Molecular Human Reproduction, Vol. 6, No. 5, 443-447,
May 2000
© 2000 European Society of Human Reproduction and Embryology
Ovary and oogenesis |
Distribution of steroidogenic enzymes involved in androgen synthesis in polycystic ovaries: an immunohistochemical study
1 Center for Reproductive Medicine, Department of Obstetrics and Gynaecology, Academic Medical Center, Amsterdam, 2 Laser Center, Academic Medical Center, Amsterdam, The Netherlands, and 3 Department of Pathology, Tohoku University School of Medicine, Aoba-ku, Sendai 980-8595, Japan
Abstract
To find an explanation for the possible working mechanism of laparoscopic ovarian electrocautery for the treatment of anovulation in polycystic ovarian syndrome (PCOS), we evaluated the distribution of steroidogenic enzymes involved in the synthesis of ovarian androgens in surgical pathology specimens of entire polycystic ovaries. A total of 13 formalin-fixed and paraffin-embedded samples of the ovaries of patients with clinically proven PCOS were immunostained with specific antibodies against cholesterol side-chain-cleavage enzyme (P450scc), 3ß-hydroxysteroid dehydrogenase (3ß-HSD), 17
-hydroxylase (P450c17) and adrenal 4-binding protein (Ad4BP), a transcription factor of steroidogenic enzymes. Follicular theca cells of all ovaries demonstrated marked immunoreactivity for Ad4BP, P450scc, 3ß-HSD and P450c17. Granulosa cells of seven ovaries expressed Ad4BP, while granulosa cells of three ovaries also showed P450scc. In the granulosa cells of all ovaries, 3ß-HSD and P450c17 immunoreactivity was not observed. In the stroma, luteinized cells of most ovaries demonstrated Ad4BP, P450scc, 3ß-HSD and P450c17 immunoreactivity, but at a much lower level compared with the follicular theca cells. Non-luteinized stromal cells sporadically demonstrated Ad4BP, P450scc, 3ß-HSD and P450c17 immunoreactivity. The stromal steroidogenic cells were mainly located in the ovarian cortex, except for some hilus steroidogenic cells. These data demonstrate that in polycystic ovaries, androgens are mainly produced in the follicular theca cells and to some extent in luteinized stromal cells. This suggests that the working mechanism of laparoscopic electrocautery of the ovary is primarily explained through the reduction of ovarian hyperandrogenism by coagulation of follicular theca cells and concomitant stroma.
17 hydroxylase/ovarian stroma/ovarian surgery/polycystic ovary/side chain cleavage enzyme
Introduction
Laparoscopic electrocautery and laparoscopic laser surgery can restore ovulation in clomiphene-resistant patients with polycystic ovary syndrome (PCOS). These treatments act on the ovarian surface, thereby creating a varying number of holes in the ovarian capsule. This results in destruction of subcapsular follicles and/or subcapsular stroma. Postoperative endocrine alterations observed in most studies after laparoscopic ovarian surgery include a decrease in LH as well as a decrease in serum androstenedione and testosterone concentrations, preceding the return of ovulatory cycles (Kaaijk et al., 1994
). The factor responsible for restoration of ovulation after laparoscopic ovarian surgery in PCOS is unknown but the endocrine alterations suggest that reduction of ovarian hyperandrogenism caused by destruction of androgen-producing tissue plays an important role in restoring ovulation (Cohen, 1996). In this view it is important to know which compartment of the polycystic ovary is responsible for androgen production. We have found four studies which measured androgen production in polycystic ovaries by in-vitro culture experiments (Warren and Salhanick, 1961; Biggs and Thomas, 1981; Mori et al., 1982; Haney et al., 1986). The limitations of these studies are that they are small, lack uniformity, and measure androgen production in only small pieces of polycystic ovarian tissue obtained by wedge resection. Two of these four studies measured androgen production in the ovarian cortex and in the medulla and found that androgens were mainly produced in the ovarian cortex (Warren and Salhanick, 1961; Biggs and Thomas, 1981). The other two studies measured androgen production in follicular theca cells and cortical stroma, and found that the follicular theca cells are much more involved in androgen production than cortical stroma (Mori et al., 1982; Haney et al., 1986). Two studies determined androgen production in the ovarian cortex of polycystic ovaries by immunohistochemical techniques (Tamura et al., 1993; Takayama et al., 1996). These studies concluded that androgens were mainly derived from the follicular theca cells. On the other hand another author (Speroff et al., 1994) states that stromal theca cells, derived from follicular atresia, secrete significant amounts of androstenedione and testosterone, although the author does not reference this statement. The available evidence at this moment therefore suggests that follicular theca cells are the main sites of androgen production in PCOS but the contribution of the ovarian stroma and medulla in androgen production is still not well established.
