Coexpression of Adrenomedullin and Its Receptors in the ...

BIOLOGY OF REPRODUCTION 79, 200–208 (2008) Published online before print 9 April 2008. DOI 10.1095/biolreprod.107.064022

Coexpression of Adrenomedullin and Its Receptors in the Reproductive System of the Rat: Effects on Steroid Secretion in Rat Ovary1 Yuk-Yin Li,3 Lei Li,3 Isabel Shui-Shan Hwang,3 Fai Tang,3,5 and Wai-Sum O2,4,6 Departments of Physiology3 and Anatomy,4 Centre of Heart, Brain, Hormone and Healthy Aging,5 Centre of Reproduction, Development and Growth,6 Faculty of Medicine, The University of Hong Kong, Hong Kong, China respectively. The ADM2 receptor binds with both ADM and CALCA [9, 10]. The functional receptors also require an intracellular protein, receptor coupling protein (CRCP), for signal transduction [10]. Immunoreactive ADM and Adm mRNA were widely distributed in various tissues both in the human [2, 11] and the rat [3, 12–15]. Gene expression and/or peptide of Adm have also been identified in reproductive organs, such as the ovary [15, 16], the uterus [15, 17], the testis [14, 18], the prostate [19, 20], and the epididymis [21]. Plasma ADM levels fluctuate with the cyclic changes in LH and 17b-estradiol during the menstrual cycle [22]. Adrenomedullin has been found in the human granulosa cells and acts as a local factor to enhance progesterone production [23]. Adrenomedullin reversibly inhibits spontaneous periodic contraction of the uterus [24], in which specific ADM binding sites were reported [17]. These findings suggest that ADM may play a role in regulating the functions of the female reproductive system. However, there were no reports of the gene expression of ADM receptor components in the ovary and oviduct. Heterozygous ADM knockout mice have been shown to have decreased fertility and a smaller litter size [25]. Yet, apart from some literature on the importance of ADM during pregnancy [26, 27], the roles of ADM are largely unknown. We therefore studied ADM level and the gene expression of Adm, Calcrl, Ramp1, Ramp2, and Ramp3 in the ovary and the oviduct of the cycling rat and characterized the ovarian-specific binding site for 125I-ADM by Scatchard plot analysis. To establish a physiological role of ADM in the ovary, the effects of ADM on estradiol and progesterone secretion in vitro were investigated.

ABSTRACT The present study demonstrates the expression of adrenomedullin (ADM) in the reproductive system of the female rat and its effect on the secretion of estradiol and progesterone. Ovarian ADM and Adm mRNA levels were decreased at estrus, whereas oviductal Adm mRNA levels were low at proestrus. Both tissues were shown to coexpress mRNAs encoding the calcitonin receptor-like receptor and receptor activity-modifying protein 1 (Ramp1), Ramp2, and Ramp3. Gel filtration chromatography of ovarian extracts showed two peaks, with the predominant one eluting at the position of authentic rat ADM (1–50) at estrus and at the position of ADM precursor at diestrus. Positive ADM immunostaining was localized in the granulosa and theca cells of the follicle and corpora lutea of the ovary. Adrenomedullin inhibited FSH-induced estradiol secretion in 2-day-old follicles and also suppressed eCG-stimulated progesterone release in corpora lutea. The inhibitory effect of ADM on the follicles and the corpora lutea was abolished by calcitonin gene-related peptide (8–37) and ADM (22–52), respectively. The presence of ADM and the gene expression of ADM and its receptor components in the female reproductive system suggest a paracrine effect of ADM on ovarian steroidogenesis. estradiol, follicle-stimulating hormone, human chorionic gonadotropin, ovary, progesterone

INTRODUCTION Adrenomedullin (ADM), originally isolated from human pheochromocytoma, is a pluripotent peptide characterized by its vasorelaxant activity [1]. Human and rat ADM consist of 52 and 50 amino acids, respectively [2, 3]. It is a member of the calcitonin family of peptides with sequence homology to calcitonin gene-related peptide (CALCA) [4]. Adrenomedullin can bind to the CALCA receptor in several types of tissues [5, 6], but specific ADM receptors that are insensitive to CALCA receptor antagonist CALCA (8–37) have been identified [7]. McLatchie et al. [8] demonstrated that the calcitonin receptorlike receptor (CALCRL) responsible for the majority of ADM and CALCA binding and ligand specificity depends on the type of receptor activity-modifying protein (RAMP) that is expressed [8]; namely, RAMP1, RAMP2, and RAMP3. The coupling of CALCRL with RAMP1, RAMP2, and RAMP3 gives rise to a CALCA, ADM1, and ADM2 receptors,

MATERIALS AND METHODS Animals Female Sprague-Dawley rats (12–13 wk) were obtained from the Laboratory Animal Unit, Faculty of Medicine, University of Hong Kong. Vaginal smears were obtained daily, and only those that showed a regular 4-day estrous cycle were included in this study. All rats were killed between 1000 and 1200 h. All procedures had been approved by the Committee on the Use of Live Animals for Teaching and Research, the University of Hong Kong, and were carried out in accordance with the Guide for the Care and Use of Laboratory Animals (National Academy of Science).

