- Email: [email protected]
Expression of inhibins, activins, insulin-like growth factor-I and steroidogenic enzymes in the equine placenta
Domestic Animal Endocrinology 31 (2006) 19–34
Expression of inhibins, activins, insulin-like growth factor-I and steroidogenic enzymes in the equine placenta Koji Y. Arai a , Yumiko Tanaka b , Hiroyuki Taniyama d , Noboru Tsunoda e , Yasuo Nambo f , Natsuko Nagamine b , Gen Watanabe b,c , Kazuyoshi Taya b,c,∗ a
Department of Tissue Physiology, Tokyo University of Agriculture and Technology, Tokyo 183-8509, Japan b Laboratory of Veterinary Physiology, Faculty of Agriculture, Tokyo University of Agriculture and Technology, Fuchu, Tokyo 183-8509, Japan c Department of Basic Veterinary Science, The United Graduate School of Veterinary Science, Gifu University, Gifu 501-1193, Japan d Department of Veterinary Pathology, Rakuno Gakuen University, Hokkaido 069-8501, Japan e Shadai Corporation, Hokkaido 059-1432, Japan f Hidaka Training Research Center, Japan Racing Association, Hokkaido 057-171, Japan Received 29 June 2005; received in revised form 25 August 2005; accepted 1 September 2005
Abstract In this study, the expression patterns of inhibins, activins, insulin-like growth factor-I (IGF-I) and steroidogenic enzymes in equine placentae recovered during the latter two-thirds of gestation were examined. Concentrations of inhibin A and inhibin pro-␣C in endometrial and fetal placental tissue homogenates were very low during the period examined, whereas these tissues contained high concentrations of activin A. In both maternal endometrial and fetal placental tissues, activin A levels decreased as pregnancy progressed. Expression of inhibin ␣-subunit was not observed in the placenta using either immunohistochemistry or in situ hybridization. Inhibin/activin A-subunit and its mRNA were confined to maternal endometrial glands, whereas immunopositive B-subunit was not detected in either endometrial glands or microcotyledons. Cytochrome P450 side chain cleavage enzyme was detected by immunohistochemistry in both endometrial glands and microcotyledons, whereas cytochrome P450 17␣-hydroxylase/lyase was absent in these tissues. Immunopositive signals for 3∗
Corresponding author. Tel.: +81 423 67 5767; fax: +81 423 67 5767. E-mail address: [email protected] (K. Taya).
0739-7240/$ – see front matter © 2005 Published by Elsevier Inc. doi:10.1016/j.domaniend.2005.09.005
20
K.Y. Arai et al. / Domestic Animal Endocrinology 31 (2006) 19–34
hydroxysteroid dehydrogenase and cytochrome P450 aromatase were localized in microcotyledons but not in endometrial glands. Immunohistochemistry revealed that IGF-I was highly expressed in microcotyledons around Day 130, and decreased as pregnancy progressed. Changes in the expression of IGF-I were correlated with the number of PCNA positive cells in the placenta. The present study demonstrated the presence and localized the site of expression of activin, IGF-I and steroidogenic enzymes in equine placental tissues during the latter two-thirds of gestation; the results suggest that activin and IGF-I may be involved in the regulation of placental development. © 2005 Published by Elsevier Inc. Keywords: Inhibin; Activin; IGF-I; Steroidogenic enzymes; Equine placenta
1. Introduction The equine placenta is diffuse, non-invasive and epitheliochorial. The trophoblastic villi are organized into complexes known as microcotyledons, which cover the surface of the allantochorion and fit into endometrial crypts [1,2]. Both fetal and maternal tissues of the equine placneta are important for fetal nutrition, respiration and waste removal [3,4]. The placenta is also an endocrine and paracrine organ, which secretes diverse hormones and growth factors to maintain pregnancy and to support fetal development. Inhibins, which belong to the transforming growth factor- (TGF-) superfamily, were initially isolated as gonadal peptides that inhibit FSH secretion from the pituitary gland [5–8]. Two inhibins have been identified. Each inhibin is composed of a common ␣-subunit and either a A-subunit to give rise inhibin A or a B-subunit to form inhibin B [5–8]. Homo- and hetero-dimers of the -subunits of inhibin are called activin A, activin AB and activin B [5–8], and are multifunctional growth factors. Inhibin/activin subunits have been detected in the placenta in humans and several animal species [9–14]. It has been proposed that placental function and development are, at least in part, regulated by activin produced locally because activin has been shown to affect functions of human placental cells [15–17]. The presence of A-subunit mRNA and the absence of ␣-subunit mRNA in the equine placenta have been shown by Yamanouchi et al. [14]. However, it is not known whether the equine placenta expresses inhibin/activin B-subunit and whether it secretes inhibin/activin dimers. In addition, contribution of the placenta to the high concentrations of inhibins in equine fetal plasma [18–20] is not known, although fetal gonads are likely the major source of inhibins in fetal circulation. Insulin-like growth factors (IGFs) are polypeptides of approximately 7 kDa that have structural homology to proinsulin and their biological actions are regulated by a family of high affinity binding proteins (IGFBPs) [21]. IGFs and IGFBPs are expressed in the placenta and are important in regulating fetal and placental growth in humans and other mammalian species [22,23]. A previous study has shown that IGF-II null mice but not IGF-I null mice show smaller placentae than those of wild type mice [24]. Although placental growth is not decreased in IGF-I null mice, the mice show slower development during prenatal and postnatal periods [24]. Expression patterns of IGFs, IGF receptors and IGFBPs are different between species, even though the species are very close [23]. Despite the accumulating evidences in other species, expression patterns of IGFs in the equine pla-
K.Y. Arai et al. / Domestic Animal Endocrinology 31 (2006) 19–34
21
centa have not been reported except for IGF-II expression during the first half of gestation [25]. The equine placenta is known to secrete various steroid hormones. Progestagens are known to be required for maintaining pregnancy. After regression of maternal corpora lutea in mid-gestation, the equine placenta is the main source of progestagens during the second half of pregnancy [26,27]. In addition, high concentrations of estrogens are present in the blood and urine during the latter two-thirds of gestation in the mare [26]. They include estrone, estradiol-17, equilin and equilenin, all of which derive from placental aromatization of C-19 precursors secreted by the fetal gonads [28–32]. The roles of these large quantities of placental estrogens are not altogether clear. However, the fetal gonadectomy studies of Pashen and Allen [31] have indicated that estrogen production in equine pregnancy may be important to stimulate the development of both endometrial and placental blood vessels. A previous study examined localization of cytochrome P450 cholesterol side chain cleavage enzyme (P450scc), 3-hydroxysteroid dehydrogenase (3-HSD) and cytochrome P450 17␣-hydroxylase/lyase (P450cl7) in the equine placenta and demonstrated that P450cl7 was not detected in this tissue, whereas the other two enzymes were present [33]. These results confirmed the importance of androgens from the fetal gonads for placental estrogen synthesis [34–36]. With respect to P450 aromatase (P450arom), histological localization of this enzyme in the placenta has not been examined, although the equine placenta contains high levels of aromatase activity [37]. In this study, we focused our attention on inhibins, activins, IGF-I and steroidogenic enzymes and examined their expression in the equine placenta to elucidate physiological features of the equine placenta as an endocrine and paracrine organ.
2. Materials and methods 2.1. Animals Placentae were collected from normal pregnant Thoroughbred and Anglo-Arab mares at Days 100–312 of gestation after euthanasia. Final mating day was designated as Day 0 of gestation. The pregnant mares were killed using an overdose of a mixture of barbiturate (Ravonal; Tanabe Pharmaceutical Co. Ltd., Osaka, Japan) and suxamethonium chloride (Succine; Yamanouchi Pharmaceutical Co. Ltd., Osaka, Japan), prior to tissue recovery. Fetal gonads were also collected as previously described [18–20]. The fetal gonad samples contained both ovaries and testes, which were not distinguished from each other. All procedures were carried out in accordance with the guidelines established by the Rakuno Gakuen University for use of laboratory animals. 2.2. Tissue samples Maternal endometrial and fetal placental (allantochorionic) tissues were separated manually from the placentae before homogenization. To measure concentrations of inhibins and activin A in maternal endometrial tissues, fetal placental tissues and fetal gonads, a portion
22
K.Y. Arai et al. / Domestic Animal Endocrinology 31 (2006) 19–34
of the tissue (1.0 g) was homogenized in 2 ml of 0.01 M phosphate-buffered saline (PBS, pH 7.4) on ice using a homogenizer (Phiscotron, Nichion, Tokyo, Japan) for 30 s at 20,000 rpm. The homogenized tissue samples were centrifuged at 20,000 × g for 30 min at 4 ◦ C. The supernatant was removed and stored at −80 ◦ C until assayed for inhibins and activin A. The homogenized samples were divided into the following groups; Days 100–140 (n = 3); Days 140–180 (n = 11); Days 180–220 (n = 8); Days 220–260 (n = 3) of gestation. For histochemical studies, six placentae recovered at different stages (Days 130, 132, 162, 208, 227 and 312 of gestation) were fixed in freshly prepared 4% (w/v) paraformaldehyde (PFA; Sigma Chemical Co., St. Louis, MO) in 0.01 M PBS and embedded in paraffin. Five- and 8-m-thick tissue sections were prepared for immunohistochemistry and in situ hybridization, respectively. These sections were placed on glass slides coated with 3-aminopropyltriethoxysilane. 2.3. ELISA Concentrations of inhibin pro-␣C in tissue homogenates were measured as previously described [19,20] using a two-site ELISA kit (Serotec, Oxford, UK) designed for measurement of human inhibin pro-␣C [38]. This assay system does not detect mature ␣-subunit monomer or dimers that contain mature ␣-subunit, but does detect pro-␣C monomer, ␣subunit precursor (pro-␣N␣C) and dimers that contain either pro-␣C peptide or ␣-subunit precursor. The amount of inhibin pro-␣C was expressed in terms of the purified inhibin pro␣C derived from human follicular fluid. Concentrations of inhibin A in tissue homogenates were measured using a two-site ELISA kit (Serotec) as previously described [19,20]. The amount of inhibin A was expressed in terms of the purified 32-kDa inhibin A derived from bovine follicular fluid. These assays have previously been validated for equine samples [19,20]. Concentrations of activin A in tissue homogenates were measured using a two-site ELISA kit (Serotec) [39]. Amounts of activin A were expressed in terms of recombinant activin A. Serial dilutions of pooled equine maternal endometrial and fetal placental homogenates produced dose–response curves that were parallel to the standard curve produced with activin A (Fig. 1). The intra- and interassay coefficients of variation were 3.3 and 12.6% for inhibin pro-␣C, 4.2 and 8.7% for inhibin A and 4.4 and 5.1% for activin A, respectively. The sensitivities of the assays were 3.0 pg/ml for inhibin pro-␣C, 15.6 pg/ml for inhibin A and 7.3 pg/ml for activin A, respectively. 2.4. Immunohistochemistry Immunohistochemistory was done as previously described [20]. After deparaffinization with xylene, the tissue sections were subjected to antigen retrieval by autoclaving in 0.01 M sodium citrate buffer, pH 6.0, at 121 ◦ C for 15 min. Sections were then incubated in 3% H2 O2 in methanol at room temperature for 15 min, followed by incubation in Block Ace (Dainippon Pharmaceutical Co. Ltd., Osaka, Japan) at 37 ◦ C for 1 h to quench non-specific staining. Then they were incubated at 37 ◦ C for 16–18 h with primary antibodies diluted in Block Ace. Primary antibodies used in the present study were a rabbit polyclonal antibody to [Tyr30]-porcine inhibin ␣ chain (1–30)-NH2 conjugated to rabbit serum albumin [40] (diluted 1:4,000; [Tyr30]-porcine inhibin ␣ chain (1–30)-NH2 was provided by Dr. Nicholas Ling, Neuroendocrine Inc., San Diego, CA), a monoclonal antibody raised against the
K.Y. Arai et al. / Domestic Animal Endocrinology 31 (2006) 19–34
23
Fig. 1. Validation of the activin A ELISA system for equine samples. Different dilutions of equine maternal endometrial tissue homogenates (squares) and fetal placental tissue homogenates (triangles) produced dose–response curves that were parallel to the standard curve (circles) produced with recombinant activin A. Values are means ± S.D. of duplicate determinations. Error bars of some values are hidden by the marks.
synthetic A-subunit (E4) [41] (final concentration, 4.5 g/ml; Provided by Dr. Nigel P. Groome, Oxford Brookes University, Oxford, UK), a monoclonal antibody raised against the synthetic human B-subunit (C5) [42] (final concentration, 4 g/ml; Provided by Dr. Nigel P. Groome), a rabbit polyclonal antibody to rat P450scc [43] (diluted 1:3000; provided by Dr. Michael J. Soares, University of Kansas Medical Center, Kansas City, KS), a rabbit polyclonal antibody to rat P450cl7 [44] (diluted 1:1000; provided by Dr. Donald C. Johnson, University of Kansas Medical Center), a rabbit polyclonal antibody to human placental 3-HSD [45] (diluted 1:1000; provided by Dr. J. Ian Mason, University of Edinburgh, Edinburgh, Scotland), a rabbit polyclonal antibody to aromatase (diluted 1:1000; purchased from Biogenesis, England, UK), a purified rabbit polyclonal antibody to human IGF-I (final concentration, 10 g/ml; purchased form AUSTRAL Biologicals, San Ramon, CA), and a mouse monoclonal antibody to human proliferating cell nuclear antigen (PCNA) (final concentration, 4 g/ml; purchased from Biomeda corp, Foster City, CA). After incubation with the primary antibodies, sections were treated with 0.25% (v/v) biotinylated goat antirabbit or rabbit anti-mouse secondary antibody (Elite ABC kit; Vector Labs, Burlingame, CA) in Block Ace at 37 ◦ C for 1 h. These sections were subsequently incubated with 2% (v/v) avidin–biotin complex (Elite ABC kit) in casein-Tris-saline (CTS) at 37 ◦ C for 30 min. The reaction products were visualized by treating with 0.025% (w/v) 3,3 -diaminobenzidine tetrachloride (Sigma) in 100 mM Tris-buffered saline containing 0.01% H2 O2 for 1–30 min. As a negative control, the primary antibody was replaced with normal rabbit serum, normal rabbit IgG or normal mouse IgG. The sections were counter-stained with haematoxylin.
24
K.Y. Arai et al. / Domestic Animal Endocrinology 31 (2006) 19–34
2.5. Complementary RNA probes and in situ hybridization The 1286-base pair (bp) equine inhibin ␣-subunit cDNA clone [46] and the 1479-bp equine inhibin/activin A-subunit cDNA clone [47] (provided by Dr. Keitaro Yamanouchi, University of Tokyo, Tokyo, Japan) were subcloned into a pBluescript KS(−) vector (Stratagene, La Jolla, CA). Template inhibin ␣ and inhibin/activin A cDNAs represent 309-bp subclones corresponding to the 5 region from the first restriction site PstI and 325-bp subclones corresponding to the 5 region from the first restriction site HindIII, respectively. Digoxigenin (DIG)-labeled RNA probes were synthesized with DIG RNA labeling mixture (Roche Diagnostics, Tokyo, Japan) according to the manufacturer’s protocols. In situ hybridization was done as previously described [20] using the DIG-labeled RNA probes. 2.6. Statistical analysis To compare inhibin and activin levels in tissue samples, one-way analysis of variance (ANOVA) was carried out. When a significant effect was obtained with analysis of variance, the significance of the difference between means was determined by a Duncan multiple
Fig. 2. Concentrations of inhibin pro-␣C (A and B) and inhibin A (C and D) in maternal endometrial tissues (A and C), fetal placental tissues (A and C) and fetal gonads (B and D) at Days 100–140 (n = 3), Days 140–180 (n = 11), Days 180–220 (n = 8) and Days 220–260 (n = 3) of gestation. Values represent means ± S.E.M. ND, not detected.
K.Y. Arai et al. / Domestic Animal Endocrinology 31 (2006) 19–34
25
range test. Data are presented as means ± S.E.M. A probability value less than 0.05 was considered to be significant.
3. Results 3.1. Concentrations of inhibin pro-αC, inhibin A and activin A in maternal endometrial and fetal placental tissues Inhibin pro-␣C was not detected in fetal placental tissues recovered between Days 100 and 260 of gestation (Fig. 2A). Concentrations of inhibin pro-␣C in maternal endometrial homogenates were much lower than those in homogenates of fetal gonads (Fig. 2A and B) and neither changed significantly during the period of pregnancy that we examined. Inhibin A was detected in both maternal endometrial and fetal placental tissues (Fig. 2C). However, the levels were much lower than that in homogenates of fetal gonads (Fig. 2D). Inhibin A levels in fetal placental tissues, maternal endometrial tissues or fetal gonads did not show significant change during the period studied, although inhibin A levels in fetal gonads tended to decrease as pregnancy progressed. Activin concentrations in fetal gonads were very low during the period that we examined (Fig. 3). By contrast, high levels of activin A were detected in maternal endometrial and fetal placental tissues that were obtained between Days 100 and 140 of gestation (Fig. 3). Levels of activin A in fetal placental tissues significantly decreased between Days 100 and 180, and remained low thereafter. Levels of
Fig. 3. Concentrations of activin A in fetal gonads, maternal endometrial tissues and fetal placental tissues at Days 100–140 (n = 3), Days 140–180 (n = 11), Days 180–220 (n = 8) and Days 220–260 (n = 3) of gestation. Values are means ± S.E.M. Columns marked with the same letter at each stage did not differ significantly (P > 0.05).
26
K.Y. Arai et al. / Domestic Animal Endocrinology 31 (2006) 19–34
activin A in maternal endometrial tissues gradually decreased as pregnancy progressed, although the change is not statistically significant. 3.2. Localization of inhibin/activin subunit mRNAs in the equine placenta Sections at Day 208 of gestation hybridized with ␣- or A-subunit probe are shown in Fig. 4. In all six placentae examined (Days 130, 132, 162, 208, 227 and 312 of gestation),
Fig. 4. Localization of inhibin ␣- (A) and inhibin/activin A-subunit (C and E) mRNAs in the equine placenta at Day 208 of gestation. Only A-subunit mRNA was detected in maternal endometrial glands. Sense controls for inhibin ␣-subunits (B) and inhibin/activin A-subunits (D) show no signal. Bar = 100 m. EG, endometrial glands. Cot, microcotyledons.
K.Y. Arai et al. / Domestic Animal Endocrinology 31 (2006) 19–34
27
inhibin ␣-subunit mRNA was not detected in either endometrial glands or microcotyledons (Fig. 4A). On the other hand, endometrial glands were strongly positive for A-subunit mRNA (Fig. 4C and E) during the period investigated. Negative control sections hybridized with sense probes showed no signal (Fig. 4B and D).
Fig. 5. Photomicrographs of histological sections of the equine placenta at Day 208 of gestation. Tissue sections were immunostained for inhibin subunits and steroidogenic enzymes. A tissue section stained with hematoxylin and eosin is given in (A). Tissue sections incubated with antibodies to inhibin A-subunit (B), P450scc (C), 3-HSD (D), P450arom (E) are shown. Inhibin ␣-subunit, B-subunit and P450cl7 were not detected in the equine placenta (data not shown). Representative negative control sections for immunohistochemistry are shown in Fig. 6G and H, respectively. Bar = 100 m. EG, endometrial glands. Cot, microcotyledons.
28
K.Y. Arai et al. / Domestic Animal Endocrinology 31 (2006) 19–34
K.Y. Arai et al. / Domestic Animal Endocrinology 31 (2006) 19–34
29
3.3. Immunohistochemical localization of inhibin/activin subunits and steroidogenic enzymes in the equine placenta Expression patterns of inhibin/activin subunits and steroidogenic enzymes did not change significantly throughout the period of gestation examined (Days 130, 132, 162, 208, 227 and 312). Sections at Day 208 of gestation are shown in Fig. 5 as representatives. A section stained with hematoxylin and eosin is also shown in Fig. 5A. Immunoreactivity for inhibin ␣-subunit was not detected in either endometrial glands or microcotyledons at any stage of gestation investigated (data not shown). Immunopositive staining of inhibin/activin A-subunit was observed in endometrial glands but not in microcotyledons (Fig. 5B). These results are consistent with the expression patterns of mRNAs for these subunits. Inhibin/activin B-subunit was not detected in either endometrial glands or microcotyledons (data not shown). P450scc was localized in both endometrial glands and microcotlyledons (Fig. 5C). Immunopositive staining of P450scc in endometrial glands was more intense than in the microcotyledons. P450cl7 was not detected in either endometrial glands or microcotyledons (data not shown). Both 3-HSD and P450arom were localized in microcotyledons but not in endometrial glands (Fig. 5D and E). 3.4. Localization of IGF-I and PCNA in the equine placenta during pregnancy Sections at Days 130, 208 and 312 of gestation stained with specific antibodies to IGF-I and PCNA are shown in Fig. 6. IGF-I immunoreactivity was detected in microcotyledons throughout at all stages investigated (Days 130, 132, 162, 208, 227 and 312 of gestation), whereas no signal was observed in endometrial glands at any stage (Fig. 6A, C and E). Signals for IGF-I were most intense around Day 130 of gestation and decreased toward parturition. Similarly, PCNA was localized in nuclei of microcotyledons during the latter two-thirds of pregnancy (Fig. 6B, D and F). More than 95% trophoblast cells were positively stained with anti-PCNA antibody around Day 130, whereas PCNA-positive trophoblast cells decreased to 80–85% around Day 200 and to about 40% at Day 312. Fig. 6G and H are negative control sections for immunohistochemistry shown in Figs. 5 and 6.
4. Discussion In the present study, high concentrations of activin A were detected in equine endometrial and fetal placental tissues, whereas levels of inhibins in these tissues were very low. Because inhibin/activin A-subunit and its mRNA were detected in endometrial glands but not in fetal placental tissues, activin A in the fetal placental tissue homogenates had proba-
Fig. 6. Immunohistochemical staining of IGF-I (A, C and E) and PCNA (B, D and F) in the equine placenta at Days 130 (A and B), 208 (C and D) and 312 (E and F). Representative negative control sections incubated with normal rabbit IgG (G, 10 g/ml) and normal mouse IgG (H, 4 g/ml) instead of the primary antibodies are also shown. Sections incubated with normal rabbit serum showed similar results. Bar = 100 m. EG, endometrial glands. Cot, microcotyledons.
30
K.Y. Arai et al. / Domestic Animal Endocrinology 31 (2006) 19–34
bly diffused across from the endometrial glands. Although neither inhibin ␣-subunit protein nor its mRNA was expressed in the equine placenta, low levels of inhibin A was detected in both maternal endometrial and fetal placental tissues. Maternal endometrial tissues also contained low concentrations of inhibin pro-␣C. Our previous studies have shown that large amounts of inhibin pro-␣C and inhibin A were present in fetal circulation, whereas levels of inhibins in maternal circulation were very low [19,20]. Therefore, inhibins, which were detected in the placental tissues, were probably derived from fetal circulation. However, the endometrial inhibin pro-␣C might come from maternal circulation because fetal placental homogenates did not contain detectable inhibin pro-␣C. The present results strongly indicated that the large amounts of inhibins in fetal circulation [19,20] were not derived from placentae but secreted form fetal gonads because fetal gonadal homogenates were abundant in inhibins. A previous study [14] has shown that levels of A-subunit mRNA in the placenta during the mid one-third of gestation were higher than those during the last one-third of gestation. The previous results are thus consistent with our present findings of decreasing activin A in the placental tissues as pregnancy progressed. On the other hand, inhibin B-subunit was not detected in the equine placenta by immunohistochemistry. Because the anti-B-subunit antibody used in the present study did detect inhibin B-subunit in the equine ovary [40], the present result indicates that the equine placenta does not express or expresses very low levels of B-subunit. Activin has been shown to affect secretion of progesterone, hCG and GnRH from human placental cells [15,16]. Therefore, activins may be involved in the regulation of hormone secretion form the equine placenta in a paracrine or autocrine fashion. On the other hand, a previous study has demonstrated that activin stimulates outgrowth of human cytotrophoblast cells and increases matrix metalloproteinase release and fibronectin synthesis by these cells [17]. Because attachment of the equine fetal placenta occurs gradually over Days 40–150 [1], placental activin may promote placentation during the first half of pregnancy. Relatively high concentrations of activin A in the placental tissues at Days 100–140 and the following decrease support this hypothesis. Further studies should be required to elucidate roles of activin in the equine placenta, such as histochemical studies for detecting activin receptors to reveal activin-responsive cells in the placenta. Furthermore, expression patterns of activin before Day 100 of gestation should be examined because attachment of the fetal placenta starts before Day 100. In the present study, P450cl7 was not detected in the equine placenta, whereas 3-HSD was observed in microcotyledons. These results are consistent with the previous report [33]. Localization of P450scc in both endometrial glands and microcotyledons was also observed previously [33]. In the previous paper, however, microcotyledons were more strongly stained with anti-P450scc antibodies than endometrial glands, the exact opposite of our findings. We do not know the reason of the discrepancy. One possibility is that the different antibodies used may be responsible for the discordance between the present and previous studies. Marshall et al. have reported that high aromatase activity was detected in the fetal allantochorionic tissue, whereas aromatase activity in the maternal endometrial tissue was low [37]. They also described that the maternal tissue contained appreciable amounts of fetal chorion and the contaminating fetal tissue might be responsible for the detectable aromatase activity in the maternal tissue. Our immunohistochemical study clearly showed that P450arom was localized in fetal tissues but not in maternal tissues, and supports the previous observation. These expression patterns of steroidogenic enzymes support the previously proposed
K.Y. Arai et al. / Domestic Animal Endocrinology 31 (2006) 19–34
31
feto–placental interaction for steroidogenesis. That is, the equine placenta is able to secrete progestagens, but requires androgens from fetal gonads as precursors for estrogen synthesis. IGF-I was found in microcotyledons throughout the latter two-thirds of pregnancy. Signals for IGF-I were strong around Day 130, when placental attachment would still have been in progress, but decreased thereafter when placental attachment would have been completed (about Day 150 of pregnancy). The number of PCNA-positive cells also decreased between Days 130 and 312. Previous studies have demonstrated that IGF-I stimulates proliferation [48–51] and migration [48–50] of mouse and human placental cells. Furthermore, maternal IGF-I levels are positively correlated with placental mass in humans [52]. The present results and the previous studies suggest that equine placental IGF-I may stimulate proliferation of placental cells and promote placentation in a paracrine fashion. However, further studies are required to elucidate physiological roles of IGF-I in the equine placental development. We only examined samples obtained after Day 130 of gestation, although attachment of the equine fetal placenta occurs over Day 40–150 [1]. Furthermore, expression patterns of IGF receptors and IGFBPs should be examined to estimate sensitivity of the cells and bioavailability of IGF-I, respectively. In addition, localization of IGF-I mRNA should be examined because IGF-I, which was detected in microcotyledons by immunohistochemistry, might come form other sources. In summary, the present study is the first to demonstrate the localization of IGF-I and aromatase in the equine placenta, and secretion of activin A by this tissue. The present results suggest involvement of activin A and IGF-I in the regulation of placental development and confirmed importance of fetal androgens for the synthesis of placental estrogens in the mare.
Acknowledgements We are grateful to Dr. Nicholas Ling, Neuroendocrine Inc., San Diego, CA, for [Tyr30]porcine inhibin ␣ chain (1–30)-NH2 ; Dr. Nigel P. Groome, Oxford Brookes University, Oxford, UK, for antibodies to human inhibin/activin A- and B-subunits; Dr. Michael J. Soares, University of Kansas Medical Center, Kansas City, KS, for antibodies to P450scc; Dr. Donald Johnson, University of Kansas Medical Center, Kansas City, KS, for antibodies to rat P450cl7; Dr. J. Ian Mason, University of Edinburgh, Edinburgh, Scotland, for antibodies to human placental 3-HSD; to Dr. Keitaro Yamanouchi, University of Tokyo, Tokyo, Japan, for equine inhibin/activin cDNAs. Grants: This work was supported in part by the Ito Foundation (to K.Y.A.), a Grant-in-Aid for Scientific Research (The 21st Century Centerof-Excellence Program, E-1) from the Ministry of Education, Culture, Sports, Science and Technology of Japan (to K.T.) and a Grant-in-aid from the Equine Research Institute of the Japan Racing Association (to K.T.).
References [1] Samuel CA, Allen WR, Steven DH. Ultrastructural development of the equine placenta. J Reprod Fertil 1975;(Suppl. 23):575–8.
32
K.Y. Arai et al. / Domestic Animal Endocrinology 31 (2006) 19–34
[2] Samuel CA, Allen WR, Steven DH. Studies on the equine placenta. III. Ultrastructure of the uterine glands and the overlying trophoblast. J Reprod Fertil 1977;51:433–7. [3] Silver M. Some aspects of equine placental exchange and foetal physiology. Equine Vet J 1984;16:227–33. [4] Antczak DF, Oriol JG, Donaldson WL, Poleman C, Stenzler L, Volsen SG, et al. Differentiation molecules of the equine trophoblast. J Reprod Fertil 1987;(Suppl. 35):371–8. [5] de Jong FH. Inhib Physiol Rev 1988;68:555–607. [6] Ying S-Y. Inhibins, activins, and follistatins: gonadal proteins modulating the secretion of follicle-stimulating hormone. Endocr Rev 1988;9:267–93. [7] Mather JP, Woodruff TK, Krummen LA. Paracrine regulation of reproductive function by inhibin and activin. Proc Soc Exp Biol Med 1992;201:1–15. [8] Mather JP, Moore A, Li RH. Activins, inhibins, and follistatins: further thoughts on a growing family of regulators. Proc Soc Exp Biol Med 1997;215:209–22. [9] Meunier H, Rivier C, Evans RM, Vale W. Gonadal and extragonadal expression of inhibin alpha, beta A, and beta B subunits in various tissues predicts diverse functions. Proc Natl Acad Sci USA 1988;85:247–51. [10] Billiar RB, Rohan R, Henson MC, Smith P, Babischkin J. Regulation of immunoreactive inhibin patterns in baboon pregnancy: maternal, placental, and fetal considerations. J Clin Endocrinol Metab 1992;75: 1345–51. [11] Petraglia F, Woodruff TK, Botticelli G, Botticelli A, Genazzani AR, Mayo KE, et al. Gonadotropin-releasing hormone, inhibin, and activin in human placenta: evidence for a common cellular localization. J Clin Endocrinol Metab 1992;74:1184–8. [12] Rabinovici J, Goldsmith PC, Librach CL, Jaffe RB. Localization and regulation of the activin-A dimer in human placental cells. J Clin Endocrinol Metab 1992;75:571–6. [13] Roberts VJ, Barth SL. Expression of messenger ribonucleic acids encoding the inhibin/activin system during mid- and late-gestation rat embryogenesis. Endocrinology 1994;134:914–23. [14] Yamanouchi K, Hirasawa K, Hasegawa T, Ikeda A, Chang KT, Matsuyama S, et al. Equine inhibin/activin A-subunit mRNA is expressed in the endometrial gland, but not in the trophoblast, during pregnancy. Mol Reprod Dev 1997;47:363–9. [15] Petraglia F, Vaughan J, Vale W. Inhibin and activin modulate the release of gonadotropin-releasing hormone, human chorionic gonadotropin, and progesterone from cultured human placental cells. Proc Natl Acad Sci USA 1989;86:5114–7. [16] Steele GL, Currie WD, Yuen BH, Jia XC, Perlas E, Leung PC. Acute stimulation of human chorionic gonadotropin secretion by recombinant human activin-A in first trimester human trophoblast. Endocrinology 1993;133:297–303. [17] Caniggia I, Lye SJ, Cross JC. Activin is a local regulator of human cytotrophoplast cell differentiation. Endocrinology 1997;138:3976–86. [18] Nambo Y, Nagata S, Oikawa M, Yoshihara T, Tsunoda N, Kohsaka T, et al. High concentrations of immunoreactive inhibin in the plasma of mares and fetal gonads during the second half of pregnancy. Reprod Fertil Dev 1996;8:1137–45. [19] Tanaka Y, Taniyama H, Tsunoda N, Shinbo H, Nagamine N, Nambo Y, et al. The testis as a major source of circulating inhibins in the male equine fetus during the second half of gestation. J Androl 2002;23:229–36. [20] Tanaka Y, Taniyama H, Tsunoda N, Herath CB, Nakai R, Shinbo H, et al. Localization and secretion of inhibins in the equine fetal ovaries. Biol Reprod 2003;68:328–35. [21] Clemmons DR. Insulin-like growth factor binding proteins and their role in controlling IGF actions. Cytokine Growth Factor Rev 1997;8:45–62. [22] Giudice LC, Mark SP, Irwin JC. Paracrine actions of insulin-like growth factors and IGF binding protein1 in non-pregnant human endometrium and at the decidual-trophoblast interface. J Reprod Immunol 1998;39:133–48. [23] Han VK, Carter AM. Spatial and temporal patterns of expression of messenger RNA for insulin-like growth factors and their binding proteins in the placenta of man and laboratory animals. Placenta 2000;21:289–305. [24] Baker J, Liu JP, Robertson EJ, Efstratiadis A. Role of insulin-like growth factors in embryonic and postnatal growth. Cell 1993;75:73–82. [25] Lennard SN, Stewart F, Allen WR. Insulin-like growth factor II gene expression in the fetus and placenta of the horse during the first half of gestation. J Reprod Fertil 1995;103:169–79.
K.Y. Arai et al. / Domestic Animal Endocrinology 31 (2006) 19–34
33
[26] Ginther OJ. Endocrinology of the pregnant mare. In: Reproductive biology of the mare: basic and applied aspects. 2nd ed. Wisconsin: Equiservices; 1991. p. 419–56. [27] Ousey JC, Forhead AJ, Rossdale PD, Grainger L, Houghton E, Fowden AL. Ontogeny of uteroplacental progestagen production in pregnant mares during the second half of gestation. Biol Reprod 2003;69: 540–8. [28] Bhavnani BR, Short RV, Solomon S. Formation of oestrogens by the pregnant mare. I. Metabolism of 7-3H-dehydroisoandrosterone and 4-14C-androstenedione injected into the umbilical vein. Endocrinology 1969;85:1172–9. [29] Bhavnani BR, Short RV, Solomon S. Formation of estrogens by the pregnant mare. II. Metabolism of 14Cacetate and 3H-cholesterol injected into the fetal circulation. Endocrinology 1971;89:1152–7. [30] Bhavnani BR, Short RV. Formation of steroids by the pregnant mare. IV Metabolism of 14C-mevalonic acid and 3H-dehydroisoandrosterone injected into the fetal circulation. Endocrinology 1973;92:1397–404. [31] Pashen RL, Allen WR. The role of the fetal gonads and placenta in steroid production, maintenance of pregnancy and parturition in the mare. J Reprod Fertil 1979;(Suppl. 27):499–509. [32] Tait AD, Santikarn LC, Allen WR. Identification of 3-hydroxy-5,7 pregnandien-20- one and 3-hydroxy-5,7 androstadien-17-one as endogenous steroids in the foetal horse gonad. J Endocrinol 1983;99:87–92. [33] Han X, Rossdale PD, Ousey J, Holdstock N, Allen WR, Silver M, et al. Localisation of 15-hydroxy prostaglandin dehydrogenase (PGDH) and steroidogenic enzymes in the equine placenta. Equine Vet J 1995;27:334–9. [34] Pashen RL, Sheldrick EL, Allen WR, Flint APF. Dehydro-epiandrosterone synthesis by the fetal foal and its importance as an oestrogen precursor. J Reprod Fertil 1982;(Suppl. 32):389–97. [35] Raeside JI, Liptrap RM, Milne FJ. Relationship of fetal gonads to urinary estrogen excretion by the pregnant mare. Am J Vet Res 1973;34:843–5. [36] Raeside JI. Dehydroepiandrosterone in the fetal gonads of the horse. J Reprod Fertil 1976;46:423–5. [37] Marshall DE, Dumasia MC, Wooding P, Gower DB, Houghton E. Studies into aromatase activity associated with fetal allantochorionic and maternal endometrial tissues of equine placenta. Identification of metabolites by gas chromatography mass spectrometry. J Steroid Biochem Mol Biol 1996;59:281–96. [38] Groome NP, Illingworth PJ, O’Brien M, Priddle J, Weaver K, Mc-Neilly AS. Quantification of inhibin pro␣C-containing forms in human serum by a new ultrasensitive two-site enzyme-linked immunosorbent assay. J Clin Endocrinol Metab 1995;80:2926–32. [39] Knight PG, Muttukrishna S, Groome NP. Development and application of a two-site enzyme immunoassay for the determination of total activin-A concentrations in serum and follicular fluid. J Endocrinol 1996;148:267–79. [40] Nagamine N, Nambo Y, Nagata S, Nagaoka K, Tsunoda N, Taniyama H, et al. Inhibin secretion in the mare: localization of inhibin alpha, betaA, and betaB subunits in the ovary. Biol Reprod 1998;59:1392–8. [41] Arora DS, Cooke IE, Ganesan TS, Ramsdale J, Manek S, Charnock FML, et al. Immunohistochemical expression of inhibin/activin subunits in epithelial and granulosa cell tumors of the ovary. J Pathol 1997;181:413–8. [42] Arora DS, Ramsdale J, Cooke IE, Manek S, Ganesan TS, Groom N, et al. Inhibin in epithelial and granulosacell tumors of the ovary—a comparison of immunohistochemistry with serum levels. J Pathol 1996;179:A42. [43] Roby KF, Larsen D, Deb S, Soares MJ. Generation and characterization of antipeptide antibodies to rat cytochrome P-450 side-chain cleavage enzyme. Mol Cell Endocrinol 1991;79:13–20. [44] Johnson DC. Cellular localization and factors controlling rat placental cytochrome P45017 alpha (CYP17): 17 alpha-hydroxylase/C17,20-lyase activity. Biol Reprod 1992;46:30–8. [45] Doody KM, Carr BR, Rainey WE, Byrd W, Murry BA, Strickler RC, et al. 3 -Hydroxysteroid dehydrogenase/isomerase in the fetal zone and neocortex of the human fetal adrenal gland. Endocrinology 1990;126:2487–92. [46] Yamanouchi K, Yoshida S, Hasegawa T, Ikeda A, Chang KT, Matsuyama S, et al. Molecular cloning of DNA for inhibin ␣-subunit from equine ovary. J Vet Med Sci 1995;57:905–9. [47] Yoshida S, Yamanouchi K, Hasegawa T, Ikeda A, Suzuki M, Chang KT, et al. Molecular cloning of cDNA for equine ovarian inhibin/activin A subunit. J Vet Med Sci 1995;57:469–73. [48] Kanai-Azuma M, Kanai Y, Kurohmaru M, Sakai S, Hayashi Y. Insulin-like growth factor (IGF)-I stimulates proliferation and migration of mouse ectoplacental cone cells, while IGF-II transforms them into trophoblastic giant cells in vitro. Biol Reprod 1993;48:252–61.
34
K.Y. Arai et al. / Domestic Animal Endocrinology 31 (2006) 19–34
[49] Li RH, Zhuang LZ. The effects of growth factors on human normal placental cytotrophoblast cell proliferation. Hum Reprod 1997;12:830–4. [50] Lacey H, Haigh T, Westwood M, Aplin JD. Mesenchymally-derived insulin-like growth factor 1 provides a paracrine stimulus for trophoblast migration. BMC Dev Biol 2002;2:5. [51] Hills FA, Elder MG, Chard T, Sullivan MHF. Regulation of human villous trophoblast by insulin-like growth factors and insulin-like growth factor-binding protein-1. J Endocrinol 2004;183:487–96. [52] Clapp 3rd JF, Schmidt S, Paranjape A, Lopez B. Maternal insulin-like growth factor-I levels (IGF-I) reflect placental mass and neonatal fat mass. Am J Obstet Gynecol 2004;190:730–6.
Expression of inhibins, activins, insulin-like growth factor-I and steroidogenic enzymes in the equine placenta Koji Y. Arai a , Yumiko Tanaka b , Hiroyuki Taniyama d , Noboru Tsunoda e , Yasuo Nambo f , Natsuko Nagamine b , Gen Watanabe b,c , Kazuyoshi Taya b,c,∗ a
Department of Tissue Physiology, Tokyo University of Agriculture and Technology, Tokyo 183-8509, Japan b Laboratory of Veterinary Physiology, Faculty of Agriculture, Tokyo University of Agriculture and Technology, Fuchu, Tokyo 183-8509, Japan c Department of Basic Veterinary Science, The United Graduate School of Veterinary Science, Gifu University, Gifu 501-1193, Japan d Department of Veterinary Pathology, Rakuno Gakuen University, Hokkaido 069-8501, Japan e Shadai Corporation, Hokkaido 059-1432, Japan f Hidaka Training Research Center, Japan Racing Association, Hokkaido 057-171, Japan Received 29 June 2005; received in revised form 25 August 2005; accepted 1 September 2005
Abstract In this study, the expression patterns of inhibins, activins, insulin-like growth factor-I (IGF-I) and steroidogenic enzymes in equine placentae recovered during the latter two-thirds of gestation were examined. Concentrations of inhibin A and inhibin pro-␣C in endometrial and fetal placental tissue homogenates were very low during the period examined, whereas these tissues contained high concentrations of activin A. In both maternal endometrial and fetal placental tissues, activin A levels decreased as pregnancy progressed. Expression of inhibin ␣-subunit was not observed in the placenta using either immunohistochemistry or in situ hybridization. Inhibin/activin A-subunit and its mRNA were confined to maternal endometrial glands, whereas immunopositive B-subunit was not detected in either endometrial glands or microcotyledons. Cytochrome P450 side chain cleavage enzyme was detected by immunohistochemistry in both endometrial glands and microcotyledons, whereas cytochrome P450 17␣-hydroxylase/lyase was absent in these tissues. Immunopositive signals for 3∗
Corresponding author. Tel.: +81 423 67 5767; fax: +81 423 67 5767. E-mail address: [email protected] (K. Taya).
0739-7240/$ – see front matter © 2005 Published by Elsevier Inc. doi:10.1016/j.domaniend.2005.09.005
20
K.Y. Arai et al. / Domestic Animal Endocrinology 31 (2006) 19–34
hydroxysteroid dehydrogenase and cytochrome P450 aromatase were localized in microcotyledons but not in endometrial glands. Immunohistochemistry revealed that IGF-I was highly expressed in microcotyledons around Day 130, and decreased as pregnancy progressed. Changes in the expression of IGF-I were correlated with the number of PCNA positive cells in the placenta. The present study demonstrated the presence and localized the site of expression of activin, IGF-I and steroidogenic enzymes in equine placental tissues during the latter two-thirds of gestation; the results suggest that activin and IGF-I may be involved in the regulation of placental development. © 2005 Published by Elsevier Inc. Keywords: Inhibin; Activin; IGF-I; Steroidogenic enzymes; Equine placenta
1. Introduction The equine placenta is diffuse, non-invasive and epitheliochorial. The trophoblastic villi are organized into complexes known as microcotyledons, which cover the surface of the allantochorion and fit into endometrial crypts [1,2]. Both fetal and maternal tissues of the equine placneta are important for fetal nutrition, respiration and waste removal [3,4]. The placenta is also an endocrine and paracrine organ, which secretes diverse hormones and growth factors to maintain pregnancy and to support fetal development. Inhibins, which belong to the transforming growth factor- (TGF-) superfamily, were initially isolated as gonadal peptides that inhibit FSH secretion from the pituitary gland [5–8]. Two inhibins have been identified. Each inhibin is composed of a common ␣-subunit and either a A-subunit to give rise inhibin A or a B-subunit to form inhibin B [5–8]. Homo- and hetero-dimers of the -subunits of inhibin are called activin A, activin AB and activin B [5–8], and are multifunctional growth factors. Inhibin/activin subunits have been detected in the placenta in humans and several animal species [9–14]. It has been proposed that placental function and development are, at least in part, regulated by activin produced locally because activin has been shown to affect functions of human placental cells [15–17]. The presence of A-subunit mRNA and the absence of ␣-subunit mRNA in the equine placenta have been shown by Yamanouchi et al. [14]. However, it is not known whether the equine placenta expresses inhibin/activin B-subunit and whether it secretes inhibin/activin dimers. In addition, contribution of the placenta to the high concentrations of inhibins in equine fetal plasma [18–20] is not known, although fetal gonads are likely the major source of inhibins in fetal circulation. Insulin-like growth factors (IGFs) are polypeptides of approximately 7 kDa that have structural homology to proinsulin and their biological actions are regulated by a family of high affinity binding proteins (IGFBPs) [21]. IGFs and IGFBPs are expressed in the placenta and are important in regulating fetal and placental growth in humans and other mammalian species [22,23]. A previous study has shown that IGF-II null mice but not IGF-I null mice show smaller placentae than those of wild type mice [24]. Although placental growth is not decreased in IGF-I null mice, the mice show slower development during prenatal and postnatal periods [24]. Expression patterns of IGFs, IGF receptors and IGFBPs are different between species, even though the species are very close [23]. Despite the accumulating evidences in other species, expression patterns of IGFs in the equine pla-
K.Y. Arai et al. / Domestic Animal Endocrinology 31 (2006) 19–34
21
centa have not been reported except for IGF-II expression during the first half of gestation [25]. The equine placenta is known to secrete various steroid hormones. Progestagens are known to be required for maintaining pregnancy. After regression of maternal corpora lutea in mid-gestation, the equine placenta is the main source of progestagens during the second half of pregnancy [26,27]. In addition, high concentrations of estrogens are present in the blood and urine during the latter two-thirds of gestation in the mare [26]. They include estrone, estradiol-17, equilin and equilenin, all of which derive from placental aromatization of C-19 precursors secreted by the fetal gonads [28–32]. The roles of these large quantities of placental estrogens are not altogether clear. However, the fetal gonadectomy studies of Pashen and Allen [31] have indicated that estrogen production in equine pregnancy may be important to stimulate the development of both endometrial and placental blood vessels. A previous study examined localization of cytochrome P450 cholesterol side chain cleavage enzyme (P450scc), 3-hydroxysteroid dehydrogenase (3-HSD) and cytochrome P450 17␣-hydroxylase/lyase (P450cl7) in the equine placenta and demonstrated that P450cl7 was not detected in this tissue, whereas the other two enzymes were present [33]. These results confirmed the importance of androgens from the fetal gonads for placental estrogen synthesis [34–36]. With respect to P450 aromatase (P450arom), histological localization of this enzyme in the placenta has not been examined, although the equine placenta contains high levels of aromatase activity [37]. In this study, we focused our attention on inhibins, activins, IGF-I and steroidogenic enzymes and examined their expression in the equine placenta to elucidate physiological features of the equine placenta as an endocrine and paracrine organ.
2. Materials and methods 2.1. Animals Placentae were collected from normal pregnant Thoroughbred and Anglo-Arab mares at Days 100–312 of gestation after euthanasia. Final mating day was designated as Day 0 of gestation. The pregnant mares were killed using an overdose of a mixture of barbiturate (Ravonal; Tanabe Pharmaceutical Co. Ltd., Osaka, Japan) and suxamethonium chloride (Succine; Yamanouchi Pharmaceutical Co. Ltd., Osaka, Japan), prior to tissue recovery. Fetal gonads were also collected as previously described [18–20]. The fetal gonad samples contained both ovaries and testes, which were not distinguished from each other. All procedures were carried out in accordance with the guidelines established by the Rakuno Gakuen University for use of laboratory animals. 2.2. Tissue samples Maternal endometrial and fetal placental (allantochorionic) tissues were separated manually from the placentae before homogenization. To measure concentrations of inhibins and activin A in maternal endometrial tissues, fetal placental tissues and fetal gonads, a portion
22
K.Y. Arai et al. / Domestic Animal Endocrinology 31 (2006) 19–34
of the tissue (1.0 g) was homogenized in 2 ml of 0.01 M phosphate-buffered saline (PBS, pH 7.4) on ice using a homogenizer (Phiscotron, Nichion, Tokyo, Japan) for 30 s at 20,000 rpm. The homogenized tissue samples were centrifuged at 20,000 × g for 30 min at 4 ◦ C. The supernatant was removed and stored at −80 ◦ C until assayed for inhibins and activin A. The homogenized samples were divided into the following groups; Days 100–140 (n = 3); Days 140–180 (n = 11); Days 180–220 (n = 8); Days 220–260 (n = 3) of gestation. For histochemical studies, six placentae recovered at different stages (Days 130, 132, 162, 208, 227 and 312 of gestation) were fixed in freshly prepared 4% (w/v) paraformaldehyde (PFA; Sigma Chemical Co., St. Louis, MO) in 0.01 M PBS and embedded in paraffin. Five- and 8-m-thick tissue sections were prepared for immunohistochemistry and in situ hybridization, respectively. These sections were placed on glass slides coated with 3-aminopropyltriethoxysilane. 2.3. ELISA Concentrations of inhibin pro-␣C in tissue homogenates were measured as previously described [19,20] using a two-site ELISA kit (Serotec, Oxford, UK) designed for measurement of human inhibin pro-␣C [38]. This assay system does not detect mature ␣-subunit monomer or dimers that contain mature ␣-subunit, but does detect pro-␣C monomer, ␣subunit precursor (pro-␣N␣C) and dimers that contain either pro-␣C peptide or ␣-subunit precursor. The amount of inhibin pro-␣C was expressed in terms of the purified inhibin pro␣C derived from human follicular fluid. Concentrations of inhibin A in tissue homogenates were measured using a two-site ELISA kit (Serotec) as previously described [19,20]. The amount of inhibin A was expressed in terms of the purified 32-kDa inhibin A derived from bovine follicular fluid. These assays have previously been validated for equine samples [19,20]. Concentrations of activin A in tissue homogenates were measured using a two-site ELISA kit (Serotec) [39]. Amounts of activin A were expressed in terms of recombinant activin A. Serial dilutions of pooled equine maternal endometrial and fetal placental homogenates produced dose–response curves that were parallel to the standard curve produced with activin A (Fig. 1). The intra- and interassay coefficients of variation were 3.3 and 12.6% for inhibin pro-␣C, 4.2 and 8.7% for inhibin A and 4.4 and 5.1% for activin A, respectively. The sensitivities of the assays were 3.0 pg/ml for inhibin pro-␣C, 15.6 pg/ml for inhibin A and 7.3 pg/ml for activin A, respectively. 2.4. Immunohistochemistry Immunohistochemistory was done as previously described [20]. After deparaffinization with xylene, the tissue sections were subjected to antigen retrieval by autoclaving in 0.01 M sodium citrate buffer, pH 6.0, at 121 ◦ C for 15 min. Sections were then incubated in 3% H2 O2 in methanol at room temperature for 15 min, followed by incubation in Block Ace (Dainippon Pharmaceutical Co. Ltd., Osaka, Japan) at 37 ◦ C for 1 h to quench non-specific staining. Then they were incubated at 37 ◦ C for 16–18 h with primary antibodies diluted in Block Ace. Primary antibodies used in the present study were a rabbit polyclonal antibody to [Tyr30]-porcine inhibin ␣ chain (1–30)-NH2 conjugated to rabbit serum albumin [40] (diluted 1:4,000; [Tyr30]-porcine inhibin ␣ chain (1–30)-NH2 was provided by Dr. Nicholas Ling, Neuroendocrine Inc., San Diego, CA), a monoclonal antibody raised against the
K.Y. Arai et al. / Domestic Animal Endocrinology 31 (2006) 19–34
23
Fig. 1. Validation of the activin A ELISA system for equine samples. Different dilutions of equine maternal endometrial tissue homogenates (squares) and fetal placental tissue homogenates (triangles) produced dose–response curves that were parallel to the standard curve (circles) produced with recombinant activin A. Values are means ± S.D. of duplicate determinations. Error bars of some values are hidden by the marks.
synthetic A-subunit (E4) [41] (final concentration, 4.5 g/ml; Provided by Dr. Nigel P. Groome, Oxford Brookes University, Oxford, UK), a monoclonal antibody raised against the synthetic human B-subunit (C5) [42] (final concentration, 4 g/ml; Provided by Dr. Nigel P. Groome), a rabbit polyclonal antibody to rat P450scc [43] (diluted 1:3000; provided by Dr. Michael J. Soares, University of Kansas Medical Center, Kansas City, KS), a rabbit polyclonal antibody to rat P450cl7 [44] (diluted 1:1000; provided by Dr. Donald C. Johnson, University of Kansas Medical Center), a rabbit polyclonal antibody to human placental 3-HSD [45] (diluted 1:1000; provided by Dr. J. Ian Mason, University of Edinburgh, Edinburgh, Scotland), a rabbit polyclonal antibody to aromatase (diluted 1:1000; purchased from Biogenesis, England, UK), a purified rabbit polyclonal antibody to human IGF-I (final concentration, 10 g/ml; purchased form AUSTRAL Biologicals, San Ramon, CA), and a mouse monoclonal antibody to human proliferating cell nuclear antigen (PCNA) (final concentration, 4 g/ml; purchased from Biomeda corp, Foster City, CA). After incubation with the primary antibodies, sections were treated with 0.25% (v/v) biotinylated goat antirabbit or rabbit anti-mouse secondary antibody (Elite ABC kit; Vector Labs, Burlingame, CA) in Block Ace at 37 ◦ C for 1 h. These sections were subsequently incubated with 2% (v/v) avidin–biotin complex (Elite ABC kit) in casein-Tris-saline (CTS) at 37 ◦ C for 30 min. The reaction products were visualized by treating with 0.025% (w/v) 3,3 -diaminobenzidine tetrachloride (Sigma) in 100 mM Tris-buffered saline containing 0.01% H2 O2 for 1–30 min. As a negative control, the primary antibody was replaced with normal rabbit serum, normal rabbit IgG or normal mouse IgG. The sections were counter-stained with haematoxylin.
24
K.Y. Arai et al. / Domestic Animal Endocrinology 31 (2006) 19–34
2.5. Complementary RNA probes and in situ hybridization The 1286-base pair (bp) equine inhibin ␣-subunit cDNA clone [46] and the 1479-bp equine inhibin/activin A-subunit cDNA clone [47] (provided by Dr. Keitaro Yamanouchi, University of Tokyo, Tokyo, Japan) were subcloned into a pBluescript KS(−) vector (Stratagene, La Jolla, CA). Template inhibin ␣ and inhibin/activin A cDNAs represent 309-bp subclones corresponding to the 5 region from the first restriction site PstI and 325-bp subclones corresponding to the 5 region from the first restriction site HindIII, respectively. Digoxigenin (DIG)-labeled RNA probes were synthesized with DIG RNA labeling mixture (Roche Diagnostics, Tokyo, Japan) according to the manufacturer’s protocols. In situ hybridization was done as previously described [20] using the DIG-labeled RNA probes. 2.6. Statistical analysis To compare inhibin and activin levels in tissue samples, one-way analysis of variance (ANOVA) was carried out. When a significant effect was obtained with analysis of variance, the significance of the difference between means was determined by a Duncan multiple
Fig. 2. Concentrations of inhibin pro-␣C (A and B) and inhibin A (C and D) in maternal endometrial tissues (A and C), fetal placental tissues (A and C) and fetal gonads (B and D) at Days 100–140 (n = 3), Days 140–180 (n = 11), Days 180–220 (n = 8) and Days 220–260 (n = 3) of gestation. Values represent means ± S.E.M. ND, not detected.
K.Y. Arai et al. / Domestic Animal Endocrinology 31 (2006) 19–34
25
range test. Data are presented as means ± S.E.M. A probability value less than 0.05 was considered to be significant.
3. Results 3.1. Concentrations of inhibin pro-αC, inhibin A and activin A in maternal endometrial and fetal placental tissues Inhibin pro-␣C was not detected in fetal placental tissues recovered between Days 100 and 260 of gestation (Fig. 2A). Concentrations of inhibin pro-␣C in maternal endometrial homogenates were much lower than those in homogenates of fetal gonads (Fig. 2A and B) and neither changed significantly during the period of pregnancy that we examined. Inhibin A was detected in both maternal endometrial and fetal placental tissues (Fig. 2C). However, the levels were much lower than that in homogenates of fetal gonads (Fig. 2D). Inhibin A levels in fetal placental tissues, maternal endometrial tissues or fetal gonads did not show significant change during the period studied, although inhibin A levels in fetal gonads tended to decrease as pregnancy progressed. Activin concentrations in fetal gonads were very low during the period that we examined (Fig. 3). By contrast, high levels of activin A were detected in maternal endometrial and fetal placental tissues that were obtained between Days 100 and 140 of gestation (Fig. 3). Levels of activin A in fetal placental tissues significantly decreased between Days 100 and 180, and remained low thereafter. Levels of
Fig. 3. Concentrations of activin A in fetal gonads, maternal endometrial tissues and fetal placental tissues at Days 100–140 (n = 3), Days 140–180 (n = 11), Days 180–220 (n = 8) and Days 220–260 (n = 3) of gestation. Values are means ± S.E.M. Columns marked with the same letter at each stage did not differ significantly (P > 0.05).
26
K.Y. Arai et al. / Domestic Animal Endocrinology 31 (2006) 19–34
activin A in maternal endometrial tissues gradually decreased as pregnancy progressed, although the change is not statistically significant. 3.2. Localization of inhibin/activin subunit mRNAs in the equine placenta Sections at Day 208 of gestation hybridized with ␣- or A-subunit probe are shown in Fig. 4. In all six placentae examined (Days 130, 132, 162, 208, 227 and 312 of gestation),
Fig. 4. Localization of inhibin ␣- (A) and inhibin/activin A-subunit (C and E) mRNAs in the equine placenta at Day 208 of gestation. Only A-subunit mRNA was detected in maternal endometrial glands. Sense controls for inhibin ␣-subunits (B) and inhibin/activin A-subunits (D) show no signal. Bar = 100 m. EG, endometrial glands. Cot, microcotyledons.
K.Y. Arai et al. / Domestic Animal Endocrinology 31 (2006) 19–34
27
inhibin ␣-subunit mRNA was not detected in either endometrial glands or microcotyledons (Fig. 4A). On the other hand, endometrial glands were strongly positive for A-subunit mRNA (Fig. 4C and E) during the period investigated. Negative control sections hybridized with sense probes showed no signal (Fig. 4B and D).
Fig. 5. Photomicrographs of histological sections of the equine placenta at Day 208 of gestation. Tissue sections were immunostained for inhibin subunits and steroidogenic enzymes. A tissue section stained with hematoxylin and eosin is given in (A). Tissue sections incubated with antibodies to inhibin A-subunit (B), P450scc (C), 3-HSD (D), P450arom (E) are shown. Inhibin ␣-subunit, B-subunit and P450cl7 were not detected in the equine placenta (data not shown). Representative negative control sections for immunohistochemistry are shown in Fig. 6G and H, respectively. Bar = 100 m. EG, endometrial glands. Cot, microcotyledons.
28
K.Y. Arai et al. / Domestic Animal Endocrinology 31 (2006) 19–34
K.Y. Arai et al. / Domestic Animal Endocrinology 31 (2006) 19–34
29
3.3. Immunohistochemical localization of inhibin/activin subunits and steroidogenic enzymes in the equine placenta Expression patterns of inhibin/activin subunits and steroidogenic enzymes did not change significantly throughout the period of gestation examined (Days 130, 132, 162, 208, 227 and 312). Sections at Day 208 of gestation are shown in Fig. 5 as representatives. A section stained with hematoxylin and eosin is also shown in Fig. 5A. Immunoreactivity for inhibin ␣-subunit was not detected in either endometrial glands or microcotyledons at any stage of gestation investigated (data not shown). Immunopositive staining of inhibin/activin A-subunit was observed in endometrial glands but not in microcotyledons (Fig. 5B). These results are consistent with the expression patterns of mRNAs for these subunits. Inhibin/activin B-subunit was not detected in either endometrial glands or microcotyledons (data not shown). P450scc was localized in both endometrial glands and microcotlyledons (Fig. 5C). Immunopositive staining of P450scc in endometrial glands was more intense than in the microcotyledons. P450cl7 was not detected in either endometrial glands or microcotyledons (data not shown). Both 3-HSD and P450arom were localized in microcotyledons but not in endometrial glands (Fig. 5D and E). 3.4. Localization of IGF-I and PCNA in the equine placenta during pregnancy Sections at Days 130, 208 and 312 of gestation stained with specific antibodies to IGF-I and PCNA are shown in Fig. 6. IGF-I immunoreactivity was detected in microcotyledons throughout at all stages investigated (Days 130, 132, 162, 208, 227 and 312 of gestation), whereas no signal was observed in endometrial glands at any stage (Fig. 6A, C and E). Signals for IGF-I were most intense around Day 130 of gestation and decreased toward parturition. Similarly, PCNA was localized in nuclei of microcotyledons during the latter two-thirds of pregnancy (Fig. 6B, D and F). More than 95% trophoblast cells were positively stained with anti-PCNA antibody around Day 130, whereas PCNA-positive trophoblast cells decreased to 80–85% around Day 200 and to about 40% at Day 312. Fig. 6G and H are negative control sections for immunohistochemistry shown in Figs. 5 and 6.
4. Discussion In the present study, high concentrations of activin A were detected in equine endometrial and fetal placental tissues, whereas levels of inhibins in these tissues were very low. Because inhibin/activin A-subunit and its mRNA were detected in endometrial glands but not in fetal placental tissues, activin A in the fetal placental tissue homogenates had proba-
Fig. 6. Immunohistochemical staining of IGF-I (A, C and E) and PCNA (B, D and F) in the equine placenta at Days 130 (A and B), 208 (C and D) and 312 (E and F). Representative negative control sections incubated with normal rabbit IgG (G, 10 g/ml) and normal mouse IgG (H, 4 g/ml) instead of the primary antibodies are also shown. Sections incubated with normal rabbit serum showed similar results. Bar = 100 m. EG, endometrial glands. Cot, microcotyledons.
30
K.Y. Arai et al. / Domestic Animal Endocrinology 31 (2006) 19–34
bly diffused across from the endometrial glands. Although neither inhibin ␣-subunit protein nor its mRNA was expressed in the equine placenta, low levels of inhibin A was detected in both maternal endometrial and fetal placental tissues. Maternal endometrial tissues also contained low concentrations of inhibin pro-␣C. Our previous studies have shown that large amounts of inhibin pro-␣C and inhibin A were present in fetal circulation, whereas levels of inhibins in maternal circulation were very low [19,20]. Therefore, inhibins, which were detected in the placental tissues, were probably derived from fetal circulation. However, the endometrial inhibin pro-␣C might come from maternal circulation because fetal placental homogenates did not contain detectable inhibin pro-␣C. The present results strongly indicated that the large amounts of inhibins in fetal circulation [19,20] were not derived from placentae but secreted form fetal gonads because fetal gonadal homogenates were abundant in inhibins. A previous study [14] has shown that levels of A-subunit mRNA in the placenta during the mid one-third of gestation were higher than those during the last one-third of gestation. The previous results are thus consistent with our present findings of decreasing activin A in the placental tissues as pregnancy progressed. On the other hand, inhibin B-subunit was not detected in the equine placenta by immunohistochemistry. Because the anti-B-subunit antibody used in the present study did detect inhibin B-subunit in the equine ovary [40], the present result indicates that the equine placenta does not express or expresses very low levels of B-subunit. Activin has been shown to affect secretion of progesterone, hCG and GnRH from human placental cells [15,16]. Therefore, activins may be involved in the regulation of hormone secretion form the equine placenta in a paracrine or autocrine fashion. On the other hand, a previous study has demonstrated that activin stimulates outgrowth of human cytotrophoblast cells and increases matrix metalloproteinase release and fibronectin synthesis by these cells [17]. Because attachment of the equine fetal placenta occurs gradually over Days 40–150 [1], placental activin may promote placentation during the first half of pregnancy. Relatively high concentrations of activin A in the placental tissues at Days 100–140 and the following decrease support this hypothesis. Further studies should be required to elucidate roles of activin in the equine placenta, such as histochemical studies for detecting activin receptors to reveal activin-responsive cells in the placenta. Furthermore, expression patterns of activin before Day 100 of gestation should be examined because attachment of the fetal placenta starts before Day 100. In the present study, P450cl7 was not detected in the equine placenta, whereas 3-HSD was observed in microcotyledons. These results are consistent with the previous report [33]. Localization of P450scc in both endometrial glands and microcotyledons was also observed previously [33]. In the previous paper, however, microcotyledons were more strongly stained with anti-P450scc antibodies than endometrial glands, the exact opposite of our findings. We do not know the reason of the discrepancy. One possibility is that the different antibodies used may be responsible for the discordance between the present and previous studies. Marshall et al. have reported that high aromatase activity was detected in the fetal allantochorionic tissue, whereas aromatase activity in the maternal endometrial tissue was low [37]. They also described that the maternal tissue contained appreciable amounts of fetal chorion and the contaminating fetal tissue might be responsible for the detectable aromatase activity in the maternal tissue. Our immunohistochemical study clearly showed that P450arom was localized in fetal tissues but not in maternal tissues, and supports the previous observation. These expression patterns of steroidogenic enzymes support the previously proposed
K.Y. Arai et al. / Domestic Animal Endocrinology 31 (2006) 19–34
31
feto–placental interaction for steroidogenesis. That is, the equine placenta is able to secrete progestagens, but requires androgens from fetal gonads as precursors for estrogen synthesis. IGF-I was found in microcotyledons throughout the latter two-thirds of pregnancy. Signals for IGF-I were strong around Day 130, when placental attachment would still have been in progress, but decreased thereafter when placental attachment would have been completed (about Day 150 of pregnancy). The number of PCNA-positive cells also decreased between Days 130 and 312. Previous studies have demonstrated that IGF-I stimulates proliferation [48–51] and migration [48–50] of mouse and human placental cells. Furthermore, maternal IGF-I levels are positively correlated with placental mass in humans [52]. The present results and the previous studies suggest that equine placental IGF-I may stimulate proliferation of placental cells and promote placentation in a paracrine fashion. However, further studies are required to elucidate physiological roles of IGF-I in the equine placental development. We only examined samples obtained after Day 130 of gestation, although attachment of the equine fetal placenta occurs over Day 40–150 [1]. Furthermore, expression patterns of IGF receptors and IGFBPs should be examined to estimate sensitivity of the cells and bioavailability of IGF-I, respectively. In addition, localization of IGF-I mRNA should be examined because IGF-I, which was detected in microcotyledons by immunohistochemistry, might come form other sources. In summary, the present study is the first to demonstrate the localization of IGF-I and aromatase in the equine placenta, and secretion of activin A by this tissue. The present results suggest involvement of activin A and IGF-I in the regulation of placental development and confirmed importance of fetal androgens for the synthesis of placental estrogens in the mare.
Acknowledgements We are grateful to Dr. Nicholas Ling, Neuroendocrine Inc., San Diego, CA, for [Tyr30]porcine inhibin ␣ chain (1–30)-NH2 ; Dr. Nigel P. Groome, Oxford Brookes University, Oxford, UK, for antibodies to human inhibin/activin A- and B-subunits; Dr. Michael J. Soares, University of Kansas Medical Center, Kansas City, KS, for antibodies to P450scc; Dr. Donald Johnson, University of Kansas Medical Center, Kansas City, KS, for antibodies to rat P450cl7; Dr. J. Ian Mason, University of Edinburgh, Edinburgh, Scotland, for antibodies to human placental 3-HSD; to Dr. Keitaro Yamanouchi, University of Tokyo, Tokyo, Japan, for equine inhibin/activin cDNAs. Grants: This work was supported in part by the Ito Foundation (to K.Y.A.), a Grant-in-Aid for Scientific Research (The 21st Century Centerof-Excellence Program, E-1) from the Ministry of Education, Culture, Sports, Science and Technology of Japan (to K.T.) and a Grant-in-aid from the Equine Research Institute of the Japan Racing Association (to K.T.).
References [1] Samuel CA, Allen WR, Steven DH. Ultrastructural development of the equine placenta. J Reprod Fertil 1975;(Suppl. 23):575–8.
32
K.Y. Arai et al. / Domestic Animal Endocrinology 31 (2006) 19–34
[2] Samuel CA, Allen WR, Steven DH. Studies on the equine placenta. III. Ultrastructure of the uterine glands and the overlying trophoblast. J Reprod Fertil 1977;51:433–7. [3] Silver M. Some aspects of equine placental exchange and foetal physiology. Equine Vet J 1984;16:227–33. [4] Antczak DF, Oriol JG, Donaldson WL, Poleman C, Stenzler L, Volsen SG, et al. Differentiation molecules of the equine trophoblast. J Reprod Fertil 1987;(Suppl. 35):371–8. [5] de Jong FH. Inhib Physiol Rev 1988;68:555–607. [6] Ying S-Y. Inhibins, activins, and follistatins: gonadal proteins modulating the secretion of follicle-stimulating hormone. Endocr Rev 1988;9:267–93. [7] Mather JP, Woodruff TK, Krummen LA. Paracrine regulation of reproductive function by inhibin and activin. Proc Soc Exp Biol Med 1992;201:1–15. [8] Mather JP, Moore A, Li RH. Activins, inhibins, and follistatins: further thoughts on a growing family of regulators. Proc Soc Exp Biol Med 1997;215:209–22. [9] Meunier H, Rivier C, Evans RM, Vale W. Gonadal and extragonadal expression of inhibin alpha, beta A, and beta B subunits in various tissues predicts diverse functions. Proc Natl Acad Sci USA 1988;85:247–51. [10] Billiar RB, Rohan R, Henson MC, Smith P, Babischkin J. Regulation of immunoreactive inhibin patterns in baboon pregnancy: maternal, placental, and fetal considerations. J Clin Endocrinol Metab 1992;75: 1345–51. [11] Petraglia F, Woodruff TK, Botticelli G, Botticelli A, Genazzani AR, Mayo KE, et al. Gonadotropin-releasing hormone, inhibin, and activin in human placenta: evidence for a common cellular localization. J Clin Endocrinol Metab 1992;74:1184–8. [12] Rabinovici J, Goldsmith PC, Librach CL, Jaffe RB. Localization and regulation of the activin-A dimer in human placental cells. J Clin Endocrinol Metab 1992;75:571–6. [13] Roberts VJ, Barth SL. Expression of messenger ribonucleic acids encoding the inhibin/activin system during mid- and late-gestation rat embryogenesis. Endocrinology 1994;134:914–23. [14] Yamanouchi K, Hirasawa K, Hasegawa T, Ikeda A, Chang KT, Matsuyama S, et al. Equine inhibin/activin A-subunit mRNA is expressed in the endometrial gland, but not in the trophoblast, during pregnancy. Mol Reprod Dev 1997;47:363–9. [15] Petraglia F, Vaughan J, Vale W. Inhibin and activin modulate the release of gonadotropin-releasing hormone, human chorionic gonadotropin, and progesterone from cultured human placental cells. Proc Natl Acad Sci USA 1989;86:5114–7. [16] Steele GL, Currie WD, Yuen BH, Jia XC, Perlas E, Leung PC. Acute stimulation of human chorionic gonadotropin secretion by recombinant human activin-A in first trimester human trophoblast. Endocrinology 1993;133:297–303. [17] Caniggia I, Lye SJ, Cross JC. Activin is a local regulator of human cytotrophoplast cell differentiation. Endocrinology 1997;138:3976–86. [18] Nambo Y, Nagata S, Oikawa M, Yoshihara T, Tsunoda N, Kohsaka T, et al. High concentrations of immunoreactive inhibin in the plasma of mares and fetal gonads during the second half of pregnancy. Reprod Fertil Dev 1996;8:1137–45. [19] Tanaka Y, Taniyama H, Tsunoda N, Shinbo H, Nagamine N, Nambo Y, et al. The testis as a major source of circulating inhibins in the male equine fetus during the second half of gestation. J Androl 2002;23:229–36. [20] Tanaka Y, Taniyama H, Tsunoda N, Herath CB, Nakai R, Shinbo H, et al. Localization and secretion of inhibins in the equine fetal ovaries. Biol Reprod 2003;68:328–35. [21] Clemmons DR. Insulin-like growth factor binding proteins and their role in controlling IGF actions. Cytokine Growth Factor Rev 1997;8:45–62. [22] Giudice LC, Mark SP, Irwin JC. Paracrine actions of insulin-like growth factors and IGF binding protein1 in non-pregnant human endometrium and at the decidual-trophoblast interface. J Reprod Immunol 1998;39:133–48. [23] Han VK, Carter AM. Spatial and temporal patterns of expression of messenger RNA for insulin-like growth factors and their binding proteins in the placenta of man and laboratory animals. Placenta 2000;21:289–305. [24] Baker J, Liu JP, Robertson EJ, Efstratiadis A. Role of insulin-like growth factors in embryonic and postnatal growth. Cell 1993;75:73–82. [25] Lennard SN, Stewart F, Allen WR. Insulin-like growth factor II gene expression in the fetus and placenta of the horse during the first half of gestation. J Reprod Fertil 1995;103:169–79.
K.Y. Arai et al. / Domestic Animal Endocrinology 31 (2006) 19–34
33
[26] Ginther OJ. Endocrinology of the pregnant mare. In: Reproductive biology of the mare: basic and applied aspects. 2nd ed. Wisconsin: Equiservices; 1991. p. 419–56. [27] Ousey JC, Forhead AJ, Rossdale PD, Grainger L, Houghton E, Fowden AL. Ontogeny of uteroplacental progestagen production in pregnant mares during the second half of gestation. Biol Reprod 2003;69: 540–8. [28] Bhavnani BR, Short RV, Solomon S. Formation of oestrogens by the pregnant mare. I. Metabolism of 7-3H-dehydroisoandrosterone and 4-14C-androstenedione injected into the umbilical vein. Endocrinology 1969;85:1172–9. [29] Bhavnani BR, Short RV, Solomon S. Formation of estrogens by the pregnant mare. II. Metabolism of 14Cacetate and 3H-cholesterol injected into the fetal circulation. Endocrinology 1971;89:1152–7. [30] Bhavnani BR, Short RV. Formation of steroids by the pregnant mare. IV Metabolism of 14C-mevalonic acid and 3H-dehydroisoandrosterone injected into the fetal circulation. Endocrinology 1973;92:1397–404. [31] Pashen RL, Allen WR. The role of the fetal gonads and placenta in steroid production, maintenance of pregnancy and parturition in the mare. J Reprod Fertil 1979;(Suppl. 27):499–509. [32] Tait AD, Santikarn LC, Allen WR. Identification of 3-hydroxy-5,7 pregnandien-20- one and 3-hydroxy-5,7 androstadien-17-one as endogenous steroids in the foetal horse gonad. J Endocrinol 1983;99:87–92. [33] Han X, Rossdale PD, Ousey J, Holdstock N, Allen WR, Silver M, et al. Localisation of 15-hydroxy prostaglandin dehydrogenase (PGDH) and steroidogenic enzymes in the equine placenta. Equine Vet J 1995;27:334–9. [34] Pashen RL, Sheldrick EL, Allen WR, Flint APF. Dehydro-epiandrosterone synthesis by the fetal foal and its importance as an oestrogen precursor. J Reprod Fertil 1982;(Suppl. 32):389–97. [35] Raeside JI, Liptrap RM, Milne FJ. Relationship of fetal gonads to urinary estrogen excretion by the pregnant mare. Am J Vet Res 1973;34:843–5. [36] Raeside JI. Dehydroepiandrosterone in the fetal gonads of the horse. J Reprod Fertil 1976;46:423–5. [37] Marshall DE, Dumasia MC, Wooding P, Gower DB, Houghton E. Studies into aromatase activity associated with fetal allantochorionic and maternal endometrial tissues of equine placenta. Identification of metabolites by gas chromatography mass spectrometry. J Steroid Biochem Mol Biol 1996;59:281–96. [38] Groome NP, Illingworth PJ, O’Brien M, Priddle J, Weaver K, Mc-Neilly AS. Quantification of inhibin pro␣C-containing forms in human serum by a new ultrasensitive two-site enzyme-linked immunosorbent assay. J Clin Endocrinol Metab 1995;80:2926–32. [39] Knight PG, Muttukrishna S, Groome NP. Development and application of a two-site enzyme immunoassay for the determination of total activin-A concentrations in serum and follicular fluid. J Endocrinol 1996;148:267–79. [40] Nagamine N, Nambo Y, Nagata S, Nagaoka K, Tsunoda N, Taniyama H, et al. Inhibin secretion in the mare: localization of inhibin alpha, betaA, and betaB subunits in the ovary. Biol Reprod 1998;59:1392–8. [41] Arora DS, Cooke IE, Ganesan TS, Ramsdale J, Manek S, Charnock FML, et al. Immunohistochemical expression of inhibin/activin subunits in epithelial and granulosa cell tumors of the ovary. J Pathol 1997;181:413–8. [42] Arora DS, Ramsdale J, Cooke IE, Manek S, Ganesan TS, Groom N, et al. Inhibin in epithelial and granulosacell tumors of the ovary—a comparison of immunohistochemistry with serum levels. J Pathol 1996;179:A42. [43] Roby KF, Larsen D, Deb S, Soares MJ. Generation and characterization of antipeptide antibodies to rat cytochrome P-450 side-chain cleavage enzyme. Mol Cell Endocrinol 1991;79:13–20. [44] Johnson DC. Cellular localization and factors controlling rat placental cytochrome P45017 alpha (CYP17): 17 alpha-hydroxylase/C17,20-lyase activity. Biol Reprod 1992;46:30–8. [45] Doody KM, Carr BR, Rainey WE, Byrd W, Murry BA, Strickler RC, et al. 3 -Hydroxysteroid dehydrogenase/isomerase in the fetal zone and neocortex of the human fetal adrenal gland. Endocrinology 1990;126:2487–92. [46] Yamanouchi K, Yoshida S, Hasegawa T, Ikeda A, Chang KT, Matsuyama S, et al. Molecular cloning of DNA for inhibin ␣-subunit from equine ovary. J Vet Med Sci 1995;57:905–9. [47] Yoshida S, Yamanouchi K, Hasegawa T, Ikeda A, Suzuki M, Chang KT, et al. Molecular cloning of cDNA for equine ovarian inhibin/activin A subunit. J Vet Med Sci 1995;57:469–73. [48] Kanai-Azuma M, Kanai Y, Kurohmaru M, Sakai S, Hayashi Y. Insulin-like growth factor (IGF)-I stimulates proliferation and migration of mouse ectoplacental cone cells, while IGF-II transforms them into trophoblastic giant cells in vitro. Biol Reprod 1993;48:252–61.
34
K.Y. Arai et al. / Domestic Animal Endocrinology 31 (2006) 19–34
[49] Li RH, Zhuang LZ. The effects of growth factors on human normal placental cytotrophoblast cell proliferation. Hum Reprod 1997;12:830–4. [50] Lacey H, Haigh T, Westwood M, Aplin JD. Mesenchymally-derived insulin-like growth factor 1 provides a paracrine stimulus for trophoblast migration. BMC Dev Biol 2002;2:5. [51] Hills FA, Elder MG, Chard T, Sullivan MHF. Regulation of human villous trophoblast by insulin-like growth factors and insulin-like growth factor-binding protein-1. J Endocrinol 2004;183:487–96. [52] Clapp 3rd JF, Schmidt S, Paranjape A, Lopez B. Maternal insulin-like growth factor-I levels (IGF-I) reflect placental mass and neonatal fat mass. Am J Obstet Gynecol 2004;190:730–6.