Prostaglandin E2 as a Regulator of Germ Cells during Ovarian ...

ORIGINAL E n d o c r i n e

ARTICLE R e s e a r c h

Prostaglandin E2 as a Regulator of Germ Cells during Ovarian Development Rosemary A. L. Bayne, Sharon L. Eddie, Craig S. Collins, Andrew J. Childs, Henry N. Jabbour, and Richard A. Anderson Medical Research Council Human Reproductive Sciences Unit (R.A.L.B., C.S.C., A.J.C., H.N.J.), and Division of Reproductive and Developmental Sciences (S.L.E., R.A.A.), University of Edinburgh Centre for Reproductive Biology, The Queen’s Medical Research Institute, Edinburgh EH16 4TJ, United Kingdom

Context: The formation of primordial follicles occurs during fetal life yet is critical to the determination of adult female fertility. Prior to this stage, germ cells proliferate, enter meiosis, and associate with somatic cells. Growth and survival factors implicated in these processes include activin A (INHBA), the neurotrophins BDNF and NT4 (NTF5), and MCL1. The prostaglandins have pleiotrophic roles in reproduction, notably in ovulation and implantation, but there are no data regarding roles for prostaglandins in human fetal ovarian development. Objective: The aim of the study was to investigate a possible role for prostaglandin (PG) E2 in human fetal ovary development. Design: In vitro analysis of ovarian development between 8 and 20 wk gestation was performed. Main Outcome Measure(s): The expression patterns of PG synthesis enzymes and the PGE2 receptors EP2 and EP4 in the ovary were assessed, and downstream effects of PGE2 on gene expression were analyzed. Results: Ovarian germ cells express the PG synthetic enzymes COX2 and PTGES as well as the EP2 and EP4 receptors, whereas COX1 is expressed by ovarian somatic cells. Treatment in vitro with PGE2 increased the expression of BDNF mRNA 1.7 ⫾ 0.16-fold (P ⫽ 0.004); INHBA mRNA, 2.1 ⫾ 0.51-fold (P ⫽ 0.04); and MCL1 mRNA, 1.15 ⫾ 0.06-fold (P ⫽ 0.04), but not that of OCT4, DAZL, VASA, NTF5, or SMAD3. Conclusions: These data indicate novel roles for PGE2 in the regulation of germ cell development in the human ovary and show that these effects may be mediated by the regulation of factors including BDNF, activin A, and MCL1. (J Clin Endocrinol Metab 94: 4053– 4060, 2009)

O

varian development in the first trimester of pregnancy is characterized by germ cell proliferation. Some germ cells continue to undergo mitosis throughout the second trimester (1), but there is increasing entry into meiosis and changing associations with somatic cells such that ultimately some germ cells form primordial follicles whereas the majority are believed to undergo apoptosis (2– 4). During this period, a number of germ-cell markers (OCT4, DAZL, VASA) are expressed associated with dif-

ferent stages of germ cell development (5). Additionally, growth factors including activin A (encoded by INHBA) and the neurotrophins brain-derived neurotrophic factor (BDNF) and NT4 (encoded by NTF5) have been demonstrated previously to be important during fetal ovary development (4, 6 –14), but there is very limited information regarding their control and interactions. Prostaglandins (PGs) mediate many important biological processes and are recognized as key molecules in re-

ISSN Print 0021-972X ISSN Online 1945-7197 Printed in U.S.A. Copyright © 2009 by The Endocrine Society doi: 10.1210/jc.2009-0755 Received April 7, 2009. Accepted July 8, 2009. First Published Online July 14, 2009

Abbreviations: BDNF, Brain-derived neurotrophic factor; COX, cyclooxygenase; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; PG, prostaglandin; PTGES, PGE synthase.

J Clin Endocrinol Metab, October 2009, 94(10):4053– 4060

jcem.endojournals.org

4053

4054

Bayne et al.

PGE2 in Ovarian Development

productive function (15), particularly in ovulation and implantation. Involvement in sex determination has been proposed (16), but roles in early stages of oogenesis or folliculogenesis have not been identified. PGs are produced from arachidonic acid by cyclooxygenase (COX) enzymes and specific terminal prostanoid synthase enzymes. There are two predominant COX enzyme isoforms (COX1 and COX2) (17). COX1 is constitutively expressed in many cell types, whereas COX2 is induced by factors including cytokines and tumor promoters (17). COX enzymes convert arachidonic acid to an intermediate PG, PGH2, and this is converted to PGE2 by PGE synthase (PTGES). Once synthesized, PGs act in an autocrine or paracrine manner through binding to a family of prostanoid-specific G protein-coupled receptors (18). For PGE2, there are four different receptors (EP1–EP4) that have distinct but overlapping tissue distributions and activate different intracellular signaling pathways and gene expression (19). EP2 and EP4 receptors elevate intracellular cAMP accumulation via G␣s (20). In light of the important roles for PGs in later follicle function, we have investigated whether PG synthetic enzymes and receptors are expressed in the human fetal ovary during the developmental period when oogonia proliferate, enter meiosis, and start to form primordial follicles. We then proceeded to examine whether PGE2 has any effect on the expression of genes already identified to be important at this stage. We demonstrate that PGE2 may have important roles in oocyte development and survival during this crucial period when a woman’s endowment of eggs for the whole of her reproductive life is determined.