RT-PCR of Adm, Calcrl, Ramp, and Crcp 1

Substantially supported by a grant from the Research Grants Council of the Hong Kong Special Administrative Region, China (HKU 7736/07M). 2 Correspondence: W.S.O., Department of Anatomy, Faculty of Medicine, The University of Hong Kong, 21 Sassoon Rd., Pokfulam, Hong Kong, China. FAX: 852 28170857; e-mail: [email protected]

Total RNA of ovary or oviduct was prepared using TRIZOL reagent [18, 21] and subjected to real-time RT-PCR. RNA samples (5 lg) were reverse transcribed into cDNA with the SuperScript II reverse transcriptase according to the manufacturer’s instructions (Life Technologies, Carlsbad, CA). Polymerase chain reactions were conducted by iCycler iQ real-time PCR detection system using iQ SYBR Green Supermix (Bio-Rad Laboratories, Hercules, CA). Standard curves for each primer pair were prepared using serial dilutions of cDNA to determine the PCR efficiency. The PCR efficiencies for Adm, Calcrl, Ramp1, Ramp2, Ramp3, and Actb were all above 0.95 and similar. The relative gene expression levels were analyzed using the DDCT method, where CT is cycle threshold. The reaction mixtures contained 25 ll iQ SYBR

Received: 5 July 2007. First decision: 2 August 2007. Accepted: 27 March 2008. Ó 2008 by the Society for the Study of Reproduction, Inc. ISSN: 0006-3363. http://www.biolreprod.org

200

ACTION OF ADRENOMEDULLIN ON STEROID RELEASE

FIG. 1. Expression of Adm and Calcrl (A) and Ramp1, Ramp2, and Ramp3 (B) in the ovary across the estrous cycle. The mRNA levels were normalized to Actb mRNAs, and each value represents the means 6 SEM (n ¼ 6–13 for proestrus, n ¼ 6–12 for estrus, n ¼ 3–8 for metestrus, n ¼ 3–7 for diestrus). *P , 0.05 compared with the respective diestrus stage as calculated by t-test. **P , 0.01 compared with the respective diestrus stage as calculated by Mann-Whitney rank sum test.

201

FIG. 2. Expression of Adm and Calcrl (A) and Ramp1, Ramp2, and Ramp3 (B) in the oviduct across the estrous cycle. The mRNA levels were normalized to Actb mRNAs, and each value represents the mean 6 SEM (n ¼ 8–11 for proestrus, n ¼ 10–12 for estrus, n ¼ 7–8 for metestrus, n ¼ 5–6 for diestrus). *P , 0.05 compared with the respective proestrus stage as calculated by t-test.

acetonitrile in 1% TFA, dried in a speed-vacuum concentrator (Savant, Farmingdale, NY), redissolved in 0.5 ml radioimmunoassay (RIA) buffer (0.1% sodium phosphate [pH 7.4], 0.1% heat-inactivated BSA, 0.05 M sodium chloride, 0.01% sodium azide, and 0.1% Triton X-100), and further purified with a nanospin column (molecular mass cutoff point is 10 kDa; Millipore) [13].

Green Supermix (Bio-Rad Laboratories), 500 nM of each primer, 1 ll template cDNA, and DNase-free water to a final volume of 50 ll. Cycle conditions were 958C for 10 min, followed by 40 cycles of 958C for 45 sec, 598C for 30 sec, and 728C for 45 sec. The reaction was completed with a dissociation step for melting point analysis with 508C to 958C (in increments of 0.58C) for 10 sec each. The primers were designed on the basis of the published sequence of Adm (caggacaagcagagcacgtc, forward; tctggcggtagcgtttgac, reverse); Calcrl (ccaaacagacttgggagtcactagg, forward; gctgtcttctctttctcatgcgtgc, reverse); Ramp1 (cactcactgcaccaaactcgtg, forward; cagtcatgagcagtgtgaccgtaa, reverse); Ramp2 (aggtattacagcaacctgcggt, forward; acatcctctgggggatcggaga, reverse); Ramp3 (acctgtcggagttcatcgtg, forward; acttcatccggggggtcttc, reverse) and Actb (ggaaatcgtgcgtgacatta, forward; aggaaggaaggctggaagag, reverse). Melt curve analysis for each primer set revealed only one peak for each product. The size of the PCR products was confirmed by comparing the size of product with a commercial ladder after agarose gel electrophoresis.

Whole ovary or oviduct was homogenized in 2 N acetic acid for ADM immunoreactivity, followed by boiling for 10 min. A 50-ll aliquot was taken for protein assay, and the remaining homogenates were centrifuged at 18 600 3 g for 20 min at 48C. The supernatants were lyophilized and stored at 208C until assay.

Extraction of ADM from Plasma

Radioimmunoassay

A volume of 2 ml plasma was acidified with an equal volume of 1% trifluoroacetic acid (TFA) and centrifuged at 18 600 3 g for 20 min at 48C (Sorvall SM 24; Thermo Fisher Scientific Inc., Waltham, MA) [12]. The diluted plasma samples were loaded onto pre-equilibrated C-18 Sep-columns (Waters Corp., Milford, MA), which were then washed twice with 3 ml of 1% TFA. The peptide was eluted with 60% high-performance liquid chromatography-grade

The lyophilized tissue sample was reconstituted in RIA buffer for the determination of immunoreactive ADM concentration [12]. Rat ADM and ADM antiserum were purchased from Peninsula (Belmont, CA). The 125Ilabeled ADM was from Phoenix (Belmont, CA). The sensitivity for the assay was 5 pg/tube. The intraassay and interassay coefficients of variation were 7% and 10%, respectively.

