| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Department of Pathology (Y.S., T.S., K.H., H.S.), Department of Medicine (Y.S., H.S., S.I.), Division of Nephrology, Endocrinology, and Vascular Medicine, and Department of Obstetrics and Gynecology (K.I.), Tohoku University School of Medicine, Sendai, Japan
Address all correspondence and requests for reprints to: Yoko Sato, M.D., Division of Nephrology, Endocrinology, and Vascular Medicine, Tohoku University School of Medicine, 2-1 Seiryo-machi, Aoba-ku, Sendai 980-8575, Japan. E-mail: Yoko21tako{at}aol.com.
| Abstract |
|---|
|
|
|---|
| Introduction |
|---|
|
|
|---|
DAX-1 and COUP-TFII have been previously reported to function as a transcriptional suppressor of Ad4BP/SF-1, a transcription factor that regulates the expression of the steroidogenic P450 genes (3, 4, 9, 10, 11). In addition, the promoter of the murine DAX-1 gene has been shown to be activated by Ad4BP/SF-1 and to be suppressed by COUP-TFII (12, 13).
It is well known that human ovarian steroidogenesis is regulated by both the temporal and spatial expression patterns of enzymes specifically involved in each step of steroidogenesis. However, to date the expression of the nuclear DNA-binding proteins, DAX-1 and COUP-TFII, has not been reported in the normal cycling human ovary, and thus, their biological roles as transcription factors of steroidogenic enzymes in the human ovary still remain unclear. Therefore, in this study we examined the temporal and spatial immunolocalization of DAX-1 and COUP-TFII and compared the results with those of P450 enzymes involved in human ovarian steroidogenesis and Ad4BP/SF-1.
| Materials and Methods |
|---|
|
|
|---|
A total of 24 specimens of normal cycling human ovaries were available for examination in this study. These ovaries were retrieved from the surgical pathology files of Tohoku University Hospital (Sendai, Japan). The protocol for this study was approved by the ethics committee at Tohoku University School of Medicine (Sendai, Japan). In addition to hysterectomy for invasive squamous cell carcinoma or adenocarcinoma of the uterine cervix, all patients underwent unilateral oophorectomy to examine the possible presence of metastasis in the ovary. After obtaining informed consent, unilateral oophorectomy was performed in patients diagnosed at or beyond clinical stage Ib. This was done to exclude the possibility of ovarian metastasis in these patients (14, 15, 16, 17). All patients enrolled in this study were premenopausal Japanese subjects, ranging in age from 2649 yr, did not have any sex steroid abnormalities, and were not taking any steroid-inducing drugs, such as oral contraceptives and/or related steroids. Moreover, a previous study by Konishi et al. (18) found that patients diagnosed with cervical carcinoma were rarely found to have sex steroid abnormalities. No significant histopathological abnormalities were detected by microscopic examination of hematoxylin- and eosin-stained slides in any of the 24 cases. Ovaries were immediately fixed in 4% paraformaldehyde (pH 7.4) for 12 h at 4 C and subsequently embedded in paraffin wax.
The phase of the menstrual cycle was determined in each case by taking a detailed patient history and performing endometrial dating. Ovarian follicles and corpus luteum for each specimen were classified according to the phase of the menstrual cycle, histological features, and immunolocalization patterns of steroidogenic enzymes, as reported by Suzuki et al. (19).
Six follicular stages were classified in our series. In the present study we examined primordial follicles (n = 54); primary follicles (n = 40); preantral follicles (n = 16); nondominant antral follicles, including P450 aromatase (P450arom)-negative, P450 cholesterol side-chain cleavage (P450scc)-positive, 3ß-hydroxysteroid dehydrogenase (3ß-HSD)-positive, and P450 17
-hydroxylase (P450C17)-positive antral follicles (n = 12); dominant follicles, including all enzyme-positive antral follicles (n = 7); and atretic follicles (n = 20).
Four luteal stages were identified. We examined corpora lutea that were positive for all enzymes (n = 7); early degenerating corpora lutea that were P450arom negative, P450scc positive, 3ß-HSD positive, and P450C17 positive (n = 8); late degenerating corpora lutea that were negative for all enzymes (n = 9); and corpus albicans (n = 10). The immunolocalization of P450arom, P450scc, 3ß-HSD, P450C17, and Ad4BP/SF-1 in these human ovaries has been previously reported by Suzuki et al. (19) and Takayama et al. (20).