Materials and Methods Tissue First- and second-trimester ovaries were obtained after medical termination of pregnancy for social reasons as described previously (7). Maternal consent was obtained, and the study was approved by the Lothian Research Ethics Committee. Gestation was determined by ultrasound scan and subsequent direct measurement of foot length. The sex of first-trimester ovaries was confirmed by PCR for the male-specific gene SRY. Ovaries were removed and used directly for culture, snap-frozen, and stored at ⫺80 C or fixed in Bouins solution before processing into paraffin using standard methods. A total of 41 fetal specimens were used in this study.

Immunohistochemistry Paraffin-embedded second-trimester ovaries were cut into 5-␮m sections and mounted onto electrostatically charged microscope slides (VWR, Leicestershire, UK). Immunohistochemistry was performed as previously described (21), except where stated. For those antibodies requiring antigen retrieval (EP2,


PTGES, COX1, and COX2), the slides were pressure cooked for 5 min in 0.01 M citrate buffer and left to stand for 20 min before cooling in water. Sections were blocked with avidin and then biotin (both from Vector, Peterborough, UK). Primary antibodies were rabbit polyclonal antibodies to EP2 (101750), EP4 (101775), and PTGES (160140) (all from Cayman Chemical Co., Ann Arbor, MI; diluted 1:100, 1:500, and 1:50, respectively), or goat polyclonal antibodies to COX1 (sc-1752) and COX2 (sc-1745) (both from Santa Cruz Biotechnology, Santa Cruz, CA; each diluted 1:50). Secondary antibodies were swine antirabbit biotinylated (E0353), diluted 1:500 (EP2 and EP4); goat antirabbit biotinylated (E0437), diluted 1:500 (PTGES) (both from Dako, Cambridge, UK); or chicken antigoat biotinylated (sc-2984) (Santa Cruz Biotechnology) diluted 1:200 (COX1 and COX2). Primary antibodies were preincubated with the supplied blocking peptide as a negative control. Sections were developed with streptavidin-horseradish peroxidase and 3,3⬘-diaminobenzidine tetrahydrochloride (Dako) and counterstained with hematoxylin. Images were captured using an Olympus Provis microscope (Olympus Optical Co., London, UK) equipped with a Kodak DCS330 camera (Eastman Kodak Co, Rochester, NY). For dual staining immunofluorescence, antigen retrieval, and peroxidase, avidin/biotin and serum blocks were performed as above. Slides were then sequentially incubated with EP2 antibody diluted 1:4000 in normal goat serum (Diagnostics Scotland, Carluke, UK; 1:4 in PBS containing 5% BSA), goat antirabbit peroxidase IgG H&L Fab fragments (ab7171; Abcam plc, Cambridge, UK), and then the TSA Plus Cy3 System (PerkinElmer Life Sciences, Beaconsfield, Buckinghamshire, UK) diluted 1:50 in substrate with two washes in PBS ⫹ 0.05% Tween 20 and PBS between each incubation. After a second antigen retrieval by microwaving in 0.01 M citrate buffer for 4 min, slides were blocked again in normal goat serum/PBS/BSA before adding EP4 antibody diluted 1:1000; goat antirabbit peroxidase IgG H&L Fab fragments diluted 1:500, and then the TSA Plus Fluorescein System (Perkin-Elmer Life Sciences) diluted 1:50 in substrate, again with washes in between. Sections were counterstained with 4⬘,6-diamidino-2-phenylindole (DAPI) (1:1000 in PBS; Sigma, Poole, UK), washed again, and mounted in Permafluor (Beckman Coulter, High Wycombe, UK). Fluorescent images were captured using a LSM510 confocal microscope (Carl Zeiss Ltd., Welwyn Garden City, Hertfordshire, UK).

