Histochem Cell Biol (2004) 122:461471 DOI 10.

1007/s00418-004-0709-6

ORIGINAL PAPER

Ioannis Mylonas Udo Jeschke Irmgard Wiest Anna Hoeing Julia Vogl Naim Shabani Christina Kuhn Sandra Schulze Markus S. Kupka Klaus Friese

Inhibin/activin subunits alpha, beta-A and beta-B are differentially expressed in normal human endometrium throughout the menstrual cycle
Accepted: 31 August 2004 / Published online: 12 October 2004  Springer-Verlag 2004

Abstract Inhibins are dimeric glycoproteins composed of an alpha (a) subunit and one of two possible beta (b-) subunits (bA or bB). The aims of this study were to assess the frequency and tissue distribution patterns of the inhibin subunits in normal human endometrium. Samples from human endometrium from proliferative phase (PP; n=32), early secretory phase (ES; n=10) and late secretory phase (LS; n=12) were obtained. Immunohistochemistry, immunofluorescence and a statistical analysis were performed. All three inhibin subunits were expressed by normal endometrium by immunohistochemistry and immunofluorescence. Inhibin-a was primarily detected in glandular epithelial cells, while inhibin-b subunits were additionally localised in stromal tissue. Inhibin-a staining reaction increased significantly between PP and ES (P<0.05), PP and LS (P<0.01), and ES and LS (P<0.02). Inhibin-bA and -bB were significant higher in LS than PP (P<0.05) and LS than ES (P<0.05). All three inhibin subunits were expressed by human endometrium varying across the menstrual cycle. This suggests substantial functions in human implantation of inhibin-a subunit, while stromal expression of the b subunits could be important in the paracrine signalling for adequate endometrial maturation. The distinct expression in human endometrial tissue suggests a synthesis of inhibins into the lumen and a predominant secretion of activins into the stroma. Keywords Endometrium Normal Immunohistochemistry Immunofluorescence
I. Mylonas and U. Jeschke contributed equally to this work I. Mylonas ()) U. Jeschke I. Wiest A. Hoeing J. Vogl N. Shabani C. Kuhn S. Schulze M. S. Kupka K. Friese First Department of Obstetrics and Gynaecology, Ludwig-Maximilians-University Munich, Maistrasse 11, 80337 Munich, Germany e-mail: ioannis.mylonas@med.uni-muenchen.de Tel.: +49-89-51604266 Fax: +49-89-51604916

Inhibin/activin subunits Inhibin-alpha Inhibin-beta A Inhibin-beta B

Introduction
The endometrium undergoes morphological and biochemical changes during the menstrual cycle. Histologically, human endometrium contains a variety of cell types: surface and glandular epithelium associated with extracellular matrix and stromal components, including vascular and neuronal elements. The endometrium is not only a target tissue for hormones, but also an endocrine organ itself, synthesising substances that are thought to influence the endometrial function by autocrine and paracrine ways. Inhibins are dimeric glycoproteins, composed of an a subunit and one of two possible b subunits (bA or bB), that were initially isolated from the gonads and identified as modulators of FSH production from the anterior pituitary gland (Vale et al. 1988; Ying 1988; de Kretser et al. 2002). Two forms of inhibin, namely inhibin A and inhibin B exist. These proteins were shown to be disulphidelinked dimers that have a common a subunit but just one of two b subunits, differentiated in inhibin A (a-bA) and in inhibin B (a-bB). Each of the activins also represent disulphide-linked dimers of the b subunits of inhibin, either homodimers of the b subunit (activin A: bA-bA, activin B: bB-bB) or a heterodimer (activin AB: bA-bB). The inhibin/activin subunits are homologues of each other and belong to the transforming growth factor-beta (TGF-b) family of proteins (Vale et al. 1988; Ying 1988; de Kretser et al. 2002). Recently, three additional b subunits have been identified, determined as bC (Htten et al. 1995), bD (Xenopus only; Oda et al. 1995) and bE (Fang et al. 1996), although their precise function remains unknown. In the regulation of human menstrual cycle, inhibin and oestradiol (E2) seem to be the negative controls of FSH secretion that disappear at the time of luteal regression (Chabbert Buffet et al. 1998). The expression of

