doi:10.1093/humrep/den197
Human Reproduction Vol.23, No.8 pp. 1895–1901, 2008
Advance Access publication on June 4, 2008
The developing human ovary: immunohistochemical
analysis of germ-cell-specific VASA protein, BCL-2/BAX
expression balance and apoptosis
Mirta S. Albamonte, Miguel A. Willis, Marı́a I. Albamonte, Federico Jensen,
Marı́a B. Espinosa and Alfredo D. Vitullo1
Departamento de Estudios Biomédicos y Biotecnológicos, Centro de Estudios Biomédicos, Biotecnológicos, Ambientales y Diagnóstico,
CEBBAD, Universidad Maimónides, Hidalgo 775, C1405BCK Buenos Aires, Argentina
1
Correspondence address. E-mail: [email protected]
BACKGROUND: Germ cell number during ovarian organogenesis is regulated through programmed cell death. We
investigated the expression of germ-cell-specific VASA protein, apoptosis-related proteins BAX and BCL-2 and DNA
fragmentation in developing human ovaries from gestation week 12 to term. METHODS: Human fetal ovaries from 13
women undergoing spontaneous abortion were fixed, paraffin-embedded and processed for immunohistochemistry to
analyse temporal and cellular localization of VASA, BCL-2 and BAX, and to detect apoptosis by TUNEL assay.
RESULTS: VASA showed a differential pattern of expression throughout the differentiation and proliferative
phase and prophase I to finally associate with Balbiani’s body in primordial and primary follicles. BCL-2 was detected
from week 12 to 17 and became undetectable thereafter. Strong BAX signal was detected in oogonia and oocytes from
week 12 to term. Low levels (10%) of TUNEL positive germ cells were detectable throughout gestation with a higher
incidence (around 20%) at 18 –20 weeks. CONCLUSIONS: VASA was specifically expressed in germ cells and
displayed a stage-specific intracellular localization enabling one to follow oogenesis throughout gestation. Apoptosisinhibiting BCL-2 was associated with the germ cell proliferative phase and prophase I, whereas BAX remained positive
throughout gestation. The highest incidence of apoptotic germ cells was coincident with the lack of detectable BCL-2
protein, and when primordial follicle formation became widespread.
Keywords: human fetal ovary; female germ cells; apoptosis; VASA; BCL-2/BAX
Introduction
Primordial germ cells (PGC) migrate early during mammalian
fetal life to colonize the genital ridges where they undergo an
extensive mitotic proliferation giving rise to a large number
of oogonia, and finally entering meiosis to produce by birth a
pool of oocytes arrested at the diplotene stage of the first
meiotic division (Hirschfield, 1991). As meiosis progresses,
somatic cells surround the oocytes to form primordial follicles
and thereafter folliculogenesis dominates the formation of
germinal tissue (Hirschfield, 1991). Once PGCs differentiate
to enter meiosis and then folliculogenesis, a degenerative
process classically referred to as atresia will continuously
clear them (Kaipia and Hsueh, 1997). In the rat, germ cell
number decreases by two-thirds from the onset of meiosis to
48 h after birth (Hirschfield, 1991). In mice, 66% of the original
oogonia are decimated by the end of gestation (Flaws et al.,
2001). In humans, the process of oocyte loss is particularly
intensive; from 7 106 potential oocytes in a 20-week-old
fetus, around 85% are already lost by birth, and .99% will
be lost by the beginning of puberty (Baker, 1963; Forabosco
et al., 1991). This massive, constitutive germ cell death
depends on the genetic machinery of apoptosis, programmed
cell death (PCD) (Hsueh et al., 1996; Tilly, 1996; Tilly et al.,
1997). The majority of studies of ovarian germ cell demise in
mammals have shown that the expression of BCL-2, BCL-X
and BAX genes plays an essential role (Tilly, 2001). The
balance between apoptosis-inducing BAX and BCL-Xs and
apoptosis-inhibiting BCL-2 and BCL-Xl proteins determines
death or survival of the germ cell (Tilly, 1996; Boise et al.,
1993; Oltvai et al., 1993). As a general rule, it has been
observed that expression of BAX is normally enhanced in the
mammalian ovary, whereas BCL-2 protein is found at low
levels or undetected at all (Kim and Tilly, 2004). This observation gives support to the high rate of PCD characterizing
the mammalian ovary, and especially the human ovary.
