracrine factors. The presence of proliferating cell nuclear antigen in cell extracts indicated that the expression of c-myc was associated with phases of increased proliferation of follicular cells after hormonal stimulation. The results demonstrate that c-myc is regulated by hormones (E2, gonadotropins) in the rat ovary during follicular development to the preovulatory stage. The pronounced increase prior to ovulation also suggests a role for c-myc in the regulation of proliferative events involved in luteal formation.