462

inhibin/activin subunits in different female tissue suggests diverse functions, such as paracrine modulators of reproductive function (de Kretser et al. 2002; Welt 2002) and gonadal carcinogenesis (Matzuk et al. 1992). Inhibin has been demonstrated by immunohistochemical means in human endometrium (Leung et al. 1998; Mylonas et al. 2004). Human endometrial cells in culture also synthesise all three inhibin subunits, as detected by RT-PCR and confirmed by two-site immunoassays (Petraglia et al. 1998), although the precise role of inhibin in human endometrium remains unknown. Since inhibins and activins are produced in endometrial cells, they might play an important role in the regulation of endometrial cell function, regulating endometrial maturation, decidualisation and implantation processes (Jones et al. 2002). The secretion of the a subunit might determine which active dimers of inhibins or activins are synthesised, whereas the expression of b subunits could facilitate and mediate autocrine/paracrine signals in this dynamic tissue. Recently, a higher expression of the inhibin-a chain during the late secretory phase compared to the other phases was observed (Mylonas et al. 2003a, 2004) and a higher amount of inhibin was detected in culture medium of isolated glandular epithelial cells from the secretory phase (Mylonas et al. 2003b). Additionally, a relationship between inhibin, oestradiol and cortisol secretion in cultured human endometrial glandular epithelial cells was also suggested (Mylonas et al. 2003a). Because specific monoclonal antibodies against all inhibin subunits have only been available for a short period of time, systematic investigations on the combined expression of inhibin/activin subunits have not been performed yet. Therefore, the aims of the present study were: A. The determination of the frequency and tissue distribution patterns of the inhibin/activin subunits in normal human endometrium by immunohistochemical means. B. The assessment of a combined expression of inhibin-a and both b subunits (bA and bB subunits) using a double immunofluorescence technique.

with no hormonal treatment for 3 months prior to surgery. All pathological and hyperplastic endometrial samples were excluded from this study. Endometrium samples were classified according to anamnestic and histological dating (Dallenbach-Hellweg and Poulsen 1985) into proliferative (days 114, n=32), early secretory (days 1522, n=10) and late secretory phase (days 2328, n=12), as previously described (Mylonas et al. 2000). Immunohistochemistry Immunohistochemistry was performed using a combination of pressure cooker heating and the standard streptavidin-biotin-peroxidase complex with the use of the mouse IgGVectastain Elite ABC kit (Vector Laboratories, Burlingame, CA, USA). Mouse monoclonal antibodies used for these experiments are listed in Table 1. The standardisation, dilution and optimisation of this protocol was primarily tested on normal premenopausal ovary tissue. Briefly, paraffin-fixed tissue sections were dewaxed using xylol for 15 min, rehydrated in an ascending series of alcohol (70%, 96% and 100%) and subjected to antigen retrieval for 10 min in a pressure cooker using sodium citrate buffer (pH 6.0) containing 0.1 M citric acid and 0.1 M sodium citrate in distilled water. After cooling, sections were washed twice in phosphate-buffered saline (PBS). Endogenous peroxidase activity was quenched by immersion in 3% hydrogen peroxide (Merck, Darmstadt, Germany) in methanol for 20 min. Non-specific binding of the primary antibodies was blocked by incubating the sections with diluted normal serum (10 ml PBS containing 150 ml horse serum; provided by Vector Laboratories) for 20 min at room temperature. Sections were then incubated at room temperature for 60 min with the primary antibodies. Inhibin-a, -bA and -bB were diluted in PBS. After washing with PBS, sections were incubated in diluted biotinylated serum (10 ml PBS containing 50 ml horse serum; provided by Vector Laboratories) for 30 min at room temperature. After incubation with the avidin-biotin peroxidase complex (diluted in 10 ml PBS; Vector Laboratories) for 30 min and repeated washing steps with PBS, visualisation was performed with substrate and chromogen 3,30 -diaminobenzidine (DAB; Dako, Glostrup, Denmark) for 810 min. Sections were counterstained with Mayers acidic haematoxylin and dehydrated in an ascending series of alcohol (50 98%). After xylol treatment, sections were covered. Negative controls were performed by replacing the primary antibody with normal mouse serum and/or a bhCG antibody in the same dilution used for inhibin detection. Immunohistochemical staining was performed using an appropriate positive control comprising ovaries containing follicular cysts. Positive cells showed a brownish colour and negative controls as well as unstained cells were blue. Immunofluorescence For the immunohistochemical characterisation cryosections from proliferative, early and late secretory phase were examined as previously described (Jeschke et al. 2002). Antibodies are listed in Table 1. The inhibin-a polyclonal antibody was used. Briefly, all samples were fixed in 5% buffered formalin. They were diluted to 10 mg/ml with PBS and incubated with the slides overnight at 4C. After washing, Cy2-labelled goat anti-rabbit IgG and Cy3-labelled

Materials and methods

Tissue samples Samples of human endometrium were obtained from 54 premenopausal, non-pregnant patients undergoing gynaecological surgery either by D&C or hysterectomy for benign diseases, mainly uterine leiomyomata. All women had a normal and regular menstrual cycle

Table 1 Antibodies used for immunohistochemical characterisation of endometrial tissue samples by immunohistochemistry and immunofluorescence