The causes that determine massive constitutive death of
mammalian female germ cells are poorly understood. This
massive elimination may avoid the persistence in the ovary
# The Author 2008. Published by Oxford University Press on behalf of the European Society of Human Reproduction and Embryology.
All rights reserved. For Permissions, please email: [email protected]
1895
Albamonte et al.
of germ cells exhibiting nuclear or mitochondrial chromosomal/genetics defects (Tilly, 2001). Alternatively, death may
relate to the exhaustion of germ cells acting as nurse cells to
the surviving oocyte pool (Pepling and Spradling, 2001).
Finally, it has been suggested that massive death may enable
the appropriate association between germ cells and pregranulosa cells during ovigerous cords or ovarian cysts breakdown,
just before primordial follicles begin to form (Guigon and
Magre, 2006). In any case, the balance between germ cell
death and survival seems to be critical to preclude ovarian dysgenesis or premature ovarian failure and to ensure reproductive
success.
Most of our understanding of the processes acting during
prenatal conformation of the mammalian female germinal
tissue comes from studies in rat and mouse. In humans, a few
contributions on ovarian apoptosis have been made (De Pol
et al., 1997, 1998; Quenby et al., 1999; Vaskivuo et al.,
2001; Abir et al., 2002; Hartley et al., 2002; Modi et al.,
2003; Fulton et al., 2005) due to the scarcity of fetal germinal
tissue. Most of the reports have concentrated on mitotic proliferation and prophase I periods (De Pol et al., 1997, 1998;
Hartley et al., 2002; Modi et al., 2003; Fulton et al., 2005),
and only one has spanned a wider gestation period including
proliferative stage, prophase I and primordial follicle formation
(Vaskivuo et al., 2001). Consequently, a large systematic
analysis of the apoptosis-dependent germ cell degeneration
from initial proliferation to early folliculogenesis throughout
gestation is still lacking. Although the results so far published
for human ovarian apoptosis yielded some discrepancies,
several important features of fetal germ cell attrition in
humans came from these studies. First, the apoptotic activity
in the human fetal ovary is restricted to the germ cells
proper, both oogonia and oocytes, rather than to pregranulosa/granulosa cells (Vaskivuo et al., 2001), a situation that
will be reversed in the adult ovary in which apoptosis involves
primarily the granulosa cells (Billig et al., 1996; Albamonte
et al., 2005), especially during the final maturation of antral follicles (Hurst et al., 2006). Second, major apoptosis activity, as
detected by TUNEL assay in different reports (Vaskivuo et al.,
2001; Hartley et al., 2002) during ovarian genesis, is coincident
with mitotic proliferation of oogonia and the entrance to
meiosis occurring from 14 to 18– 20 gestation weeks.
Finally, the level of apoptosis increases notably in the
ovaries of fetuses with chromosomal abnormalities. As
shown by Modi et al. (2003) in their analysis of germ cell
demise, sex-chromosome aneuploid fetal human ovaries had
increased levels of apoptosis. This is an observation of particular interest since, in many cases, materials available for analysis come from abortions performed because of fetal
malformations and/or chromosomal abnormalities (Abir
et al., 2002).
In order to further investigate the dynamics of massive constitutive germ cell degeneration in the developing human
ovary, we undertook an extensive study on the distribution of
the specific germ cell marker VASA, on the evaluation of apoptotic germ cells by a DNA in situ labelling assay, and on the
balance of expression of BCL-2 and BAX genes in normal
fetal ovarian tissue spanning the period between 12 and 38
1896
gestational weeks; this includes the transition from oogonial
mitotic proliferation to the entrance to meiosis and primordial
follicle formation.