Extraction of ADM from the Ovary and the Oviduct

202

LI ET AL.

FIG. 3. Correlations between the serum concentrations of estradiol and the mRNA levels of Adm (A), Calcrl (B), Ramp2 (C), and Ramp3 (D).

Protein Measurement

Immunocytochemistry

Fifty-microliter aliquots of tissue homogenate or standard (BSA) were boiled with 1 N NaOH for 10 min, and 50 ll of the boiled sample was mixed with 2.5 ml protein assay reagent (Bio-Rad Laboratories). After 10 min of incubation at room temperature, the samples were measured spectrophotometrically at 595 nm (LKB Ultraspec II; Biochrom). The immunoreactive ADM was expressed as femtomoles per milligram of protein.

To localize ADM in the ovary, avidin-biotin histochemical staining procedure using a Vextastain ABC kit (Vector Laboratories, Burlingame, CA) was employed [18]. Ovary was collected and fixed in formalin overnight. The tissue sample was embedded in paraffin, and 10-lm sections were cut on a microtome. The sections were treated for 60 min with 10% methanol and 3% hydrogen peroxide, and then were incubated with the primary antibody (1:2000) for ADM at 48C overnight. The avidin-biotin-peroxidase complex in the sections was visualized with diaminobenzidine for 5–10 min.

Gel Filtration Chromatography The lyophilized sample of the ovary was reconstituted in distilled water, and the supernatant obtained after centrifugation at 18 600 3 g for 20 min at 48C was acidified with 96% glacial acetic acid to give a final concentration of 1 N acetic acid. The sample was chromatographed on a Biogel P30 (Bio-Rad Laboratories) column (0.9 3 60 cm), which was eluted with 1 N acetic acid at a flow rate of 1 ml per 10 min for 400 min [18, 21]. These 1-ml fractions were assayed for ADM immunoreactivities. The column was calibrated with authentic ADM, with blue dextran for void-volume determination, and with the proteins from the gel filtration kit, carbonic anhydrase, cytochrome c, and vitamin B12 as molecular weight markers.

Receptor-Binding Assays Membranes were prepared by differential centrifugation [28], and receptor binding was performed as described previously [18, 21]. Whole ovary was homogenized in ice-cold 50 mM Hepes (pH 7.6) comprising 0.25 M sucrose, 10 lg/ml pepstatin, 0.25 lg/ml each of leupeptin and antipain, 0.1 mg/ml each of benzamidine and bacitracin, and 30 lg/ml aprotinin. The homogenates were centrifuged at 1500 3 g for 20 min at 48C, and the supernatants were centrifuged at 100 000 3 g for 1 h at 48C. The pellets were resuspended in 10 volumes of the above-mentioned buffer without sucrose and were centrifuged at 100 000 3 g for 1 h at 48C. The final pellets were resuspended, aliquoted, and

TABLE 1. ADM concentrations in the reproductive tissues, plasma ADM, and serum estradiol and progesterone concentrations in cycling rats.a Parameter Tissue ADM concentrations Ovary (fmol/mg) Oviduct (fmol/mg) Uterus (fmol/mg) Plasma or serum hormone levels ADM (fmol/ml) Estradiol (pg/ml) Progesterone (ng/ml) a

Proestrus (n)

Estrus (n)

Metestrus (n)

Diestrus (n)

10.2 6 2.0 (13) 31.4 6 5.2 (11) 8.0 6 1.0 (13)

6.1 6 1.0 (12) 27.2 6 4.9 (12) 7.9 6 1.2 (12)

13.4 6 1.7 (8)b 30.7 6 7.0 (8) 8.9 6 2.9 (8)

15.7 6 2.3 (7)c,d 30.0 6 3.0 (6) 6.4 6 1.1 (7)

2.4 6 0.1 (8) 13.5 6 3.5 (13) 9.3 6 3.1 (13)

2.3 6 0.1 (9) 6.9 6 1.5 (12) 8.9 6 2.3 (12)

2.4 6 0.1 (7) 12.8 6 3.5 (8) 9.2 6 2.3 (8)

2.2 6 0.1 (7) 8.9 6 1.7 (7) 16.2 6 5.2 (6)

Data represent mean 6 SEM; n ¼ number of rats. Value is significantly different compared to the estrus stage; P , 0.05. c Value is significantly different compared to the estrus stage; P , 0.01. d Value is significantly different compared to the proestrus stage; P , 0.05. b


FIG. 4. Correlations between the serum concentrations of estradiol and the mRNA levels of Adm (A), Calcrl (B), and Ramp1 (C).