In this study, we therefore used entire polycystic ovaries to evaluate the contribution of the ovarian stroma, medulla and the follicular theca cells to androgen production by immunolocalization of steroidogenic enzymes involved in androgen production. In addition, the presence of adrenal 4-binding protein (Ad4BP), a universal transcription factor of steroidogenic enzymes, which has been almost exclusively detected in steroid producing cells, was studied (Sasano et al., 1995; Takayama et al., 1995).
Materials and methods
Ovaries
Ovaries were obtained from 13 patients who had undergone unilateral oophorectomy 14–18 years previously as treatment of clomiphene-resistant anovulation (n = 7) or oligomenorrhoea in combination with severe hirsutism and cycle irregularities (n = 6). The ages of the patients were 22–36 years at the time of unilateral oophorectomy. All patients had oligomenorrhoea, elevated testosterone concentrations (>4 nmol/l) and LH/FSH ratios >3.
The ovaries were fixed in 10% formalin and embedded in paraffin 14–18 years previously. Light microscopic examination of all 13 haematoxylin and eosin-stained ovaries showed the presence of a thickened ovarian capsule with multiple subcapsular cysts and a dense hyperplastic ovarian stroma (Figure 1).
|
Immunohistochemistry
Surgical pathology specimens of the ovaries were cut into 3 µm thick sections and mounted on poly-L-lysine-coated glass slides. The primary antibodies used for this study were all polyclonal antibodies produced in rabbits. The optimal dilutions and the sources including their characteristics have been previously described (Takayama et al., 1996
).
Immunostaining was performed as previously described (Suzuki et al., 1994). Briefly, after routine deparaffinization, sections were placed in 0.3% H2O2-methyl alcohol for 30 min to block endogenous peroxidase activity. Slides for Ad4BP immunostaining were placed in citric acid buffer (2 mmol/l citric acid, 9 mmol/l trisodium citrate dihydrate, pH 6.0), heated in a microwave oven (model NE-A40; Matsushita, Tokyo, Japan) for 15 min (500 W) for antigen retrieval, and subsequently allowed to cool down for ~2 h at room temperature. Sections were treated for 30 min at room temperature with 10% normal goat serum.
Immunohistochemical studies were performed using the streptavidin–biotin–peroxidase method with a Histofine kit (Nichirei Co Ltd, Tokyo, Japan). Sections were incubated with primary antibodies for 18 h at 4°C. They were then incubated for 30 min at room temperature with biotinylated anti-rabbit immunoglobulin (Ig)G, washed with 0.01 mol/l phosphate-buffered saline (PBS) and incubated with peroxidase-conjugated streptavidin under the same conditions as described above. The reaction product was subsequently detected by immersion in a solution containing 0.05 mol/l Tris–HCl, pH 7.6, 0.66 mol/l 3.3' -diaminobenzidine (DAB) and 2 mol/l H2O2. A 1% methyl green staining solution was employed for counterstaining the nuclei. As negative controls, 0.01 mol/l PBS, normal rabbit IgG and unrelated antibodies, including adrenocorticotrophic, were used instead of the primary antibodies.
The immunoreactivity of the steroidogenic enzymes and Ad4BP was independently evaluated and graded as follows by two of the authors (H.S. and T.S.): - = cells no immunoreactivity; + = number of positive cells <25% of the total number of cells; ++ = number of positive cells in 25–50% of the total number of cells; +++ = number of positive cells in >50% of the total number of cells. Discordant ovaries were re-evaluated simultaneously by the same authors, using double-headed light microscopy (BH-2, Olympus Co Ltd, Tokyo, Japan). Follicular cells were classified into theca cells and granulosa cells. Stromal cells were classified into luteinized and non-luteinized cells. Luteinized stromal cells were histologically defined as cells associated with abundant clear or eosinophilic cytoplasm. Non-luteinized stromal cells were defined as the stromal cells not presenting abundant clear or eosinophilic cytoplasm.
Results
Immunoreactivity of Ad4BP, P450scc, 3ß-HSD and P450c17 appeared brown as a result of DAB colorimetric reaction. Immunoreactivity of steroidogenic enzymes was detected in the cytoplasm, while that of Ad4BP was detected in the nucleus.
Immunoreactivity for these enzymes and grading of positive follicular theca cells (TC), granulosa cells (GC), stromal luteinized cells (LC) and non-luteinized cells (NL) are summarized in Table I.
|
In all ovaries, the follicular theca cells demonstrated Ad4BP (Figure 2
), P450scc (Figure 3), 3ß-HSD (Figure 4) and P450c17 (Figure 5). In seven ovaries, the granulosa cells were positive for Ad4BP, and in three ovaries these cells were positive for P450scc. These three ovaries were also immunopositive for Ad4BP. 3ß-HSD and P450c17 immunoreactivity was not detected in any of the granulosa cells of the follicles examined. In the stroma, luteinized cells of 11 ovaries demonstrated Ad4BP, P450scc, and P450c17 immunoreactivity, while 12 ovaries demonstrated 3ß-HSD immunoreactivity. In non-luteinized stromal cells, Ad4BP nuclear immunoreactivity was present in eight ovaries. P450scc immunoreactivity was present in the non-luteinized stromal cells of six ovaries which also presented Ad4BP nuclear immunoreactivity. 3ß-HSD and P450c17 were present in four and three ovaries respectively. The number of luteinized cells and non-luteinized cells positive for Ad4BP, 3ß-HSD and P450c17 was lower compared with that of positive theca cells.
Apart from some immunopositive hilus cells (Figure 6), the majority of the stromal cells showing immunoreactivity for the steroidogenic enzymes and Ad4BP were distributed in the outer cortical stroma of the ovary (Figures 7 and 8).
Discussion
The results of this study, the first that determined steroidogenic enzymes involved in androgen synthesis in entire polycystic ovaries, show that follicular theca cells, as well as luteinized stromal cells demonstrate Ad4BP, P450scc, 3ß-HSD and P450c17 immunoreactivity, and that therefore these cells are capable of producing androgens. Furthermore, some non-luteinized stromal cells of some ovaries showed immunoreactivity for these enzymes. However, the total number of immunopositive luteinized and non-luteinized stromal cells is much lower than that of follicular theca cells, suggesting much less involvement of the stromal cells in androgen production compared with follicular theca cells. This finding confirms the results of previous studies (Mori et al., 1982; Haney et al., 1986; Tamura et al., 1993; Takayama et al., 1996) but is in conflict with the statement of Speroff (Speroff et al., 1994). Probably, most of the stromal theca cells, derived from follicular atresia, lose their capability to produce androgens. A recent study (Kyei-Mensah et al., 1998) measuring ovarian stromal volume using three dimensional ultrasound in patients with PCOS and comparing the findings with serum androgen concentrations, demonstrated a positive correlation between stromal volume and serum androgen concentrations. This suggests that the large size of the stromal compartment in PCOS might allow stromal androgen production to become clinically significant despite a much lower involvement of stromal cells in androgen production compared to follicular theca cells, as demonstrated in our study.
An important finding of our study is that the stromal immunopositive cells are mainly present in the outer cortex, while immunopositive cells are not present in the medulla, apart from some immunopositive hilus cells. This strongly suggests that stromal androgen production in polycystic ovaries is derived from the cyst-bearing ovarian cortex. This confirms the results of the previous studies (Warren and Salhanick, 1961; Biggs and Thomas, 1981).
Based on the results of the present study, we conclude that the follicular theca cells are the main sites of androgen production, followed by luteinized stromal cells in the cortex. These results might explain why laparoscopic surgery, which acts on the ovarian surface and subcapsular follicles, results in an immediate decrease of serum androgen concentrations. This fall in serum androgen concentrations leads to a correction of the ovarian pituitary feedback, resulting in follicular growth and ovulation. The duration of regular ovulatory cycles after ovarian surgery might depend on the volume of androgen producing tissue that is destroyed and the severity of ovarian hyperandrogenism.
The drawback of laparoscopic treatments is the risk of peri-ovarian adhesion formation associated with extensive capsular damage. Donesky and Adashi have pointed out that because of the risk of adhesion formation, attention is shifting towards methods that cauterize extensive areas of the deeper stromal areas with minimal damage to the ovarian surface (Donesky and Adashi, 1995). The results of our study suggest that destruction of deeper stromal tissue alone might not reduce androgen-producing tissue sufficiently to restore ovulation. This hypothesis is supported by our preliminary clinical experience on the use of transvaginal interstitial laser treatment of the ovaries to restore ovulation in clomiphene-resistant patients with PCOS (Kaaijk et al., 1997).
In summary, we demonstrated that steroidogenic enzymes involved in androgen production in the polycystic ovary are mainly expressed in the follicular theca cells and to a lesser extent in the ovarian cortical stroma. Therefore, we suggest that the working mechanism of laparoscopic electrocautery of the ovary might be primarily explained by coagulation of follicular theca cells and concomitant stroma.
Notes
4 To whom correspondence should be addressed at: Center for Reproductive Medicine, Department of Obstetrics and Gynaecology, Academic Medical Center, PO Box 22660, 1105 AZ Amsterdam, The Netherlands 
References
Biggs, J.S.G. and Thomas, F.J. (1981) Sites of steroid production in the polycystic ovary. Br. J. Obstet. Gynaecol., 88, 42–46.[Web of Science][Medline]
Cohen, J. (1996) Laparoscopic procedures for treatment of infertility related to polycystic ovarian syndrome. Hum. Reprod. Update, 2, 337–344.
Donesky, B.W. and Adashi, E.Y. (1995) Surgically induced ovulation in the polycystic ovary syndrome: Wedge resection revisited in the age of laparoscopy. Fertil. Steril., 63, 439–463.[Web of Science][Medline]
Haney, A.F., Maxson, W.S. and Schomberg, D.W. (1986) Compartimental ovarian steroidogenesis in polycystic ovary syndrome. Obstet. Gynecol., 68, 638–644.[Web of Science][Medline]
Kaaijk, E.M., Beek, J.F. and van der Veen, F. (1994) Laparoscopic laser surgery of chronic hyperandrogenic anovulation. Laser Surg. Med., 16, 292–302.
Kaaijk, E.M., van der Veen, F., Beek, J.F. et al. (1997) Transvaginal interstitial laser treatment of the ovary for the management of chronic hyperandrogenic anovulation: First clinical experience. Gynecol. Obstet. Invest., 44, 115–119.[Web of Science][Medline]
Kyei-Mensah, A., LinTan, S., Zaidi, J. and Jacobs, H.S. (1998) Relationship of ovarian stromal volume to serum androgen concentrations in patients with polycystic ovary syndrome. Hum. Reprod., 13, 1437–1441.
Mori, T., Fujita, Y., Nihnobu, K. et al. (1982) Significance of atretic follicles as the site of androgen production in polycystic ovaries. J. Endocrinol. Invest., 5, 209–214.[Web of Science][Medline]
Takayama, K., Fukaya, T., Sasano, H. et al. (1996) Immunohistochemical study of steroidogenesis and cell proliferation in the polycystic ovarian syndrome. Hum. Reprod., 11, 1387–1392.
Takayama, K., Sasano, H., Fukaya, T. et al. (1995) Immunohistochemical localization of Ad4-binding protein with correlation to steroidogenic enzyme expression in cycling human ovaries and sex cord stromal tumors. J. Clin. Endocrinol. Metab., 80, 2815–2821.[Abstract]
Tamura, T., Kitawaki, J., Yamamoto, T. et al. (1993) Immunohistochemical localization of 17-hydroxylase/C17-20 lyase and aromatase cytochrome P-450 in polycystic human ovaries. J. Endocrinol., 139, 503–509.
Sasano, H., Shizawa, S., Suzuki, T. et al. (1995) Ad4BP in the human adrenal cortex and its disorders. J. Clin. Endocrinol. Metab., 80, 2378–2380.[Abstract]
Speroff, L., Glass, R.H. and Kase, N. (eds) (1994) Clinical Gynecologic Endocrinology and Infertility. Williams and Wilkins, Baltimore, USA, 465 pp.
Suzuki, T., Sasano, H., Kimura, N. et al. (1994) Immunohistochemical distribution of progesterone, androgen and oestrogen receptors in the human ovary during the menstrual cycle: Relationship to expression of steroidogenic enzymes. Hum. Reprod., 9, 1589–1595.
Warren, J.C. and Salhanick, H.A. (1961) Steroid biosynthesis in the human ovary. J. Clin. Endocrinol. Metab., 21, 1218–30.
Submitted on June 7, 1999; accepted on February 15, 2000.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  R. S. Viger, S. M. Guittot, M. Anttonen, D. B. Wilson, and M. Heikinheimo Role of the GATA Family of Transcription Factors in Endocrine Development, Function, and Disease Mol. Endocrinol., April 1, 2008; 22(4): 781 - 798. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Bielinska, S. Kiiveri, H. Parviainen, S. Mannisto, M. Heikinheimo, and D. B. Wilson Gonadectomy-induced Adrenocortical Neoplasia in the Domestic Ferret (Mustela putorius furo) and Laboratory Mouse Veterinary Pathology, March 1, 2006; 43(2): 97 - 117. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Bielinska, E. Genova, I. Boime, H. Parviainen, S. Kiiveri, J. Leppaluoto, N. Rahman, M. Heikinheimo, and D. B. Wilson Gonadotropin-Induced Adrenocortical Neoplasia in NU/J Nude Mice Endocrinology, September 1, 2005; 146(9): 3975 - 3984. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Bachelot, G. Meduri, N. Massin, M. Misrahi, F. Kuttenn, and P. Touraine Ovarian Steroidogenesis and Serum Androgen Levels in Patients with Premature Ovarian Failure J. Clin. Endocrinol. Metab., April 1, 2005; 90(4): 2391 - 2396. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||