Primary antibodies
Rabbit polyclonal antibody for DAX-1 was obtained from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA). The antibody for COUP-TFII was provided by Dr. Sotirios K. Karathanasis (Parkes-Davis Pharmaceutical Research Division, Warner-Lambert, Ann Arbor, MI). The generation and characterization of the primary polyclonal antibody for COUP-TFII have been described previously (21). Briefly, a bacterially expressed and affinity-purified peptide spanning the N-terminal and DNA-binding domains (amino acid residues 1170) of COUP-TFII was used to raise antibodies in rabbits. The polyclonal antibody for Ad4BP/SF-1 was provided by Dr. K. Morohashi (National Institute for Basic Biology, Okazaki, Japan). The generation and characterization of Ad4BP/SF-1 antibody have been described previously (22), and the application of this antibody in an immunohistochemical study has been previously reported (23, 24).
Immunohistochemistry
Immunohistochemical analysis was performed using the streptavidin-biotin amplification method using a Histofine kit (Nichirei, Tokyo, Japan) as described previously (23). After deparaffinization, antigen retrieval for DAX-1, COUP-TFII, and Ad4BP/SF-1 analyses was performed by heating the slides in an autoclave at 120 C for 5 min in citric acid buffer (2 mmol/liter citric acid and 9 mmol/liter trisodium citrate dehydrate, pH 6.0). The dilutions of the primary antibodies used in our study were as follows: DAX-1, 1:500; COUP-TFII, 1:1500; and Ad4BP/SF-1, 1:1000. The antigen-antibody complex was visualized with 3,3'-diaminobenzidine solution [1 mmol/liter 3,3'-diaminobenzidine, 50 mmol/liter Tris-HCl buffer (pH 7.6), and 0.006% H2O2] and counterstained with hematoxylin. Tissue sections of adrenal gland, kidney, spleen, and ovary were used as positive controls for DAX-1, COUP-TFII, and Ad4BP/SF-1, respectively. As a negative control, normal rabbit or mouse IgG was used instead of the primary antibodies. No specific immunoreactivity was detected in these sections.
Scoring of immunoreactivity
After completely reviewing the immunohistochemical sections, relative immunoreactivity for DAX-1, COUP-TFII, and Ad4BP/SF-1 in the ovarian specimens was classified into the following groups by blind evaluation of each slide by three of the authors (Y.S., T.S., and H.S.) independently: 2 = strongly positive, 1 = weakly positive, and 0 = negative or no immunopositive staining. Results found not to agree among the observers were reevaluated together using a multiheaded light microscope. Relative immunoreactivity of the cells was evaluated via an H-scoring system, as described by McCarty et al. (25) with some modifications. Briefly, more than 500 cells were counted in each case, and H-scores were subsequently generated by adding together 2x% strongly stained nuclei, and 1x% weakly stained nuclei, giving a possible range of 0200. Statistical significance was evaluated using Scheffés F test, and P < 0.05 was considered significant.
| Results |
|---|
|
|
|---|
Results for DAX-1, COUP-TFII, and Ad4BP/SF-1 immunostaining as well as those for P450scc, 3ß-HSD, P450C17, and P450arom immunoreactivity in the normal cycling ovary are summarized in Table 1
. Data on steroidogenic enzymes are cited from our previous study (19). Takayama et al. (20) reported immunohistochemical localization of Ad4BP/SF-1 in normal cycling human ovaries, but we evaluated the relative immunoreactivity of the cells in different stages using an H-scoring system. The cellular localization of steroidogenic enzymes has been previously reported in detail (19). Immunoreactivity for DAX-1, COUP-TFII, and Ad4BP/SF-1 was exclusively detected in the nuclei.
|
In these follicles, no immunoreactivity of steroidogenic enzymes was detected as previously reported (19). In both primordial and primary follicles, DAX-1 was detected in granulosa cells, whereas immunoreactive COUP-TFII was found to be localized in the stromal cells surrounding the granulosa cells. In preantral follicles, DAX-1, COUP-TFII, and Ad4BP/SF-1 immunoreactivities were detected in both granulosa and thecal cells. However, the immunolocalization patterns for these transcription factors are different, as shown in Table 1
and Fig. 1
, AC.