stored at 708C until assay. Ovarian membranes (equivalent to 30 lg protein) were incubated at 48C for 15 min with 125I-labeled ADM (10–300 pM) in 0.3 ml binding buffer (20 mM Hepes [pH 7.4], 5 mM MgCl2, 10 mM NaCl, 4 mM KCl, 1 mM EDTA, 1 lM phosphoramindon, 0.25 mg/ml of bacitracin, and 0.3% BSA). After incubation, membranes were filtered through a Brandel cell harvester (Biomedical Research and Development Laboratories, Gaithersburg, MD) using GF/B filter paper (Whatman, Clifton, NJ) that had been soaked overnight in 0.3% polyethylenimine and washed three times with 3 ml ice-cold 50 mM Tris-HCl (pH 7.4). Radioactivity retained by the filters was counted by a gamma counter (Cobra II; Packard, Shelton, CT) at 75% efficiency. Specific binding was defined as total binding minus nonspecific binding, which was determined in the presence of 500 nM unlabeled rat ADM. The ratio of

203

FIG. 5. Correlation between the serum concentrations of estradiol and the oviductal ADM concentrations (A). Correlations between the uterine ADM concentrations and the serum concentrations of estradiol (B) and progesterone (C).

specifically bound to free radioligand concentration was plotted against specific binding (Scatchard plot), in which the Kd and Bmax values were obtained.

Isolation of Follicles and Incubation Procedure Immature rats of 26 days of age were injected intraperitoneally with 30 IU eCG (Sigma, St. Louis, MO). The ovaries were excised 24 or 48 h later under an overdose of pentobarbital and transferred to ice-chilled DMEM-F12 medium (1:1; Life Technologies) containing 100 IU/ml penicillin G and 50 lg/

204

LI ET AL. under an atmosphere of 5% CO2/95% O2 for 1 h. There were a total of five treatment groups: 1) control (medium only); 2) FSH (0.2 IU/ml; Gonal-F; Laboratories Serono, Aubonne, Switzerland); 3) ADM (108 or 107 M); 4) FSH (0.2 IU/ml) plus ADM (108 or 107 M); and 5) FSH (0.2 IU/ml) plus ADM (108 or 107 M) with 107 M ADM (22–52) (Phoenix). After preincubation, the media of all wells were replaced, and ADM (22–52) was added to the wells for treatment [5]. Incubation was continued for another 30 min before the drugs for the various treatment groups were added. After 3 h of incubation, the media were centrifuged at 600 3 g for 10 min. The supernatant was collected and stored at 208C until it was analyzed for estradiol by enzyme immunoassay (EIA). Total proteins were determined by protein assay as described above. As it was subsequently found that the ADM receptor blocker, ADM (22–52), failed to block the ADM effect in the 2-day-old follicles, this study was repeated using CALCA (8–37) (Phoenix), the CALCA receptor antagonist, at 106 M.

Isolation of Corpora Lutea and Incubation Procedure To obtain a well-defined generation of corpora lutea, immature rats aged 26 days were injected intraperitoneally with 30 IU eCG. Two days later, the rats were injected intraperitoneally with 20 IU eCG (Sigma) to induce ovulation. Pentobarbital was administrated intraperitoneally (0.1 ml/100 g) 24 h later, and the ovaries were then excised and transferred to ice-chilled DMEM-F12 medium containing 100 IU/ml penicillin G and 100 lg/ml streptomycin sulfate. Corpora lutea were isolated under a stereomicroscope. The corpora lutea (five per well in a 24-well tissue culture plate) were incubated in 0.5 ml of the above medium at 378C under an atmosphere of 5% CO2/95% O2 for 2 h. There were a total of five treatment groups: 1) control (medium only); 2) eCG (0.5 IU/ml); 3) ADM (108 or 107 M); 4) eCG (0.5 IU/ml) plus ADM (108 or 107 M); and 5) eCG (0.5 IU/ml) plus ADM (108 or 107 M) with 106 M ADM (22–52). After preincubation, the media of all wells were replaced and ADM (22–52) were added to the wells for treatment [5]. Incubation was continued for another 30 min before the drugs for the various treatment groups were added. At the end of 3 more hours, the media were centrifuged at 600 3 g for 10 min. The supernatants were collected and stored at 208C until analysis for progesterone by EIA. Total proteins were determined by protein assay as described above.

Estradiol and Progesterone Determination

FIG. 6. Biogel P30 filtration chromatograms of the extracts of the whole ovary at estrus (A) and at diestrus (B). Void volume (V0) is at fraction 10, and authentic ADM elutes at fraction 22. Carbonic anhydrase, cytochrome c, and vitamin B12 eluted at fractions 11, 15, and 28, respectively. ml streptomycin sulfate (Gibco-BRL). One-day-old follicles from 24-h posteCG treatment or 2-day-old follicles from 48-h post-eCG treatment were isolated under a stereomicroscope. The follicles (six to eight per well in a 24well tissue culture plate) were incubated in 0.5 ml of the above medium at 378C

FIG. 7. Immunocytochemical study of ADM in the rat ovary. A–C) Positive immunostaining in interstitial cells (A), granulosa cells (B), theca cells (B), and granulosa lutein cells in the corpus luteum (C). D–F) Negative controls without the addition of ADM anti-serum.

Serum estradiol and progesterone levels were measured with an estradiol RIA kit (DiaSorin, Saluggia, Italy) and a progesterone RIA kit (Diagnostic Systems Laboratories Inc., Webster, TX). The sensitivities for the assays were 2 pg/ml and 0.1 ng/ml, respectively. The intraassay variations were 3.8% and 5.1%, respectively. All samples for a given experiment were measured in the same assay. Estradiol and progesterone concentrations in the culture media were measured with an estradiol immunoassay kit (MP Biomedicals, Orangeburg, NY) and a progesterone immunoassay kit (Pantex, Bio-Analysis Inc., Santa Monica, CA), respectively. The sensitivities for the assays were 10 pg/ml and 0.04 ng/ml, respectively. The intraassay variations were 10.9% and 5.7%, respectively, and the interassay variations were 12.5% and 7.5%, respectively.