Culture of fetal ovaries Human fetal ovaries of 14 –17 wk gestation (n ⫽ 6, 3 ⫻ 15 wk, 2 ⫻ 16 wk, and 1 ⫻ 17 wk) were cultured as small explants in cell culture inserts essentially as described previously (10). For each experiment, both ovaries from a single fetus were dissected cleanly in HBSS (Sigma, Poole, UK), cut into approximately 1-mm cubes, and then distributed equally between the three treatment groups (thus approximately six per treatment). Tissue was cultured in ␣MEM ⫹ GlutaMAX with 1X nonessential amino acids (both from Invitrogen, Paisley, UK); 2 mM sodium pyruvate and 3 mg/ml BSA Fraction V (both from Sigma, Poole, UK); and penicillin/streptomycin/amphotericin B (Cambrex Biosciences, Baltimore, MD) for 8 h in the presence of no treatment, 3 ␮g/ml indomethacin (Sigma, Poole, UK) ⫹ vehicle (ethanol), or 3 ␮g/ml indomethacin ⫹ 100 nM PGE2 (Sigma). Indomethacin was included to reduce endogenous PG production, thus potentially enhancing the effect of exogenous PGE2. Each set of explants was then homogenized in RLT Buffer (QIAGEN,


Crawley, UK) containing 0.14 storage at ⫺80 C.

M

2-mercaptoethanol before

RNA extraction and first strand cDNA synthesis RNA was extracted from RLT lysates of cultured ovary using the RNeasy Micro Kit (QIAGEN) or from frozen uncultured ovaries (dissected clean of mesonephros in the case of first-trimester specimens) using the RNeasy Mini Kit (QIAGEN), both with on-column DNase I digestion according to the manufacturer’s protocol. RNA concentration and purity were measured on the NanoDrop 1100 (NanoDrop Products, Wilmington, DE) and first strand cDNA synthesized from 200 ng RNA using Superscript III Reverse Transcriptase Master Mix ⫾ the RT enzyme mix (RT⫹ and RT⫺, respectively) according to the manufacturer’s recommendations (Invitrogen).

RT-PCR analysis RT-PCR for COX1, COX2, PTGES, EP2, and EP4 as well as GAPDH (housekeeping gene control) was performed on RT⫹ and RT⫺ samples of first- and second-trimester human fetal ovary or human endometrial cDNA (as a positive control) using HotStar Taq DNA Polymerase (QIAGEN) and the primers shown in Table 1. PCRs were incubated at 95 C for 15 min, followed by 35 cycles of 95 C for 30 sec; 60 C for 30 sec; and 72 C for 30 sec, with a final extension step of 72 C for 10 min. PCRs were analyzed by agarose gel electrophoresis 关2.5% agarose gels, stained with GelRed nucleic acid stain (Cambridge Bioscience Ltd., Cambridge, UK)兴.


4055

Quantitative RT-PCR analysis of gene expression Quantitative RT-PCR was performed using PowerSYBR Green Master Mix and the ABI7900HTFast system with SDS2.0 software (gestation analysis) or ABI7500 Fast system with SDS1.1 software (cultures; all Applied Biosystems, Warrington, UK) under standard run conditions using the default two-step PCR protocol with dissociation curve analysis. Standard curves for products of each gene transcript (Table 1) were performed using cDNA dilutions (1:5 to 1:10 000) generated from 19-wk human fetal ovary RNA. Then 1:10 dilutions of uncultured or 1:5 dilutions of cultured ovary cDNA were used for quantitative comparisons relative to glyceraldehyde-3-phosphate dehydrogenase (GAPDH), which has previously proved to be a suitable normalization control in the human fetal ovary. Melt curves were analyzed to confirm specific products, and standard curves yielded slopes approaching ⫺3.324 with R2 values close to 0.99, allowing quantification using the 2⫺⌬⌬Ct method. For gestational analysis of expression, samples were grouped into three developmental stages: first-trimester tissue (8 –9 wk; germ cell proliferation); early second trimester (14 –16 wk; some initiation of meiosis); later second trimester (17–21 wk; widespread meiosis, follicle formation).

Statistical analysis RT-PCR data were analyzed by ANOVA with post hoc Bonferroni and linear trend tests (gestation analysis) or paired t tests on log-transformed data comparing PGE2 ⫹ indomethacin treatment to indomethacin-only treatment (as the appropriate

TABLE 1. PCR primer details Gene COX1 COX2 PTGES EP2 EP4 GAPDH OCT4 DAZL VASA INHBA BDNF NTF5 MCL1 SMAD3