Materials and Methods
Ovarian material
This study was reviewed and approved by the Research Ethics Committee of Universidad Maimónides. Human fetal ovaries were collected from spontaneous abortions at the Hospital Municipal de
Merlo, Buenos Aires province, Argentina, after obtaining the patients
consent. The age of fetuses was estimated from the date of the last menstrual period and the crown-heel length, and foot length at autopsy. All
fetuses appeared morphologically normal. A total of 13 ovaries corresponding to 12 (n ¼ 1), 16 (n ¼ 1), 17 (n ¼ 1), 18 (n ¼ 3), 20 (n ¼ 1),
22 (n ¼ 1), 27 (n ¼ 1), 28 (n ¼ 1), 33 (n ¼ 1), 34 (n ¼ 1) and 38
(n ¼ 1) gestation weeks were analysed. Fetuses in which death was
suspected to have recently occurred before autopsy, with no sign of
necrosis, were used. Ovarian tissues were recovered in the surgery
room and immediately fixed in 10% buffered formalin. After 24 h fixation, samples were processed for routine paraffin embedding.
Embedded samples were serially sectioned at 5 mm, mounted onto
cleaned, coated slides and stored at room temperature until used. All
samples were examined for tissue integrity and general histology by
haematoxylin–eosine staining.
Immunofluorescence analysis of VASA expression
To assess VASA expression, samples were blocked in 10% bovine fetal
serum (BFS)/15% bovine serum albumin (BSA) in phosphatebuffered saline (PBS) for 1 h at room temperature, thoroughly
rinsed in PBS and incubated with polyclonal rabbit anti-Mvh (mouse
vasa homologue) primary antibodies kindly provided by Dr Noce
(Mitsubishi Kagaku Institute of Life Sciences, Tokyo, Japan) overnight at 48C. Although the primary antibody has been raised against
mouse Vasa protein, cross-reaction with human VASA protein can
be anticipated since the level of protein homology between species
is .90% (Castrillon et al., 2000), especially for the epitopes against
which antibodies were raised. After overnight incubation, slides
were appropriately rinsed in PBS and incubated with goat anti-rabbit
IgG conjugated with FITC (Santa Cruz Biotechnology, CA, USA)
45–60 min at room temperature. All slides were counterstained with
1 mg/ml propidium iodide. Negative controls were performed by
omitting the primary antibody. Samples were examined in an
Olympus BX40 microscope by conventional epifluorescence with
ultraviolet illumination and images were captured with an Olympus
Camedia C-5060 camera.
Immunohistochemical detection of BCL-2/BAX expression
Dewaxed and re-hydrated sections were incubated with rabbit polyclonal anti-BAX or anti-BCL-2 (Santa Cruz Biotechnology) primary antibodies overnight at 48C. Double-stainings were performed by using
EnVisionw Doublestain System (Dako Cytomation, USA) according
to the recommendations of the manufacturer. Immunoenzymatic reactions were performed with horse-radish peroxidase (HRP) or alkaline
phosphatase labelled polymers, revealed with 3,30 -diaminobenzidine
(DAB) and fast-red for BAX and BCL-2, respectively. All slides
were counterstained with haematoxylin. Negative controls were performed by omitting the primary antibodies.
TUNEL assay
Detection of DNA fragmentation was performed in paraffin-embedded
sections by terminal deoxynucleotidyl transferase (TdT)-mediated
Apoptosis in the human fetal ovary
deoxyuridine triphosphate (dUTP) nick end labelling technique, using
the ‘In Situ Cell Death Detection Kit’ (Roche Diagnostics, Germany)
with fluorescein tagged nucleotides. Procedure followed the suppliers’
recommendations. Treated sections were examined in an Olympus
BX40 microscope by conventional epifluorescence with ultraviolet
illumination. In order to confirm negative results, TUNEL-processed
sections were incubated with 10 IU/ml DNase II (Sigma Chemical
Co., USA) in 50 mM Tris –HCl pH 7.5, 10 mM Mg2Cl and 1 mg/ml
BSA for 10 min at room temperature. After incubation, slides were
thoroughly rinsed and treated again according to the TUNEL protocol.