205

FIG. 8. Scatchard plot of 125I-labeled ADM binding in rat ovary membranes. The Kd is 61.97 pM, and the Bmax is 14.08 fmol/mg protein. B/F, bound:free ratio.

Statistical Analysis All data are expressed as mean 6 SEM, and statistical significance was assessed by one-way ANOVA followed by Fisher’s least significance difference test for posthoc comparisons. Correlations between the ADM levels and serum estradiol or serum progesterone were calculated using two-tailed Pearson correlation coefficients. Similar correlations were done for the gene expression of Adm, Calcrl, and Ramp with the steroid levels. Statistical significance was taken at the P , 0.05 level.

RESULTS Expression of Adm and Its Receptor Components in Cycling Rats The mRNA expression of Adm and its receptor components were estimated by real-time RT-PCR, and the results are shown in Figures 1 and 2. Ovarian Adm and Ramp1 mRNA levels were significantly lower at estrus than at diestrus (P , 0.01 for Adm as calculated by Mann-Whitney rank sum test; P , 0.05 for Ramp1 as calculated by t-test; Fig. 1). Calcrl, Ramp2, and Ramp3 mRNA levels remained unchanged throughout the cycle. However, the mRNA levels of Adm, Calcrl, Ramp2, and Ramp3 were positively correlated with serum estradiol levels (r ¼ 0.59, P , 0.001 for Adm; r ¼ 0.45, P , 0.01 for Calcrl; r ¼ 0.5, P , 0.01 for Ramp2; r ¼ 0.71, P , 0.001 for Ramp3; Fig. 3). In the oviduct, Adm and Calcrl mRNA levels were higher at estrus than at proestrus (P , 0.05 as calculated by t-test; Fig. 2), whereas Ramp mRNA levels were similar at all stages. The mRNA levels of Adm were positively correlated with serum estradiol levels (r ¼ 0.45, P , 0.05; Fig. 4), but the mRNA levels of Calcrl and Ramp1 were negatively correlated with serum estradiol levels (r ¼ 0.34, P , 0.05 for Calcrl; r ¼ 0.34, P , 0.05 for Ramp1; Fig. 4). ADM Immunoreactivity, Plasma ADM, and Serum Estradiol and Progesterone Concentrations The levels of ADM in the ovary and oviduct, the plasma levels of ADM, and the serum levels of estradiol and progesterone among the four stages of the rat estrous cycle are summarized in Table 1. There were no significant changes in the levels of ADM in the oviduct among the four stages. However, ovarian ADM concentrations were significantly reduced at estrus (P , 0.01 for estrus vs. metestrus; P ,

FIG. 9. Effects of ADM on in vitro basal and FSH-stimulated estradiol secretion in rat 2-day follicles from 48-h post-eCG treatment. *P , 0.01, **P , 0.001 compared with the group without ADM and FSH. P , 0.05, P , 0.01, P , 0.001 compared with the FSH-alone groups. #P , 0.05 compared with the groups with both ADM and FSH. Data are given as mean 6 SEM (A: n ¼ 7–8 for the ADM-alone group and both ADM and FSH groups, n ¼ 4–5 for the remaining groups; B: n ¼ 4–5 for all groups) and are expressed as percentage changes from the control (i.e., no ADM, no FSH group), which is taken as 100%.

0.001 for estrus vs. diestrus) and peaked at diestrus (P , 0.05 for proestrus vs. diestrus). The plasma concentrations of ADM as well as the serum concentrations of estradiol and progesterone were not significantly different among the four stages of the estrous cycle. Adrenomedullin levels in the oviduct were positively correlated with serum estradiol levels (r ¼ 0.62, P , 0.001; Fig. 5). Characterization of ADM in the Ovary: Gel Filtration Chromatographic Analysis, Immunohistochemical Study, and Scatchard Plot Analysis Two immunoreactive ADM peaks were seen on the gel filtration chromatograms of the ovary at estrus (Fig. 6A) and diestrus (Fig. 6B). The ovary at estrus showed a predominant peak of authentic rat ADM (1–50) and a smaller peak of rat ADM precursor. The opposite was found to be true for the ovary at diestrus. Moderate positive immunostaining of ADM

206

LI ET AL.

FIG. 11. Schematic figure of the steroidogenic effect of ADM on rat ovary. Adrenomedullin inhibits the production of sex steroids. This inhibitory effect is limited by gonadotropins by suppressing Adm gene expression. Taken together with the stimulatory effect of estradiol on Adm gene expression and production, this forms a regulatory steroidogenic loop among ADM, gonadotropins, and sex steroids in the ovary. FIG. 10. Effects of ADM on in vitro basal and eCG-stimulated progesterone secretion in rat corpora lutea. *P , 0.001 compared with the group without ADM and eCG. P , 0.05, P , 0.01 compared with the groups with eCG alone. #P , 0.05, ##P , 0.01 compared with the groups with both ADM and eCG. Data are given as mean 6 SEM (n ¼ 8–9 for the control and eCG alone groups, n ¼ 3–4 for the remaining groups) and are expressed as percentage changes from the control (i.e., no ADM, no eCG group), which is taken as 100%.

was localized in the interstitial cells (Fig. 7A), granulosa cells, and theca cells (Fig. 7B), as well as an intense staining in the granulosa lutein cells of corpus luteum (Fig. 7C). The Scatchard plot of ADM showed that the Kd and Bmax for ADM binding were 67.2 6 2.73 pM and 15.0 6 1.22 fmol/mg protein, respectively. A representative Scatchard plot is shown in Figure 8. Effect of ADM on Estradiol Release In Vitro On 1-day follicles, ADM had no effects on basal or FSHstimulated estradiol release (data not shown). In contrast, ADM at 108 or 107 M inhibited the FSH-stimulated release of estradiol in 2-day follicles (P , 0.05) without affecting the basal estradiol secretion (Fig. 9). This inhibitory effect of ADM was not blocked by the 107 M ADM receptor antagonist, human ADM (22–52) (Fig. 9A), but was blocked by 106 M CALCA (8–37), the human CALCA receptor antagonist (Fig. 9B). Effect of ADM on Progesterone Secretion In Vitro Adrenomedullin at 108 or 107 M suppressed the eCGstimulated secretion of progesterone (P , 0.01 for 108 M ADM; P , 0.05 for 107 M ADM), but had no effect on the basal progesterone secretion (Fig. 10). This inhibitory effect of ADM was completely blocked by 106 M ADM receptor antagonist, human ADM (22–52). DISCUSSION Gel filtration study showed that the major peak for ADM in the ovarian extract was the active peptide at estrus and the precursor at diestrus. The peak heights of the active peptide at these two stages were similar, but the precursor peak was three times higher at diestrus than at estrus. The latter finding may account for the increase in ADM immunoactivity in the ovary at diestrus. The relative abundance of precursor at diestrus may

be due to an increase in secretion of the active peptide or a decrease in processing of the precursor protein. Specific binding sites for ADM were observed for the first time in the rat ovary (Fig. 8). The expression of Calcrl and Ramp as demonstrated by RT-PCR lends further support. Both binding affinity and the maximum binding (number of receptors) are lower than those previously reported for the uterus [17]. CALCRL can function as a CALCA receptor or an ADM receptor. Coexpression of CALCRL with RAMP1 gives rise to a CALCA receptor, whereas with RAMP2 or RAMP3 it results in an ADM receptor [8, 29, 30]. Our results revealed that Ramp1, Ramp2, and Ramp3 are expressed in the ovary and oviduct, suggesting that both ADM and CALCA receptors are present in these organs. We found that ADM was mainly localized in the corpora lutea, with relatively weak ADM staining in the interstitial cells, granulosa cells, and theca cells. This agrees with the previous finding of a high level of mRNA in the corpus luteum and a low level in the follicle [15]. ADM is expressed in human and rat granulosa cells [23, 31] and in human corpora lutea, and RAMP2 was found in the human corpora lutea [16]. The gene expression of Adm, Calcrl, Ramp2, and Ramp3 is more closely correlated with serum estradiol levels than with the stage of the estrous cycle, suggesting that the expression of these genes in ovary is regulated by the cyclic changes of estradiol. The changes in ADM mirror those of Adm mRNA levels, as both are lowest at estrus and may be related to the possible actions of ADM on follicular development and luteogenesis. At estrus, Adm gene expression is low in order to facilitate follicular development. In rat follicular culture, FSH decreases Adm mRNA in the granulosa cells and stimulates follicular growth [31]. This suggests that ADM may be inhibitory to follicular development. In the human ovary, the ADM mRNA level is low in the follicles and high in the corpus luteum, especially at the midluteal phase [16]. At diestrus, ADM gene expression is high, because ADM is important for the development and maintenance of the corpus luteum [16]. In the present study, we demonstrated that ADM diminished FSH-stimulated estradiol release in rat follicles as well as eCGstimulated progesterone release in rat corpora lutea. At estrus, the decrease in ADM would enhance the FSH effect on steroidogenesis, whereas at diestrus, the increase in ADM would attenuate the effect of eCG on progesterone secretion. The finding in corpus luteum is different from the previous report of an enhancement of progesterone secretion by ADM in human granulosa cells [23]. This discrepancy may be due to