|
P450arom-positive follicles were considered the dominant follicles among the antral follicles examined, as reported previously by Suzuki and colleagues (19). Immunoreactive proteins for DAX-1 and Ad4BP/SF-1 were markedly expressed in both granulosa and theca interna cells of these follicles (Fig. 2
, A and B). COUP-TFII was also positive in these follicles, but its immunoreactivity was shown to be more markedly expressed in thecal cells than in granulosa cells (Fig. 2C
).
|
Functioning corpus luteum
After ovulation, immunoreactive proteins for DAX-1 and Ad4BP/SF-1 were detected in both the luteinized granulosa and thecal cells, but patterns of immunolocalization for these two transcription factors were different. Immunopositive DAX-1 was markedly expressed in luteinized granulosa cells (Fig. 2G
); however, the opposite was true for immunoreactive Ad4BP/SF-1, which was found to be predominantly expressed in luteinized thecal cells (Fig. 2H
). COUP-TFII was detected only in luteinized thecal cells (Fig. 2I
).
Atretic follicle, degenerating corpus luteum, and corpus albicans
Immunoreactive proteins for DAX-1, COUP-TFII, and Ad4BP/SF-1 were shown to be expressed in the theca interna cells of atretic follicles. In the degenerating corpus luteum, DAX-1 was not detected. Both immunoreactive COUP-TFII and Ad4BP/SF-1 were detected in luteinized thecal cells of the early degenerating corpus luteum. However, DAX-1, COUP-TFII, and Ad4BP/SF-1 were not detected in the cells of the corpus albicans (Table 1
).
Stroma and surface epithelium
Both DAX-1 and COUP-TFII immunoreactive proteins were detected in the stromal cells of all specimens. However, DAX-1 and COUP-TFII immunopositive stromal cells did not express immunoreactive proteins for Ad4BP/SF-1 and steroidogenic enzymes. DAX-1 was identified in the surface epithelium of the ovary, whereas COUP-TFII, Ad4BP/SF-1, and all steroidogenic enzymes examined in this study were immunonegative in this epithelium (Fig. 2
, JL).
| Discussion |
|---|
|
|
|---|
The direct correlation among levels of DAX-1, COUP-TFII, and steroidogenic enzymes could not be demonstrated in our study. However, based on the following findings we postulated that DAX-1 and COUP-TFII may play a role in the modulation of Ad4BP/SF-1-dependent transcription of steroidogenic enzymes in different cell types and follicular stages in normal cycling ovaries. DAX-1 was originally reported to suppress the transcriptional activation of the aromatase gene via Ad4BP/SF-1 DNA-binding proteins in vitro (28). In addition, COUP-TFII was found to inhibit Ad4BP/SF-1-mediated P450C17 gene transcription (29). Therefore, DAX-1 and COUP-TFII have been proposed to inhibit Ad4BP/SF-1-dependent transcription of steroidogenic enzymes. In the present study DAX-1 was predominantly associated with granulosa cell phenotypes, whereas COUP-TFII was associated with thecal cell phenotypes in the ovarian follicle even after ovulation and formation of the corpus luteum. In human ovary it is well known that aromatase is exclusively detected in granulosa cells, whereas P450C17 is expressed in the thecal cells of dominant follicles. These patterns of enzyme distribution suggest that DNA-binding proteins are colocalized with these enzymes to regulate the expression of steroidogenic enzymes in the human ovary in vivo.