Primer pair F: 5⬘-TGTTCGGTGTCCAGTTCCAATA-3⬘; R: 5⬘-ACCTTGAAGGAGTCAGGCATGAG-3⬘ F: 5⬘-CCTTCCTCCTGTGCCTGATG-3⬘; R: 5⬘-ACAATCTCATTTGAATCAGGAAGCT-3⬘ F: 5⬘-GAAGAAGGCCTTTGCCAAC-3⬘; R: 5⬘-GGGTTAGGACCCAGAAAGGA-3⬘ F: 5⬘-GACCGCTTACCTGCAGCTGTAC-3⬘; R: 5⬘-TGAAGTTGCAGGCGAGCA-3⬘ F: 5⬘-ACGCCGCCTACTCCTACATG-3⬘; R: 5⬘-AGAGGACGGTGGCGAGAAT-3⬘ F: 5⬘-GACATCAAGAAGGTGGTGAAGC-3⬘; R: 5⬘-GTCCACCACCCTGTTGCTGTAG-3⬘ F: 5⬘-ACATCAAAGCTCTGCAGAAAGAAC-3⬘; R: 5⬘-CTGAATACCTTCCCAAATAGAACCC-3⬘ F: 5⬘-GAAGGCAAAATCATGCCAAACAC-3⬘; R: 5⬘-CTTCTGCACATCCACGTCATTA-3⬘ F: 5⬘-AAGAGAGGCGGCTATCGAGATGGA-3⬘; R: 5⬘-CGTTCACTTCCACTGCCACTTCTG-3⬘ F: 5⬘-GGCAAGTTGCTGGATTATAGTG-3⬘; R: 5⬘-CCACATACCCGTTCTCCCCGAC-3⬘ F: 5⬘-AACAATAAGGACGCAGACTT-3⬘; R: 5⬘-TGCAGTCTTTTTGTCTGCCG-3⬘ Quantitect Primer Assay Hs_NTF5_1_SG (QIAGEN) F: 5⬘-ATCTCTCGGTACCTTCGGGAGC-3⬘; R: 5⬘-GCTGAAAACATGGATCATCACTCG-3⬘ F: 5⬘-TGAGGCTGTCTACCAGTTGACC-3⬘; R: 5⬘-CTAAGACACACTGGAACAGCGG-3⬘

F, Forward; R, reverse.

Amplicon (bp) 95

GenBank accession no. NM_000962

82

NM_000963

168

NM_004878

71

NM_000956

64

NM_000958

212

NM_002046

126

NM_002701

186

NM_001351

238

NM_024415

272

NM_002192

222

NM_001709

96 221

NM_006179 NM_021960

199

NM_005902

4056

Bayne et al.



across fetal ovary development with no significant changes. COX2 mRNA levels are very low throughout (below the limit of quantitative detection, data not shown). In contrast, PTGES mRNA levels fell 3-fold (P ⫽ 0.002) between first- and early second-trimester ovary before increasing again 2-fold by later gestations (P ⫽ 0.008). Differences between the earliest and latest gestations analyzed were not significant. Regarding the receptors, EP2 mRNA did not change significantly across the range of gestations, but EP4 levels rose gradually across development with a statistically significant linear trend (P ⫽ 0.03). Immunohistochemical analysis We investigated the cellular localization of these same gene products in the fetal ovary by immunohistochemistry. COX1 was strongly expressed in the surface epithelium and in somatic cells intermingling between germ cells (Fig. 2, A–C). No staining of germ cells was observed. Staining was strongest in the cortex near the surface epithelium, with a reducing gradient of expression in soFIG. 1. A, Representative nonquantitative RT-PCR analysis of COX1, COX2, PTGES, EP2, and EP4 mRNA from 64-d (1T), 14-wk (14), and 18-wk (18) human fetal ovary or endometrium (E) matic cells among the more mature cDNA as indicated. Plus and minus signs indicate the presence or absence of reverse germ cells located more centrally (Fig. transcriptase during cDNA synthesis. M, 100bp ladder marker; bl, water blank. GAPDH was 2, A and B). Only somatic cells in and analyzed as a positive control. B, Quantitative RT-PCR analysis of mRNA expression of COX1, PTGES, EP2, and EP4 in human ovary at different gestations. *, P ⬍ 0.05; **, P ⬍ 0.01. around germ cell nests or surrounding Mean ⫾ SEM of n ⫽ 4 –10 per group. Quantities are expressed as percentage relative to follicles were stained: cells within the GAPDH. somatic cell streams showed no expression of COX1. control) and untreated to indomethacin only, all using GraphPad The pattern of COX2 expression was markedly differPrism 5.0 software (GraphPad Software, Inc. San Diego, CA). ent (Fig. 2D). Staining was confined to the germ cell compartment (Fig. 2, E and F). The intensity of staining was variable both between oocyte nests and within nests, with Results some oocytes having higher levels of COX2 than others (Fig. COX enzymes, PTGES, and PG receptors EP2 and 2E). COX2 expression was maintained within the oocytes of EP4 are expressed in the human fetal ovary primordial follicles (Fig. 2F) in later-gestation specimens. To assess whether the fetal ovary can specifically synRT-PCR analysis Expression of the genes required for PGE2 signaling thesize PGE2, we next investigated expression of microwas investigated by RT-PCR in the human fetal ovary. somal PTGES in ovaries where COX activity is clearly COX1 (PTGS1), COX2 (PTGS2), PTGES, EP2, and EP4 present. As with COX2, PTGES was predominantly lowere initially assayed on 64-d (first trimester), 14- and calized to germ cells and was absent from somatic cell 18-wk fetal ovary cDNAs. All genes were found to be streams (Fig. 2, G–I), with expression appearing to inexpressed in all RT⫹ cDNA samples (Fig. 1A), although crease at later gestations. The presence of the PGE2 receptors, EP2 and EP4, was COX2 was barely detectable in first-trimester ovary. Quantitative RT-PCR analysis of these genes (Fig. 1B) also investigated. Clear germ cell expression of EP2 was demonstrated that COX1 expression remains fairly stable detected in second-trimester ovary (Fig. 3, A–C) with