species difference or the preparation used (granulosa cell vs. corpus luteum). The results of using an ADM receptor antagonist suggest that specific ADM receptors may mediate the effects of ADM on the corpus luteum but not on the follicles. Indeed, the success of the CALCA receptor antagonist in the follicles suggests that specific CALCA receptors may be involved. The ADM2 receptor is probably not involved, as these actions are not partially blocked by the ADM receptor antagonist. The decrease in Ramp1 gene expression accompanying the decrease in ADM and Adm gene expression at estrus would decrease CALCA receptors, and thus further reduce the action of ADM. There is no increase in Ramp2 gene expression (and ADM receptor) at diestrus to further increase the effect of an increase in ADM and Adm gene expression. The interrelationships between the gonadotropins, ADM, and the sex steroids are illustrated in Figure 11. Adrenomedullin and its mRNA have not previously been characterized in the oviduct. We show that immunoreactive ADM is present at relatively high levels, approximately 3- to 4fold of the concentrations of ovary and uterus, suggesting that ADM may be a major protein present in the oviductal fluid. Oviductal ADM levels and the gene expression of Adm, Calcrl, and Ramp1 are more closely correlated with serum estradiol levels than with the stage of the estrous cycle, suggesting that the expression of these genes in oviduct is regulated by the cyclic changes of estradiol. The finding that the Adm and Calcrl mRNA levels are significantly higher at estrus than at proestrus suggests that ADM may play a role in some reproductive processes, such as sperm and ovum transport. Marinoni et al. [32] suggest that ADM in the seminal fluid may play a role in sperm motility. More importantly, the presence of sperm has been shown to increase Adm gene expression in the mouse oviduct [33]. The oviduct is lined by ciliated cells, and ADM may initially mediate ovum transport by its effects on ciliary action and later facilitate fertilization of the ovum by inhibiting the contractility of the oviduct. In support of the latter, CALCA is known to inhibit contractility in the human fallopian tube [34] and intense CALCA-binding sites present on the vascular smooth muscle of the fallopian tube [35]. Therefore, ADM may be one of the local factors in regulating the oviductal blood flow as well as the muscular tone of oviduct, playing a role in both oocyte and sperm transport [32, 36] and in fertilization. As Calcrl mRNA level is elevated, there may be increases in the responses to ADM via both ADM and CALCA receptors. In summary, ADM and the gene expression of its receptor components are differentially regulated in the ovary and the oviduct throughout the estrous cycle. The cyclic changes in ovarian ADM release may be regulated by estradiol via the latter’s effects on gonadotropin secretion. Functionally, ADM inhibits the effects of gonadotropins on both estradiol secretion by the follicles and progesterone secretion by corpora lutea. REFERENCES 1. Kitamura K, Kangawa K, Kawamoto M, Ichiki Y, Nakamura S, Matsuo H, Eto T. Adrenomedullin: a novel hypotensive peptide isolated from human pheochromocytoma. Biochem Biophys Res Commun 1993; 192:553–560 2. Kitamura K, Sakata J, Kangawa K, Kojima M, Matsuo H, Eto T. Cloning and characterization of cDNA encoding a precursor for human adrenomedullin. Biochem Biophys Res Commun 1993; 194:720–725 3. Sakata J, Shimobuko T, Kitamura K, Nakamura S, Kangawa K, Matsuo H, Eto T. Molecular cloning and biological activities of rat adrenomedullin, a hypotensive peptide. Biochem Biophys Res Commun 1993; 195:921–927. 4. Eguchi S, Hirata Y, Iwasaki H, Sato K, Watanabe TX, Inui Y, Nakajima K, Sakakibara S, Marumo F. Structure-activity relationship of adrenomedullin, a novel vasodilatory peptide, in cultured rat vascular smooth muscle cells. Endocrinology 1994; 135:2454–2458.