Wang and co-workers (28) reported that DAX-1 suppressed transcriptional activation of the aromatase gene via Ad4BP/SF-1 DNA-binding proteins in vitro. Ad4BP/SF-1 is generally required for the induction of various steroidogenic enzymes. In our study both Ad4BP/SF-1 and DAX-1 immunoreactivities were shown to be expressed in granulosa cells of the ovarian follicle after the preantral follicular stage. Therefore, we postulated that Ad4BP/SF-1-mediated transcription of steroidogenic enzymes may be inhibited by DAX-1 from premordial to preantral stages and in nondominant follicles, because DAX-1 immunoreactivity was relatively high compared with that of Ad4BP/SF-1 in these stages of human ovarian follicular development. From the preantral to dominant follicles, our findings suggest that the increase in expression of Ad4BP/SF-1 immunoreactivity was greater than that in DAX-1. In dominant follicles, this significant increase in Ad4BP/SF-1 expression may induce the transcription of steroidogenic enzymes through this nuclear protein. However, the results of our present study have demonstrated that its expression is relatively low in luteinized granulosa cells and was associated with aromatase and P450scc immunoreactivity. In a very recent study by Hinshelwood et al. (30), liver receptor homolog 1 (LRH-1) was reported to act in a similar fashion as Ad4BP/SF-1. In luteinized granulosa cells of human cycling ovaries, LRH-1 may be associated with steroidogenic enzyme regulation; however, further investigations are required to elucidate the temporal and spatial expression patterns of LRH-1 in the human ovary and its possible function in steroidogenic enzyme regulation. The increase in Ad4BP/SF-1 expression detected in these granulosa cells was parallel that of DAX-1 expression. In these stages of ovarian follicular development, DAX-1 expression may be activated by Ad4BP/SF-1 as reported previously in vitro (17, 26).
Shibata et al. (29) reported that COUP-TFII inhibited Ad4BP/SF-1-mediated P450C17 gene transcription. In our study COUP-TFII immunoreactive proteins were found to be colocalized with P450C17 in thecal cells of normal human ovaries. Throughout the menstrual cycle, COUP-TFII protein levels remain consistently high in thecal cells, whereas the Ad4BP/SF-1 protein level appears to increase during the development of the follicle. However, due to the increase in the expression level of Ad4BP with respect to COUP-TFII, it is possible to speculate that Ad4BP/SF-1 can induce the Ad4BP/SF-1-mediated transcription of steroidogenic enzymes in normal cycling human ovaries. On the contrary, Suzuki et al. (31) reported that protein expression levels for Ad4BP/SF-1 remained constant, whereas protein expression levels for COUP-TFII changed in the adrenal gland and its disorders. In contrast to the adrenal gland, steroidogenesis in granulosa cells of the human ovary may be controlled with alterations in the level of Ad4BP/SF-1 expression.
DAX-1 was detected in stromal cells, primordial follicles, and primary follicles, in contrast to Ad4BP/SF-1, which was not detected in these tissues. These findings appear to suggest that DAX-1 may be regulated by factors other than Ad4BP/SF-1 in stromal cells, primordial follicles, and primary follicles of the human ovary. However, further investigations are required to clarify the role that DNA-binding proteins play with respect to the expression and regulation of steroidogenic enzymes in the normal human ovary.
| Footnotes |
|---|
-hydroxylase; SF-1, steroidogenic factor-1. Received November 4, 2002.
Accepted March 5, 2003.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
J. Guo, S.-X. Tao, M. Chen, Y.-Q. Shi, Z.-Q. Zhang, Y.-C. Li, X.-S. Zhang, Z.-Y. Hu, and Y.-X. Liu Heat Treatment Induces Liver Receptor Homolog-1 Expression in Monkey and Rat Sertoli Cells Endocrinology, March 1, 2007; 148(3): 1255 - 1265. [Abstract] [Full Text] [PDF] |
||||
![]() |
F.-Q. Yu, C.-S. Han, W. Yang, X. Jin, Z.-Y. Hu, and Y.-X. Liu Activation of the p38 MAPK pathway by follicle-stimulating hormone regulates steroidogenesis in granulosa cells differentially J. Endocrinol., July 1, 2005; 186(1): 85 - 96. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. W. Kim, N. Peng, W. E. Rainey, B. R. Carr, and G. R. Attia Liver Receptor Homolog-1 Regulates the Expression of Steroidogenic Acute Regulatory Protein in Human Granulosa Cells J. Clin. Endocrinol. Metab., June 1, 2004; 89(6): 3042 - 3047. [Abstract] [Full Text] [PDF] |
||||
![]() |
Q. Huang, A. Raya, P. DeJesus, S.-H. Chao, K. C. Quon, J. S. Caldwell, S. K. Chanda, J. C. Izpisua-Belmonte, and P. G. Schultz Identification of p53 regulators by genome-wide functional analysis PNAS, March 9, 2004; 101(10): 3456 - 3461. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |