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,
J Clin Endocrinol Metab, October 2009, 94(10):4053– 4060
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,
J Clin Endocrinol Metab, October 2009, 94(10):4053– 4060
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.
jcem.endojournals.org
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.
PGE2 in Ovarian Development
J Clin Endocrinol Metab, October 2009, 94(10):4053– 4060
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
J Clin Endocrinol Metab, October 2009, 94(10):4053– 4060
jcem.endojournals.org
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.
PGE2 in Ovarian Development
J Clin Endocrinol Metab, October 2009, 94(10):4053– 4060
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
J Clin Endocrinol Metab, October 2009, 94(10):4053– 4060
jcem.endojournals.org
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.
PGE2 in Ovarian Development
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.
J Clin Endocrinol Metab, October 2009, 94(10):4053– 4060
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, Erämaa 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