207

5. Entzeroth M, Doods HN, Wieland HA, Wienen W. Adrenomedullin mediates vasodilation via CGRP1 receptors. Life Sci 1995; 56:PL19– PL25. 6. Eguchi S, Hirata Y, Kano H, Sato K, Watanabe Y, Watanabe TX, Nakajima K, Sakakibara S, Marumo F. Specific receptors for adrenomedullin in cultured rat vascular smooth muscle cells. FEBS Lett 1994; 340: 226–230. 7. Kapas S, Catt K, Clark A. Cloning and expression of cDNA encoding a rat adrenomedullin receptor. J Biol Chem 1995; 270:25344–25347. 8. McLatchie LM, Fraser NJ, Main MJ, Wise A, Brown J, Thompson N, Solari R, Lee MG, Foord SM. RAMPs regulate the transport, and ligand specificity of the calcitonin-receptor-like receptor. Nature 1998; 393:333– 339. 9. Prado MA, Evans-Bain B, Oliver KR, Dickerson IM. The role of the CGRP-receptor component protein (RCP) in adrenomedullin receptor signal transduction. Peptides 2001; 22:1773–1781. 10. Hay DL, Conner AC, Howitt SG, Smith DM, Poyner DR. The pharmacology of adrenomedullin receptors and their relationship to CGRP receptors. J Mol Neurosci 2004; 22:105–113. 11. Ichiki Y, Kitamura K, Kangawa K, Kawamoto M, Matsuo H, Eto T. Distribution and characterization of immunoreactive adrenomedullin in human tissue and plasma. FEBS Lett 1994; 338:6–10. 12. Hwang ISS, Tang F. The distribution and gene expression of adrenomedullin in the rat brain: peptide/mRNA and precursor/active peptide relationships. Neurosci Lett 1999; 267:85–88. 13. Hwang ISS, Tang F. Peripheral distribution and gene expression of adrenomedullin in the rat: possible source of blood adrenomedullin. Neuropeptides 2000; 34:32–37. 14. Sakata J, Shimokubo T, Kitamura K, Nishizono M, Iehiki Y, Kangawa K, Matsuo H, Eto T. Distribution and characterization of immunoreactive rat adrenomedullin in tissue and plasma. FEBS Lett 1994; 352:105–108. 15. Cameron VA, Fleming AM. Novel sites of adrenomedullin gene expression in mouse and rat tissues. Endocrinology 1998; 139:2253–2264. 16. Abe K, Minegishi T, Ibuki Y, Kojima M, Kangawa K. Expression of adrenomedullin in the human corpus luteum. Fertil Steril 2000; 74:141– 145. 17. Upton PD, Austin C, Taylor GM, Nandha KA, Clark AJL, Ghatei MA, Bloom SR, Smith DM. Expression of adrenomedullin and its binding sites in the rat uterus: increased number of binding sites and AM messenger ribonucleic acids in 20-day pregnant rats compared with non-pregnant rats. Endocrinology 1997; 138:2508–2513. 18. Li YY, Hwang ISS, O WS, Tang F. Adrenomedullin peptide: gene expression of adrenomedullin, its receptors and receptor activity modifying proteins, and receptor binding in rat testis – actions on testosterone secretion. Biol Reprod 2006; 75:183–188 19. Pewitt EB, Haleem R, Wang Z. Adrenomedullin gene is abundantly expressed and directly regulated by androgen in the rat ventral prostate. Endocrinology 1999; 140:2382–2386. 20. Jimenez N, Calvo A, Martinez A, Rosell D, Cuttitta F, Montuenga LM. Expression of adrenomedullin and proadrenomedullin N-terminal 20 peptide in human and rat prostate. J Histochem Cytochem 1999; 47:1167– 1177. 21. Hwang ISS, Autelitano DJ, Wong PYD, Leung GPH, Tang F. Coexpression of adrenomedullin and adrenomedullin receptors in rat epididymis: distinct physiological actions on anion transport. Biol Reprod 2003; 68:2005–2012. 22. Marinoni E, Di Iorio R, Letizia C, Lucchini C, Alo P, Cosmi EV. Changes in plasma adrenomedullin levels during the menstrual cycle. Regul Pept 2000; 87:15–18. 23. Moriyama T, Otani T, Maruo T. Expression of adrenomedullin by human granulosa lutein cells and its effect on progesterone production. Eur J Endocrinol 2000; 142:671–676. 24. Yanagita T, Yamamoto R, Sugano T, Kobayashi H, Uezono Y, Yokoo H, Shiraishi S, Minami S, Wada A. Adrenomedullin inhibits spontaneous and bradykinin-induced but not oxytocin- or prostaglandin F2a-induced periodic contraction of rat uterus. Br J Pharmacol 2000; 130:1727–1730. 25. Li M, Yee D, Magnuson TR, Smithies O, Caron KM. Reduced maternal expression of adrenomedullin disrupts fertility, placentation, and fetal growth in mice. J Clin Invest 2006; 116:2653–2662. 26. Wilson c, Nikitenko LL, Sargent IL, Rees MC. Adrenomedullin: multiple functions in human pregnancy. Angiogenesis 2004; 7:203–212. 27. Penchalaneni J, Wimalawansa SJ, Yallampalli C. Adrenomedullin antagonist treatment during early gestation in rats causes fetoplacental growth restriction through apoptosis. Biol Reprod 2004; 71:1475–1483. 28. Owji AA, smith Coppock HA, Morgan DG, Bhogal R, Ghatei MA, Bloom SR. An abundant and specific binding site for the novel vasodilator adrenomedullin in the rat. Endocrinology 1995; 136:2127–2134.

208

LI ET AL.

29. Buhlmann N, Leuthauser K, Muff R, Fischer JA, Born W. A receptor activity modifying protein (RAMP)2-dependent adrenomedullin receptor is a calcitonin gene-related peptide receptor when coexpressed with human RMAP1. Endocrinology 1999; 140:2883–2890. 30. Chakravarty P, Suthar TP, Coppock HA, Nicholl CG, Bloom SR, Legon S, Smith DM. CGRP and adrenomedullin binding correlates with transcripts levels for calcitonin receptor-like receptor (CRLR) and receptor activity modifying proteins (RAMPs) in rat tissues. Br J Pharmacol 2000; 130:189–195. 31. Abe K, Minegishi T, Tano M, Hirakawa T, Tsuchiya M, Kangawa K, Kojima M, Ibuki Y. Expression and effect of adrenomedullin on rat granulosa cell. Endocrinology 1998; 139:5263–5266. 32. Marinoni E, Di Iorio R, Villaccio B, Vellucci O, Di Netta T, Sessa M,

33. 34. 35. 36.

Letizia C, Cosmi EV. Adrenomedullin in human male reproductive system. Eur J Obstet Gynecol Reprod Biol 2005; 122:195–198. Fazeli A, Affara NA, Hubank M, Holt WV. Sperm-induced modification of the oviductal gene expression profile after natural insemination in mice. Biol Reprod 2004; 71:60–65. Samuelson UE, Dalsgaard CJ, Lundberg JM, Hokfelt T. Calcitonin generelated peptide inhibits spontaneous contractions in human uterus and fallopian tube. Neurosci Lett 1985; 62:225–230. Nimmo AJ, Whitaker EM, Carstairs JR, Morrison JF. The autoadiographic localization of calcitonin gene-related peptide and substance P receptors in human fallopian tube. Q J Exp Physiol 1989; 74:955–958. Jansen RP. Endocrine response in the fallopian tube. Endocr Rev 1984; 5: 525–551.