Antibody b-hCG Inhibin-a Inhibin-a Inhibin-bA Inhibin-bB Cy2 and Cy3 Cy2 and Cy3

Clone Polyclonal R1 Polyclonal E4 C5

Isotype Rabbit IgG Mouse IgG2a Rabbit IgG Mouse IgG2b Mouse IgG2a Goat anti-rabbit IgG Goat anti-mouse IgG

Dilution 1:200 1:50 1:40 1:50 1:10 1:200 1:200

Source Dianova, Hamburg, Germany Serotec, Oxford, UK Signet, Dedham, MA, USA Serotec Serotec Dianova Dianova

463 goat anti-mouse IgG, diluted 1:200, served as second antibody. The slides were finally embedded in mounting buffer containing 4,6-diamino-2-phenylindole (DAPI) resulting in blue staining of the nucleus. Slides were examined with a Zeiss (Jena, Germany) Axiophot photomicroscope. Digital images were obtained with a digital camera system (CF20DXC; KAPPA Messtechnik, Gleichen, Germany) and saved on computer. Western blot analysis of inhibin-a, -bA and -bB subunits Endometrial adenocarcinoma cells of the cell line HEC (obtained from ECACC, Salisbury, UK; 3107) were lysed in 400 ml Laemmli sample buffer for 5 min at 100C. Lysate proteins were separated by SDS-PAGE and transferred to nitrocellulose membranes. The gels used are 16%, according to the manufacturer (Biorad, Germany). After blocking with 4% BSA in TBS/Tween (20 mM TRIS/HCl, pH 7.2, 1 M NaCl, 0.05% Tween 20), blots were incubated with monoclonal mouse antibodies against inhibina, -bA and -bB (10 mg/ml), mouse IgG as a control or TBS/Tween (blank) in TBS/Tween with 1% BSA for 16 h at 6C. Blots were washed and incubated with a goat anti-mouse IgGHRP conjugate (diluted 1:500 in TBS/Tween with 1% BSA) for 2 h. Staining was performed with 10 mg 3-amino-9-ethylcarbazole in 1 ml acetone, 25 ml 50 mM sodium acetate, pH 5.0, and 15 ml H2O2 (30%). Evaluation and statistical analysis The intensity and distribution patterns of specific inhibin-a, -bA and -bB subunit immunohistochemical staining was evaluated using the semiquantitative IRS score, as previously described (Remmele and Stegner 1987; Mylonas et al. 2000). The IRS score was calculated by multiplication of optical staining intensity (graded as 0=no, 1=weak, 2=moderate and 3=strong staining) and the percentage of positively stained cells (0=no staining, 1=<10% of the cells, 2=1150% of the cells, 3=5180% of the cells and 4=>81% of the cells). The slides were examined by two independent observers, including a gynaecological pathologist (N.S.). Sections were examined using an Olympus (Tokyo, Japan) photomicroscope. Digital images were obtained with a digital camera system (Olympus) and were saved on computer. The results were evaluated using the non-parametric Mann-Whitney U rank-sum test (SPSS; Chigaco, IL, USA). Significance was assumed at P<0.05. Fig. 1 Western blot analysis of the used inhibin/activin subunits. Detection of inhibin-a (lane 2), inhibin-bA (lane 3) and inhibin-bB (lane 4) in HEC cell lysates. Lane 1 molecular weight marker proteins, lane 5 negative control

antibody. Immunohistochemical staining was performed using an appropriate positive control comprising ovaries containing follicular cysts. Inhibin-a stained positive with ovarian granulosa cells and theca interna cells (Fig. 3a, b), while inhibin-bA and -bB subunits also showed a positive immunohistochemical staining reaction with human ovarian tissue (Fig. 3cf) Immunohistochemical expression of inhibin-a, -bA and -bB All three inhibin/activin subunits were detected in normal human endometrial tissue across the menstrual cycle. Inhibin-a and both the bA and bB subunits showed a similar expression pattern, being primarily localised in glandular epithelial cells. However, stromal cells begin to synthesise both b subunits at the onset of late secretory phase. An overview of the staining intensity is provided in Table 2. The inhibin-a subunit was expressed in human endometrium primarily in the glandular and surface epithelium. During the proliferative phase a minimal staining reaction could be observed at the apical component of glandular cells (Fig. 4a). The staining intensity increased during the early secretory phase being localised at the basal and apical parts (Fig. 4b). The strongest staining reaction was observed during late secretory phase (Fig. 4c). With the onset of early secretory phase the a subunit could also be detected in distinct stromal cells (Fig. 4d). The immunoreactive score for inhibin-a in-