4057

FIG. 2. Immunohistochemical localization of COX1 (A–C), COX2 (D–F), and PTGES (G–I) in second-trimester human fetal ovaries. Gestations: A, B, E, and G, 14 wk; H and I, 17 wk; C, D, and F, 19 wk. Insets show peptide-blocked negative controls. Positive 3,3⬘-diaminobenzidine tetrahydrochloride staining in all panels is brown, and sections are counterstained with hematoxylin. S, Somatic cell (S1, somatic cell within “cord”; S2, somatic cell among germ cells); gc, germ cell; gc*, germ cell staining more intensely for COX2; PF, primordial follicle; se, surface epithelium. Scale bars are all 50 ␮m.

staining stronger toward the inner cortex and absent from those germ cells at the outermost cortex (Fig. 3C). Staining of the cell surface of the oocytes was more obvious at 19 wk gestation, both within nests and in primordial follicles (Fig. 3B). EP4 was also expressed by germ cells (Fig. 3, D–F) but, in contrast to EP2, expression was strongest in germ cells at the periphery of the ovary with a reducing gradient of expression toward the inner cortex (Fig. 3D). The differential but overlapping expression of EP2 and EP4 was particularly evident after dual staining immunofluorescence (Fig. 3, G–I), confirming exclusive expression of EP4 by germ cells most peripherally located and expression of both by some germ cells located away from the peripheral zone. There was no evidence of either EP2 or EP4 in the somatic cells at any gestation. BDNF, INHBA, and MCL1 gene expression are up-regulated by PGE2 in cultured ovary To determine whether exogenous PGE2 alters germ cell gene expression, organ explants were cultured in the presence or absence of PGE2 for 8 h (n ⫽ 6), and the tissue was immediately lysed for RNA extraction. No significant differences in gene expression of the germ cell markers OCT4, DAZL, or VASA were noted between PGE2-

treated and appropriate control cultures (Fig. 4, top panel). Despite natural variation in expression between specimens that was not gestation related (reflected in the error bars in Fig. 4), expression of INHBA (activin A) and BDNF showed robust and reproducible increases with PGE2 treatment. INHBA showed a 2.1 ⫾ 0.51-fold increase (P ⫽ 0.04) with PGE2 (Fig. 4, middle panel), and BDNF expression rose 1.7 ⫾ 0.16-fold with PGE2 treatment (P ⫽ 0.004). NTF5 expression tended to fall but did not show consistent changes between experiments (Fig. 4, middle panel). Expression of the antiapoptotic factor MCL1 (expressed exclusively in germ cells (4)) also showed a significant increase in expression (Fig. 4, lower panel; P ⫽ 0.04) but only of 1.15 ⫾ 0.06-fold. The activin signaling factor SMAD3 mRNA (expressed in somatic cells) was unchanged by PGE2 (Fig. 4, lower panel). There were no significant changes in expression of any gene between indomethacin-treated and untreated samples.

Discussion PGs are well-recognized as essential mediators of several processes in female reproduction, most notably ovulation

4058

Bayne et al.