Images were captured with an Olympus Camedia C-5060 camera.
Semi-quantification of expression levels
In order to assess trends in the expression of BCL-2, BAX and VASA,
cells showing a positive signal were counted using an Olympus
AX-0049, OC-M, 10/10 10 mm counting frame. Six frames were
entirely counted per section in all samples (magnification 200) in
order to have comparable areas of homogeneous cortical tissue for
each time-point. Counting was performed in the cortical region of
the ovary by two independent observers, and the results were
expressed as a mean value of absolute number for each time-point.
Results
VASA expression in the developing human ovary
VASA immunolabelling was detectable in the cytoplasm of
germ cells, both oogonia and oocytes, at all gestational ages
analysed. The immunolabelling pattern of VASA expression
in the earliest sample (gestation week 12) was faint and homogeneously distributed within the germ cell cytoplasm (Fig. 1A).
At weeks 16 and 18, the signal intensity increased progressively, became brighter and punctuated, and was localized
mainly as a perinuclear ring (Fig. 1B and C). From week 20
to the end of gestation, VASA showed the strongest signal
observed, with a clear para-nuclear localization, corresponding
to the Balbiani’s vitelline space, in oocytes lying in primordial
Figure 1: Immunodetection of the specific germ cell marker VASA
in fetal human ovaries.
(A) VASA is detected as a faint cytoplasmic staining in oogonia at
gestation week 12, (B) the signal becomes stronger by gestation
week 16 and (C) localizes mainly as a brightly and punctuated perinuclear ring by week 18. (D) With the development of primordial follicles, VASA localizes in a half-moon area with a para-nuclear
position corresponding to the Balbiani’s vitelline space. Magnification: 1000.
follicles (Fig. 1D). This was also the case for primary follicles
found at week 38.
During the mitotic proliferative stage, week 12, most germ
cells in any visual field were positive for VASA although the
signal was weak (see above). As germ cells progressed to
meiosis and finally to primordial follicle formation, absolute
numbers of VASA positive cells decreased but remained
easily detectable at any time-point.
In all gestational ages analysed, ovarian sections incubated
in the absence of primary antibodies exhibited no detectable
immunolabelling.
BCL-2/BAX expression balance
At gestational week 12, the majority of cells (around 95%),
morphologically identified as oogonia, were labelled with
anti-BCL-2 antibody (Figs. 2 and 3A). However, this number
dropped to around 50% by week 16, ,10% by week 17, and
BCL-2 immunolabelling was no longer detectable afterwards
(Figs 2 and 3B and C). It is interesting to note that the lack
of detectable BCL-2 was coincident with the appearance of primordial follicles and was maintained from 20 to 38 week. In all
cases, the BCL-2 signal localized to the germ cell cytoplasm; it
is worthwhile, however, to mention that immunolabelling in
somatic cells was also detected.
The pattern of BAX expression was quite different to that of
BCL-2. Immunolabelling with anti-BAX antibodies remained
positive throughout the entire gestation period, both in
oogonia and oocytes, and during primordial follicle formation
(Fig. 4). Positive immunostaining for BAX was stronger than
that found for BCL-2 (compare Figs 3 and 4) although the treatments followed in both cases were the same. In order to confirm
that differences in signal intensity were not related to the
enzymatic reaction employed in each case, we reversed in
some sections the use of DAB or fast-red for revealing immunoenzymatic reactions. Irrespective of this, signals were
always stronger for BAX than BCL-2. BAX immunostaining
was widespread and pregranulosa and granulosa cells were
found to be positive for BAX.
Figure 2: Absolute number of germ cells showing positive staining
for BCL-2 and BAX proteins in the developing human ovary from
gestation week 12 to term.
BAX remains positive throughout prenatal oogenesis, whereas BCL-2
is specially associated with differentiation and proliferation of germ
cells.
1897
Albamonte et al.
Figure 3: Immunodetection of apoptosis-inhibiting BCL-2 protein in
the developing human ovary.
Positive signals were detected in oogonia and oocytes at gestation
weeks 12 (A) and 16 (B) From week 18 (C) BCL-2 was no longer
detected. (D) Representative panoramic view of negative control performed by omitting the primary antibody. Magnification: 1000
(A, B, C); 200 (D).
Figure 5: Apoptosis-related DNA fragmentation detected by TUNEL
assay in the developing human ovary.
At most time-points, detection of positive cells was scarce, no .10%,
as illustrated in a 16 week section shown in (A) However, at week 20,
20% of the cells were positive (B) After treatment with DNase and
re-staining for TUNEL, all nuclei became positive indicating that
DNA was originally not fragmented (C) Magnification: 200
(A and C); 1000 (B).
Figure 4: Immunodetection of apoptosis-inducing BAX protein in
the developing human ovary.
Positive signals, before primordial follicle formation, were detectable
in oogonia and oocytes at weeks 12 and 16 (A) thereafter a stronger
mark was especially detected in oocytes from primordial follicles
until term: (B) 20 weeks, (C) 27 weeks, (D) 28 weeks and (E) 38
weeks. At term, primary follicles also showed detectable BAX
(F) Magnification: 1000 (A–C and F); 400 (D and E).
Plotting the results of semi-quantitative assessment of
BCL-2 and BAX clearly showed that the expression of
BCL-2 was specially associated with germ cell differentiation
and proliferation, from gestation week 12 to 18, whereas BAX
remained detectable throughout gestation including differentiation, proliferation, entrance into meiosis and follicle formation (Fig. 2). The majority (around 95%) of round celled
1898
oogonia, with a chromatin-homogeneous central nucleus, and
early oocytes showing a heterogeneous chromatin distribution
indicative of variable degrees of chromosome condensation as
they progress through prophase I were found positive for
BCL-2 protein at week 12, dropping to 50% by week 16 and
,1% at week 18, a situation also reflected by absolute
numbers (Fig. 2). Occasional primordial follicles, easily distinguishable by a layer of flattened somatic cells surrounding
the oocyte, were also found positive for those time-points.
BAX protein was expressed throughout gestation associated
with the three stages of germ cell development described
above, e.g. oogonia, early oocytes and primordial
follicle-enclosed oocytes. Although the absolute number
reflects a decrease from week 18 to term, the percent proportion
of BAX positive cells remained constant, just fluctuating
between 95% and 98% of the cells from week 12 to 38. It
has to be taken into account that germ cells increase in size
with maturity and the decrease in absolute number reflects
this fact.
DNA fragmentation in germ cells
Samples from all of the gestation time-points analysed showed
positive cells for TUNEL reaction both in oogonia and oocytes
(Fig. 5). In all cases, no .10% of the cells showed a positive
reaction except for samples between weeks 18 and 20, in
which the extent of apoptosis attained a 20% detection level.
TUNEL assayed sections were incubated with DNase in
order to confirm DNA integrity in negative cells. After
Apoptosis in the human fetal ovary
enzymatic treatment all nuclei became positive (Fig. 5C).
Negative controls performed by omitting TdT showed no
positive reaction.
Discussion
During the period analysed in this study, germ cells in the
developing human ovary progress in oogenesis through
mitotic proliferation of PGC, enter into meiotic prophase
I and finally associate with somatic cells to form primordial
follicles. All those essential changes in ovarian organogenesis
were reflected by the stage-specific immunodetection pattern
displayed by VASA protein. In agreement with previous
observations (Castrillon et al., 2000; Stoop et al., 2005),
signal intensity was stronger as germ cells progressed from
oogonia to oocytes, and cellular localization changed from
widespread cytoplasmic distribution to a restricted aggregation
in Balbiani’s vitelline space, in primordial-follicle enclosed
oocytes. Balbiani’s body (Balbiani, 1864) is a cytoplasmic,
para-nuclear complex composed of mitochondria, endoplasmic
reticulum and lamellar aggregates (Hertig and Adams, 1967).