Results
Western blot analysis was recently described for the used inhibin/activin subunit antibodies by Groome et al. (1990) and Groome and Lawrence (1991). We performed a western blot analysis of the monoclonal inhibin-a, -bA and -bB antibodies to test specificity of the antibodies used for immunohistochemistry. Our western blot analysis (Fig. 1) confirmed previous described results (Groome et al. 1990; Groome and Lawrence 1991). Inhibin-a antibody produced a main protein band in the 32 kDa molecular mass range, while both b antibodies generated a main protein band in the 28 kDa mass range, correlating with the known molecular mass described for inhibin/ activin subunits. For immunohistochemistry negative controls were performed by replacing the primary antibodies with normal mouse serum and bhCG antibody in the same dilution. While placenta reacted positive with the bhCG antibody (Fig. 2a), both proliferative (Fig. 2b) and late secretory endometrial tissue (Fig. 2c) did not react with this

464 Fig. 2ac Immunohistochemical staining reaction of bhCG in endometrial samples for negative control. Normal placental tissue reacted positive with the bhCG antibody (a). Both proliferative (b) and late secretory endometrial tissue (c) did not react with this antibody when applied with the same procedure as used for inhibin/activin subunits. Magnification ac 125

creased significantly between proliferative and early secretory phase (P<0.05), proliferative and late secretory phase (P<0.01) as well as between early and late secretory phase (P<0.02) (Fig. 7). The bA subunit was also detected in endometrial glandular epithelial cells. A diffuse immunohistochemical staining reaction could be observed across the menstrual cycle. Proliferative (Fig. 5a) and early secretory (Fig. 5b) endometria expressed the bA subunit with a similar pattern. The maximal immunostaining reaction was observed during late secretory phase (Fig. 5c). Interestingly, a strong synthesis of the bA subunit could be demonstrated at the beginning of secretory phase in stromal cells (Fig. 5d). The immunoreactive score for inhibin-bA increased significantly between proliferative and late secretory phase (P<0.03) as well as between early and late secretory phase (P<0.03) (Fig. 7). Endometrial glandular epithelial cells also express the bB subunit throughout the menstrual cycle. The staining intensity was localised at the basal and apical parts of glands without a distinct expression pattern between proliferative (Fig. 6a), early (Fig. 6b) and late secretory phase (Fig. 6c). In the secretory phase endometrial stromal cells also synthesised the bB subunit (Fig. 6d). The immunoreactive score for inhibin-bB subunit presents statistical significance across the menstrual cycle, being higher during late secretory phase than proliferative and early secretory phase (Fig. 7).

Expression of inhibin-a, -bA and -bB demonstrated by immunofluorescence All three inhibin/activin subunits were detected in normal human endometrial tissue across the menstrual cycle using immunofluorescence, with a polyclonal inhibin-a subunit antibody and monoclonal antibodies for the two distinct inhibin-b subunits. During proliferative phase the inhibin-a subunit was primarily expressed in glandular epithelial cells, with some expression in endometrial blood vessels and minimal detection in stromal cells (Fig. 8a green). It was localised at the apical and basal parts of the gland (Fig. 8a, d). Inhibin-bA was expressed in both glandular and stromal cells (Fig. 8b red). With the triple filter excitation, a coexpression of inhibin-a and -bA subunits could be demonstrated being localised at the basal membrane of glandular cells and the stroma, while in the apical part no or minimal colocalisation could be observed (Fig. 8c yellow). The inhibin-bB subunit could also be demonstrated in glands and stromal cells (Fig. 8e red), while the coexpression showed primary reaction in the stromal compartments with minimal colocalisation of the a and bB subunits in the apical part of the glands (Fig. 8f). In late secretory phase the inhibin-a subunit was primarily expressed in glandular epithelial cells with a higher intensity than proliferative phase. It was localised at the apical and basal parts of the gland (Fig. 9a, d green). Inhibin-bA was expressed in both glandular and stromal cells at the apical and basal parts of glands (Fig. 9b red). A coexpression of inhibin-a and -bA subunits could be demonstrated localised at the basal membrane of glandular cells and the stroma, with

465 Fig. 3af Immunohistochemical staining reaction of inhibin/ activin subunits in normal premenopausal ovary tissue for positive control. Immunohistochemical staining was performed using an appropriate positive control comprising premenopausal ovaries containing follicular cysts. Inhibin-a stained positive with ovarian granulosa cells and cells from the theca interna (a, b), while inhibin-bA and -bB subunits also showed a positive immunohistochemical staining reaction with human ovarian tissue (cf). Magnification a, c, e 125; b, d, f 250

Table 2 Immunohistochemical findings determining localisation and intensity of immunostaining for inhibin-a, -bA and -bB subunits in human endometrium across the normal menstrual cycle. ( No staining, minimal staining, + positive staining, ++ intense staining) Inhibin/activin subunits Immunohistochemistry Glandular epithelium Stromal cells Immunofluorescence Glandular epithelium Stromal cells a bA bB a bA bB a bA bB a bA bB Proliferative phase + + + + + ++ ++ Early secretory phase + + + + + + + ++ + ++ ++ Late secretory phase ++ ++ ++ ++ ++ ++ ++ ++ + + ++ ++

minimal colocalisation observed towards the lumen (Fig. 9c). The inhibin-bB subunit could primarily be demonstrated in stromal cells, with little reaction within the glandular lumina (Fig. 9e red), while the coex-

pression showed strong primary reaction in the stromal compartments with minimal colocalisation of the a and bB subunits in the apical part of the glands (Fig. 9f).