FIG. 3. Immunohistochemical localization of EP2 (A–C, H, and I) and EP4 (D–F, G, and I) in second-trimester human fetal ovaries. Gestations: A, D, and E, 14 wk; G–I, 17 wk; B and F, 19 wk; and C, 20 wk. Insets show peptide-blocked negative controls. A–F, Positive 3,3⬘-diaminobenzidine tetrahydrochloride staining is brown, and sections are counterstained with hematoxylin. S1, Somatic cell within “cord”; S2, somatic cell among germ cells; gc, germ cell; PF, primordial follicle. Scale bars are 50 ␮m. G–I, Dual immunofluorescence for EP2 (red, H and I) and EP4 (green, G and I). Counterstain is DAPI (blue, I). Separate channel (G and H) and merged (I) images are shown. Yellow in I indicates colocalization, although the intensity of the red staining tends to swamp out the green staining (compare I and G). Scale bars are 20 ␮m.

and implantation. The data presented here provide evidence that PGs may also have important roles in the development of the ovary, providing a basis for autocrine germ cell PGE2 signaling as well as paracrine signaling of PGs from somatic granulosa precursor cells to adjacent germ cells. Furthermore, addition of PGE2 to ovarian tissue in vitro was shown to selectively increase expression of germ cell-expressed genes identified as important regulators of this stage of ovarian development. The data demonstrate the expression of genes encoding the key enzymes required for the synthesis of PGE2, i.e. COX1, COX2, and PTGES, and the mediators of its effects, i.e. its receptors EP2 and EP4. Levels of COX enzyme mRNAs did not change across gestation, indicating that fairly constant levels of precursors are available, but levels of PTGES mRNA were reduced at 14 –16 wk and then rose again from 17 wk. This rise at later gestations also appears to be reflected at the protein level, although a

quantitative analysis of the immunohistochemistry was not performed. Levels of EP4 receptor mRNA rose gradually with development, and EP4 was found to be exclusively expressed by germ cells. Whether this reflects an increasing role for this receptor as the ovary develops or simply a relative increase in the number of germ cells that express it is not yet clear, but EP4 continues to be expressed in the more mature, more centrally located germ cells that express EP2. EP2 mRNA levels did not change across the gestational range, yet the EP2 receptor protein is only expressed in more mature germ cells. These changes may reflect multiple roles for PGE2 as it interacts with different EP receptors as oocytes mature. The striking differences in distribution of COX1 protein 关somatic cells, mostly at the periphery of the ovary where germ cells are less mature and express the pluripotency marker OCT4 (5)兴 and COX2 (germ cells, including maturing oocytes within nests and those already enclosed



4059

a proportion of larger oocytes within cell nests that are about to undergo nest breakdown and primordial follicle formation (10). It is largely believed to act on adjacent somatic cells, based on the localization of its transcription factors phosphorylated SMADs 2 and 3, and may regulate oocyte development through the regulation of kit ligand/cKIT signaling (7, 23). PGE2 increased gene expression of INHBA, thus potentially promoting this transition. PGE2 has been shown to induce INHBA, as well as INHA, in cultured human granulosa-lutein cells (24), thus this appears to be a conserved pathway in ovarian function. BDNF mRNA levels were similarly observed to rise after PGE2 treatment. FIG. 4. Quantitative RT-PCR analysis of mRNA expression of genes indicated after culture of PGE2 and PGD2 have been shown to be human ovary explants in base medium (Untreated) or in the presence of 3 ␮g/ml indomethacin powerful inducers of the neurotrophins (Indo) or 3 ␮g/ml Indomethacin ⫹ 100 nM PGE2 (Indo ⫹ PGE2). *, P ⬍ 0.05 vs. indomethacin NGF and BDNF in mouse astrocyte culalone; **, P ⬍ 0.01 vs. indomethacin alone. Data are mean ⫾ SEM of n ⫽ 6 per group. Expression levels are percentage relative to GAPDH. tures (25). As well as their roles in neuronal cells, BDNF and NT4 (NTF5), in primordial follicles) indicates a range of functions, which bind to and activate the Tropomyosin-related Kiwhich may be mediated by several PGs. In addition to the nase B receptor, are important signaling molecules in the data on PGE2 reported here, PGD2 has been implicated in developing mammalian ovary (6, 8, 9, 11–14). However, gonadal development in the mouse, specifically in male PGE2 had no significant effect on NTF5 expression (insex-specific Sertoli cell specification (22) at the time of sex deed the trend was toward a decrease in expression), indetermination. No previous data on PG function are availdicating the specificity of the effect on BDNF. BDNF and able in the developing ovary. Although a range of other NTF5 are expressed predominantly in somatic cells, but PGs may derive from the actions of COX1 and COX2, BDNF is also present in germ cells at later gestations (our PTGES specifically converts PGH2 to PGE2, and this enunpublished observations). Our results cannot differentizyme was found to be predominantly localized to germ cells. The two receptors for PGE2, EP2 and EP4, were also ate between changes in the somatic or germ cell compartgerm cell specific, suggesting an autocrine role for germ ment; but, given the localization of PGE2 receptors, the relatively short time-course of the treatment, and the lack of cell-derived PGE2. Subsequent studies investigated the regulation of po- any effect on SMAD3 mRNA, another somatic cell marker, tential target genes that can be activated by PGE2. The EP2 it may be that BDNF mRNA is elevated specifically in germ and EP4 receptors are found solely on germ cells, so this is cells and thus explain why NTF5 expression is not affected. MCL1 is an antiapoptotic member of the BCL2 family where the first changes in gene expression are likely to (26). We have previously demonstrated an increase in exoccur. We therefore assessed expression of the germ cell pression of the long isoform mRNA during the second markers OCT4, DAZL, and VASA whose expression levels are associated with different stages of germ cell devel- trimester and that it is exclusively localized to more mature opment within the fetal ovary (5). No significant changes germ cells including those within primordial follicles (4). were observed that might suggest a general effect on a We observed a small but significant increase in MCL1 certain stage of germ cell development. However, the ab- mRNA levels after PGE2 treatment, suggesting a positive sence of changes in these germ cell genes supports the spec- role for PGE2 in germ cell survival. Because MCL1 exificity of the changes in expression of the germ cell growth pression is also reported to be regulated by activin in other factor genes, rather than their being nonspecific effects on systems (27), it is possible that the small change detected here is secondary to increased activin expression, which germ cell survival. Expression of INHBA mRNA in the human fetal ovary could be explored further in longer-duration cultures. In conclusion, these data provide novel evidence for rises during the second trimester from about 17 wk gestation (10). Activin ␤A protein expression is restricted to roles for PGs, and specifically PGE2, in ovarian develop-