466 Fig. 4ad Immunohistochemical staining reaction of inhibina. During proliferative phase a minimal staining reaction of inhibin-a subunit could be observed at the apical component of glandular cells (a). The staining intensity increased during early secretory phase being localised at the basal and apical parts (b). The strongest staining reaction was observed during late secretory phase (c). With the onset of early secretory phase the a subunit could also be detected in distinct stromal cells (d). Magnification ad 400

Fig. 5ad Immunohistochemical staining reaction of inhibinbA. Proliferative (a) and early secretory (b) endometria expressed the bA subunit with a similar pattern. The maximal immunostaining reaction was observed during late secretory phase (c). A strong synthesis of the bA subunit could be demonstrated at the beginning of secretory phase in stromal cells (d). Magnification ad 400

Discussion
While inhibins/activins were initially characterised as endocrine and paracrine hormonal regulators of the hypothalamicpituitarygonadal axis, it is now clear that

they are expressed in a wide range of tissues including endometrium (Leung et al. 1998; Otani et al. 1998; Jones et al. 2000; Mylonas et al. 2004) and cultured endometrial cells (Petraglia et al. 1998; Mylonas et al. 2003a, b). We have demonstrated the immunohistochemical staining

467 Fig. 6ad Immunohistochemical staining reaction of inhibinbB. The inhibin-bB subunit was mostly localised at the basal and apical parts of glands with a distinct expression pattern between proliferative (a), early (b) and late secretory phase (c), with stromal cells also synthesising this subunit (d). Magnification ad 400

Fig. 7 Immunohistochemical expression of inhibin/activin subunits. The immunoreactive score for inhibin-a increased significantly between proliferative and late secretory phase (*) as well as between proliferative and early secretory phase (**). Additionally, the immunoreactive score increased significantly between early and secretory phase (***). The immunohistochemical staining reaction

for inhibin-bA also increased significantly between proliferative and late secretory phase (*) as well as between early and late secretory phase (**), while inhibin-bB subunit showed similar statistical significance across the menstrual cycle. Significance was assumed at P<0.05

reaction of the inhibin-b and -a subunits in normal endometrial tissue using immunohistochemical means and additionally we showed a coexpression of inhibin-b and -a using immunofluorescence double staining methods. All three inhibin/activin subunits were expressed by normal human endometrial tissue. The synthesis of inhibin and activin subunits in the endometrium varied with the different stages of the menstrual cycle. It has been suggested that epithelial cells are the predominant source

for inhibin/activin subunits in normal endometrium, while stromal cells begin to produce the b subunits with the onset of decidualisation. We could confirm these results by immunohistochemistry, but also showed with the use of immunofluorescence staining techniques that a distinguished expression of these subunits is present in endometrial stroma throughout the menstrual cycle. This discrepancy between both methods might be explained by assuming a very low expression of inhibin-a, which

468 Fig. 8af Expression of inhibin-a, -bA and -bB demonstrated by immunofluorescence during proliferative phase. During proliferative phase the inhibin-a subunit was primarily expressed in glandular epithelial cells, with some expression in endometrial blood vessels and minimal detection in stromal cells (a; green). It was localised at the apical and basal parts of the gland (a, d; green). Inhibin-bA was expressed in both glandular and stromal cells (b; red). At triple filter excitation, a coexpression of inhibin-a and -bA subunits could be demonstrated localised at the basal membrane of glandular cells and the stroma, while in the apical part no colocalisation could be observed (c). The inhibin-bB subunit could also be demonstrated in glands and stromal cells (e; red), while the coexpression showed primary reaction in the stromal compartments with minimal colocalisation of the a and bB subunits in the apical part of the glands (f). Magnification ac 100; df 400

cannot be detected by classic immunohistochemical procedures. Moreover, inhibin-a subunit was detected in the apical compartments of human glandular cells throughout all phases, while inhibin-b subunits were mostly localised in the stroma. This might suggest a substantial function in human implantation of inhibin-a subunit, while the stromal expression of the b subunits could be important in the paracrine signalling for adequate endometrial maturation. Additionally, the distinct expression in glandular epithelial cells of inhibin-a in the apical part also shows a preferred synthesis of inhibins into the lumen and, since its expression is lower in endometrial stroma, suggesting a predominant secretion and production of activins into this compartment. Interestingly, immunofluorescence revealed the localisation of inhibin-a towards the glandular lumina, although just minimal colocalisation with the inhibin-b subunits was observed. The endometrium might be an important source of inhibinactivin dimers present

in the intrauterine compartment during early pregnancy (Petraglia 1997; Riley et al. 2000). These suggestions are in accordance with several recent reports, where isolated epithelial cells secreted both dimeric inhibins and activins (Petraglia et al. 1998). Recent data also indicate that high concentrations of activin A are secreted by stromal cells after in vitro decidualisation (Jones et al. 2002), equivalent to levels detected in maternal serum during the third trimester of pregnancy (Fowler et al. 1998). In early menstrual stages, the a and b subunits were observed in human endometrium, primarily in glandular epithelial cells (Leung et al. 1998; Otani et al. 1998; Jones et al. 2000; Mylonas et al. 2004). Previous analysis showed a higher expression during the late secretory phase (Mylonas et al. 2003a, 2004), while other investigators could not detect the inhibin-a subunit in human endometrium (Otani et al. 1998) or demonstrate an expression without a significant variation (Leung et al.

469 Fig. 9af Expression of inhibin-a, -bA and -bB demonstrated by immunofluorescence during late secretory phase. In the late secretory phase the inhibin-a subunit was primarily expressed in glandular epithelial cells with a higher intensity than proliferative phase (a, d; green). It was localised at the apical and basal parts of the gland (a, d; green). Inhibin-bA was expressed in both glandular and stromal cells at the apical and basal parts of glands (b; red). A coexpression of inhibin-a and -bA subunits could be demonstrated localised at the basal membrane of glandular cells and the stroma, with minimal colocalisation observed towards the lumen (c). The inhibin-bB subunit could primarily be demonstrated in stromal cells, with little reaction within the glandular lumina (e; red), while the coexpression showed strong primary reaction in the stromal compartments with minimal colocalisation of the a and bB subunits in the apical part of the glands (f). Magnification ac 400; df 250

1998). We could demonstrate a significantly higher inhibin-a expression in the secretory phase than in the proliferative phase, confirming previous results (Mylonas et al. 2003a, 2004). Interestingly, a higher concentration of inhibin could be observed during the secretory phase compared to the proliferative phase in culture medium of normal endometrial glandular cell cultures (Mylonas et al. 2003b). This could be due to the different antibodies used in both studies and also compared to studies of other research groups (Leung et al. 1998; Otani et al. 1998; Jones et al. 2000). Additionally we observed significant rising of inhibin-a immunohistochemical staining reaction between proliferative and both early and late secretory phases and between early secretory and late secretory phase. Our previous results did not demonstrate significant differences between early and late secretory phase (Mylonas et al. 2003a, 2004), probably due to the lower amount of analysed cases.

Autocrine and paracrine actions of inhibins/activins have been reported, including modulation of ovarian and placental hormone secretion and local regulation of macrophage function (de Kretser et al. 2002; Welt 2002). The endometrium is a potential target for inhibin/activin action. Possible functions could include endometrial decidualisation (Jones et al. 2002), trophoblast differentiation (Caniggia et al. 1997), immunomodulatory function (Keelan et al. 2000) and local steroidogenesis (Ni et al. 2000). Activin A is a local regulator of cell differentiation and tissue remodelling, with important roles in wound healing (Munz et al. 1999) and mesoderm induction during embryogenesis (Smith et al. 1990). Activin can also inhibit cell proliferation in many human cell types, including breast cancer cell lines (Kalkhoven et al. 1995) and vascular endothelial cells (McCarthy and Bicknell 1993). Additional roles in promoting differentiation and epithelial cell proliferation in other organs have also been

470

described (Robinson and Hennighausen 1997; Li et al. 1998). Therefore, the higher expression of inhibin/activin in late secretory phase might be a major paracrine signal involved during endometrial maturation. Recently, it has been demonstrated that activin A promotes human endometrial stromal cell decidualisation in vitro (Jones et al. 2002). The inhibin-a subunit gene is thought to be a tumour suppressor with gonadal and adrenal specificity (Matzuk et al. 1992). Endometrial adenocarcinomas express significantly lower inhibin-a than hyperplastic endometrium, suggesting an important function in endometrial pathogenesis (Mylonas et al. 2004). The action of inhibin might be related to the influence of gonadotropins on the different tissues, with the gonads, adrenal and placenta being gonadotropin-responsive organs (Risbridger et al. 2001). No specific inhibin receptor has been identified yet, although inhibin competes with activin for binding to the activin receptor type II, but with no stimulation of the activin receptor type I receptor. Inhibin has a low affinity for activin receptor type II (Welt et al. 2002). Recently, some inhibin-specific binding proteins have been identified that are potential receptors, including inhibin-binding protein (INhBP/p120) (Chong et al. 2000) and betaglycan (TGF-b receptor type III; Lewis et al. 2000). Betaglycan can bind inhibin through the a subunit and enhance the interaction of inhibin with activin receptor type II (Lewis et al. 2000). There is substantial evidence supporting the existence of specific inhibin receptors (Robertson et al. 2000). In normal human cultured endometrial glandular cells we could show a relationship between inhibin, oestradiol and cortisol expression (Mylonas et al. 2003a), suggesting an autocrine and paracrine function and regulation in normal human endometrium. In conclusion, all three inhibin/activin subunits were expressed by normal human endometrial tissue. The synthesis of inhibin and activin subunits in the endometrium varied significantly with the different stages of the menstrual cycle. Inhibin-a subunit was detected in the apical compartments of human glandular cells throughout all phases, while inhibin-b subunits were mostly localised in the stroma. Although the precise role of these inhibin/ activin subunits in human endometrium is still unclear, they could be involved in autocrine/paracrine signalling, contributing to several aspects of endometrial maturation such as angiogenesis, decidualisation and tissue remodelling.
Acknowledgements We would like to thank the nurses, medical doctors and laboratory staff for obtaining the endometrial material. This study was supported in part by the FFoLe project of the Ludwig-Maximilians-University Munich for I. Mylonas.

References
Caniggia I, Lye SJ, Cross JC (1997) Activin is a local regulator of human cytotrophoblast cell differentiation. Endocrinology 138:39763986

Chabbert Buffet N, Djakoure C, Maitre SC, Bouchard P (1998) Regulation of the human menstrual cycle. Front Neuroendocrinol 19:151186 Chong H, Pangas SA, Bernard DJ, Wang E, Gitch J, Chen W, Draper LB, Cox ET, Woodruff TK (2000) Structure and expression of a membrane component of the inhibin receptor system. Endocrinology 141:26002607 Dallenbach-Hellweg G, Poulsen H (1985) Atlas der Histopathologie des Endometriums. Springer Berlin, Heidelberg, New York de Kretser DM, Hedger MP, Loveland KL, Phillips DJ (2002) Inhibins, activins and follistatin in reproduction. Hum Reprod Update 8:529541 Fang J, Yin W, Smiley E, Wang SQ, Bonadio J (1996) Molecular cloning of the mouse activin beta E subunit gene. Biochem Biophys Res Commun 228:669674 Fowler PA, Evans LW, Groome NP, Templeton A, Knight PG (1998) A longitudinal study of maternal serum inhibin-A, inhibin-B, activin-A, activin-AB, pro-alphaC and follistatin during pregnancy. Hum Reprod 13:35303536 Groome N, Lawrence M (1991) Preparation of monoclonal antibodies to the beta A subunit of ovarian inhibin using a synthetic peptide immunogen. Hybridoma 10:309316 Groome N, Hancock J, Betteridge A, Lawrence M, Craven R (1990) Monoclonal and polyclonal antibodies reactive with the 132 amino terminal sequence of the alpha subunit of human 32 K inhibin. Hybridoma 9:3142 Htten G, Neidhardt H, Schneider C, Pohl J (1995) Cloning of a new member of the TGF-beta family: a putative new activin beta C chain. Biochem Biophys Res Commun 206:608613 Jeschke U, Richter DU, Hammer A, Briese V, Friese K, Karsten U (2002) Expression of the Thomsen-Friedenreich antigen and of its putative carrier protein mucin 1 in the human placenta and in trophoblast cells in vitro. Histochem Cell Biol 117:219226 Jones RL, Salamonsen LA, Critchley HO, Rogers PA, Affandi B, Findlay JK (2000) Inhibin and activin subunits are differentially expressed in endometrial cells and leukocytes during the menstrual cycle, in early pregnancy and in women using progestin-only contraception. Mol Hum Reprod 6:11071117 Jones RL, Salamonsen LA, Findlay JK (2002) Activin A promotes human endometrial stromal cell decidualization in vitro. J Clin Endocrinol Metab 87:40014004 Kalkhoven E, Roelen BA, de Winter JP, Mummery CL, van den Eijnden-van Raaij AJ, van der Saag PT, van der Burg B (1995) Resistance to transforming growth factor beta and activin due to reduced receptor expression in human breast tumor cell lines. Cell Growth Differ 6:11511161 Keelan JA, Zhou RL, Mitchell MD (2000) Activin A exerts both pro- and anti-inflammatory effects on human term gestational tissues. Placenta 21:3843 Leung PH, Salamonsen LA, Findlay JK (1998) Immunolocalization of inhibin and activin subunits in human endometrium across the menstrual cycle. Hum Reprod 13:34693477 Lewis KA, Gray PC, Blount AL, MacConell LA, Wiater E, Bilezikjian LM, Vale W (2000) Betaglycan binds inhibin and can mediate functional antagonism of activin signalling. Nature 404:411414 Li Q, Karam SM, Coerver KA, Matzuk MM, Gordon JI (1998) Stimulation of activin receptor II signaling pathways inhibits differentiation of multiple gastric epithelial lineages. Mol Endocrinol 12:181192 Matzuk MM, Finegold MJ, Su JG, Hsueh AJ, Bradley A (1992) Alpha-inhibin is a tumour-suppressor gene with gonadal specificity in mice. Nature 360:313319 McCarthy SA, Bicknell R (1993) Inhibition of vascular endothelial cell growth by activin-A. J Biol Chem 268:2306623071 Munz B, Hubner G, Tretter Y, Alzheimer C, Werner S (1999) A novel role of activin in inflammation and repair. J Endocrinol 161:187193 Mylonas I, Speer R, Makovitzky J, Richter DU, Briese V, Jeschke U, Friese K (2000) Immunohistochemical analysis of steroid receptors and glycodelin A (PP14) in isolated glandular epi-

471 thelial cells of normal human endometrium. Histochem Cell Biol 114:405411 Mylonas I, Jeschke U, Winkler L, Makovitzky J, Richter DU, Briese V, Friese K (2003a) Immunohistochemical expression of inhibin-alpha in human endometrium and the in vitro secretion of inhibin, estradiol and cortisol in cultured human endometrial glandular cells. Arch Gynecol Obstet 268:142150 Mylonas I, Winkler L, Jeschke U, Briese V, Friese K (2003b) Investigations on isolation, purification and cultivation of human endometrial cells and on the in vitro inhibin expression in glandular epithelial cells. Zentralbl Gynakol 125:415423 Mylonas I, Makovitzky J, Richter DU, Jeschke U, Briese V, Friese K (2004) Expression of the inhibin-alpha subunit in normal, hyperplastic and malignant endometrial tissue: an immunohistochemical analysis. Gynecol Oncol 93:9297 Ni X, Luo S, Minegishi T, Peng C (2000) Activin A in JEG-3 cells: potential role as an autocrine regulator of steroidogenesis in humans. Biol Reprod 62:12241230 Oda S, Nishimatsu S, Murakami K, Ueno N (1995) Molecular cloning and functional analysis of a new activin beta subunit: a dorsal mesoderm-inducing activity in Xenopus. Biochem Biophys Res Commun 210:581588 Otani T, Minami S, Kokawa K, Shikone T, Yamoto M, Nakano R (1998) Immunohistochemical localization of activin A in human endometrial tissues during the menstrual cycle and in early pregnancy. Obstet Gynecol 91:685692 Petraglia F (1997) Inhibin, activin and follistatin in the human placenta: a new family of regulatory proteins. Placenta 18:38 Petraglia F, Florio P, Luisi S, Gallo R, Gadducci A, Vigano P, Di Blasio AM, Genazzani AR, Vale W (1998) Expression and secretion of inhibin and activin in normal and neoplastic uterine tissues. High levels of serum activin A in women with endometrial and cervical carcinoma. J Clin Endocrinol Metab 83:11941200 Remmele W, Stegner HE (1987) Vorschlag zur einheitlichen Definierung eines immunreaktiven Score (IRS) fr den immunhistochemischen strogenrezeptornachweis (ER-ICA) im Mammakarzinomgewebe. Pathologe 8:138140 Riley SC, Leask R, Balfour C, Brennand JE, Groome NP (2000) Production of inhibin forms by the fetal membranes, decidua, placenta and fetus at parturition. Hum Reprod 15:578583 Risbridger GP, Schmitt JF, Robertson DM (2001) Activins and inhibins in endocrine and other tumors. Endocr Rev 22:836 858 Robertson DM, Hertan R, Farnworth PG (2000) Is the action of inhibin mediated via a unique receptor? Rev Reprod 5:131135 Robinson GW, Hennighausen L (1997) Inhibins and activins regulate mammary epithelial cell differentiation through mesenchymal-epithelial interactions. Development 124:27012708 Smith JC, Price BM, Van Nimmen K, Huylebroeck D (1990) Identification of a potent Xenopus mesoderm-inducing factor as a homologue of activin A. Nature 345:729731 Vale W, Rivier C, Hsueh A, Campen C, Meunier H, Bicsak T, Vaughan J, Corrigan A, Bardin W, Sawchenko P, et al (1988) Chemical and biological characterization of the inhibin family of protein hormones. Recent Prog Horm Res 44:134 Welt CK (2002) The physiology and pathophysiology of inhibin, activin and follistatin in female reproduction. Curr Opin Obstet Gynecol 14:317323 Welt C, Sidis Y, Keutmann H, Schneyer A (2002) Activins, inhibins, and follistatins: from endocrinology to signaling. A paradigm for the new millennium. Exp Biol Med (Maywood) 227:724752 Ying SY (1988) Inhibins, activins, and follistatins: gonadal proteins modulating the secretion of follicle-stimulating hormone. Endocr Rev 9:267293

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.