4060

Bayne et al.


ment. Pathways for PGE2 synthesis and action are present in germ cells in the human fetal ovary, and PGE2 can induce expression of genes known to be important in oocyte maturation and survival during the period when germ cells become associated with somatic cells to form follicles. Future studies are likely to identify roles for other PGs in both germ cell and somatic cell function.


10.

11.

12.

Acknowledgments We are grateful to Ms. Joan Creiger, Ms. Anne Saunderson, and the staff of the Bruntsfield Suite, Royal Infirmary of Edinburgh, for their assistance in obtaining the human samples used in this study. Address all correspondence and requests for reprints to: Dr. Rosemary A. L. Bayne, Medical Research Council Human Reproductive Sciences Unit, Centre for Reproductive Biology, The Queen’s Medical Research Institute, 47 Little France Crescent, Edinburgh EH16 4TJ, United Kingdom. E-mail: r.bayne@ hrsu.mrc.ac.uk. This work was supported by the UK Medical Research Council (Grants U.1276.00.002.00001 and U.1276.00.004.00002). Disclosure Summary: All authors have nothing to declare.

References 1. Fulton N, Martins da Silva SJ, Bayne RA, Anderson RA 2005 Germ cell proliferation and apoptosis in the developing human ovary. J Clin Endocrinol Metab 90:4664 – 4670 2. Baker TG 1963 A quantitative and cytological study of germ cells in human ovaries. Proc R Soc Lond B Biol Sci 158:417– 433 3. Bendsen E, Byskov AG, Andersen CY, Westergaard LG 2006 Number of germ cells and somatic cells in human fetal ovaries during the first weeks after sex differentiation. Hum Reprod 21:30 –35 4. Hartley PS, Bayne RA, Robinson LL, Fulton N, Anderson RA 2002 Developmental changes in expression of myeloid cell leukemia-1 in human germ cells during oogenesis and early folliculogenesis. J Clin Endocrinol Metab 87:3417–3427 5. Anderson RA, Fulton N, Cowan G, Coutts S, Saunders PT 2007 Conserved and divergent patterns of expression of DAZL, VASA and OCT4 in the germ cells of the human fetal ovary and testis. BMC Dev Biol 7:136 6. Anderson RA, Robinson LL, Brooks J, Spears N 2002 Neurotropins and their receptors are expressed in the human fetal ovary. J Clin Endocrinol Metab 87:890 – 897 7. Coutts SM, Childs AJ, Fulton N, Collins C, Bayne RA, McNeilly AS, Anderson RA 2008 Activin signals via SMAD2/3 between germ and somatic cells in the human fetal ovary and regulates kit ligand expression. Dev Biol 314:189 –199 8. Dissen GA, Hirshfield AN, Malamed S, Ojeda SR 1995 Expression of neurotrophins and their receptors in the mammalian ovary is developmentally regulated: changes at the time of folliculogenesis. Endocrinology 136:4681– 4692 9. Harel S, Jin S, Fisch B, Feldberg D, Krissi H, Felz C, Freimann S, Tan SL, Ao A, Abir R 2006 Tyrosine kinase B receptor and its activated

13.

14.

15.

16.

17. 18.

19.

20. 21.

22.

23.

24.

25.

26.

27.

neurotrophins in ovaries from human fetuses and adults. Mol Hum Reprod 12:357–365 Martins da Silva SJ, Bayne RA, Cambray N, Hartley PS, McNeilly AS, Anderson RA 2004 Expression of activin subunits and receptors in the developing human ovary: activin A promotes germ cell survival and proliferation before primordial follicle formation. Dev Biol 266:334 –345 Ojeda SR, Romero C, Tapia V, Dissen GA 2000 Neurotrophic and cell-cell dependent control of early follicular development. Mol Cell Endocrinol 163:67–71 Paredes A, Romero C, Dissen GA, DeChiara TM, Reichardt L, Cornea A, Ojeda SR, Xu B 2004 TrkB receptors are required for follicular growth and oocyte survival in the mammalian ovary. Dev Biol 267: 430 – 449 Seifer DB, Feng B, Shelden RM, Chen S, Dreyfus CF 2002 Brainderived neurotrophic factor: a novel human ovarian follicular protein. J Clin Endocrinol Metab 87:655– 659 Spears N, Molinek MD, Robinson LL, Fulton N, Cameron H, Shimoda K, Telfer EE, Anderson RA, Price DJ 2003 The role of neurotrophin receptors in female germ-cell survival in mouse and human. Development 130:5481–5491 Oates JA, FitzGerald GA, Branch RA, Jackson EK, Knapp HR, Roberts 2nd LJ 1988 Clinical implications of prostaglandin and thromboxane A2 formation (1). N Engl J Med 319:689 – 698 Adams IR, McLaren A 2002 Sexually dimorphic development of mouse primordial germ cells: switching from oogenesis to spermatogenesis. Development 129:1155–1164 Morita I 2002 Distinct functions of COX-1 and COX-2. Prostaglandins Other Lipid Mediat 68 – 69:165–175 Coleman RA, Smith WL, Narumiya S 1994 International Union of Pharmacology classification of prostanoid receptors: properties, distribution, and structure of the receptors and their subtypes. Pharmacol Rev 46:205–229 Negishi M, Sugimoto Y, Ichikawa A 1995 Molecular mechanisms of diverse actions of prostanoid receptors. Biochim Biophys Acta 1259: 109 –119 Narumiya S, Sugimoto Y, Ushikubi F 1999 Prostanoid receptors: structures, properties, and functions. Physiol Rev 79:1193–1226 Robinson LL, Gaskell TL, Saunders PT, Anderson RA 2001 Germ cell specific expression of c-kit in the human fetal gonad. Mol Hum Reprod 7:845– 852 Wilhelm D, Martinson F, Bradford S, Wilson MJ, Combes AN, Beverdam A, Bowles J, Mizusaki H, Koopman P 2005 Sertoli cell differentiation is induced both cell-autonomously and through prostaglandin signaling during mammalian sex determination. Dev Biol 287:111–124 Childs AJ, Anderson RA 28 May 2009 Activin A selectively represses expression of the membrane-bound isoform of Kit Ligand in human fetal ovary. Fertil Steril 10.1016/j.fertnstert.2009.03.095 Tuuri T, Era¨maa M, Van Schaik RH, Ritvos O 1996 Differential regulation of inhibin/activin ␣- and ␤ A-subunit and follistin mRNAs by cyclic AMP and phorbol ester in cultured human granulosa-luteal cells. Mol Cell Endocrinol 121:1–10 Toyomoto M, Ohta M, Okumura K, Yano H, Matsumoto K, Inoue S, Hayashi K, Ikeda K 2004 Prostaglandins are powerful inducers of NGF and BDNF production in mouse astrocyte cultures. FEBS Lett 562:211–215 Kozopas KM, Yang T, Buchan HL, Zhou P, Craig RW 1993 MCL1, a gene expressed in programmed myeloid cell differentiation, has sequence similarity to BCL2. Proc Natl Acad Sci USA 90:3516 – 3520 Fukuchi Y, Kizaki M, Yamato K, Kawamura C, Umezawa A, Hata Ji, Nishihara T, Ikeda Y 2001 Mcl-1, an early-induction molecule, modulates activin A-induced apoptosis and differentiation of CML cells. Oncogene 20:704 –713