The association of VASA with Balbiani’s body is intriguing;
however, its presence has been considered as a distinct trait
in fetal oocytes since it was not observed in adult ovaries
(Castrillon et al., 2000). During the third trimester, we found
Balbiani’s-associated VASA signal as a dominant feature of
primordial follicles, although in a low proportion of oocytes
(10%), VASA does not associate to Balbiani’s body and
shows a perinuclear distribution. An opposite pattern in
the VASA distribution, i.e. low abundance of Balbiani’sassociated signal, has been described in the fetal ovary of the
rodent L. maximus in which over-expression of BCL-2
precludes germ cell attrition and oocyte degeneration is
,10% (Jensen et al., 2005). It is tempting to speculate that
the association to Balbiani’s body may relate to the final
fate of the oocyte, especially considering that VASA is an
RNA-binding protein belonging to the DEAD box family
that might be involved in assembly and transport of mRNAs
during oogenesis (Castrillon et al., 2000).
Our analysis of BCL-2/BAX demonstrates that apoptosisinhibiting BCL-2 protein and apoptosis-inducing BAX
protein have different patterns of expression in the
developing human ovary, both in the temporal scale and in
cellular localization. BCL-2 expression was especially associated with germ cell differentiation and proliferation periods of
ovarian development. Conversely, BAX protein was detectable
throughout gestation with strong signals in germ cells and, with
less intensity, also in somatic cells, especially pregranulosa and
granulosa cells. We found expression of BCL-2 protein
restricted to weeks 12 to 17 localizing to germ cells and to a
lesser extent to somatic cells. Positive somatic cells adjacent
to germ cells may be ascribed to pregranulosa cells, although
in other cases, positive cells were distributed in the stroma.
Consistent with our finding, a similar temporal expression
has been previously reported by others (Vaskivuo et al.,
2001; Hartley et al., 2002) and it has also been shown that
BCL-2 expression begins as early as gestation week 6
(Quenby et al., 1999). It is clear that BCL-2 expression is
associated mainly with mitotic proliferation and entry into
meiosis periods, since we were unable to detect BCL-2 from
week 18 when primordial follicle formation becomes widespread. It is likely that apoptosis-inhibiting BCL-2 protein is
involved in germ cell survival before primordial follicles
appear in the developing ovary. Interestingly, Hartley et al.
(2002) have reported that another member of the BCL-2
family, the anti-apoptotic Mcl-1 protein, is strongly expressed
in germ cells from gestation week 18, and plays a central role in
inhibiting cell death at the time of germ cell – somatic cell
association during primordial follicle formation. Both BCL-2
and Mcl-1 are known to interact with proapoptotic BCL-2
members, such as BAX, to form heterodimers and determine
cell survival (Hsu et al., 1998). Thus, it seems that different
apoptosis-inhibiting BCL-2 members act in a concerted temporal way during prenatal oogenesis and folliculogenesis in
the developing human ovary, to regulate the default pathway
committing germ cells to death.
Unlike BCL-2 protein which showed a time-restricted
expression pattern, apoptosis-inducing BAX protein displayed
conspicuous signals from week 12 to term associated with
oogonia, primordial follicle-enclosed oocytes, and pregranulosa and granulosa cells. This result is comparable with previous reports showing widespread immunolocalization of
BAX throughout proliferation, meiotic prophase and primordial follicle formation in the developing human ovary
(Vaskivuo et al., 2001; Hartley et al., 2002). It is worthwhile
to note that throughout the period studied, BAX signal was
strong, in contrast with the slight staining found in ovaries
expressing BCL-2. Repetition of staining on different days
and using different enzymatic reactions for revealing BAX
protein confirmed that the staining was reproducible. Thus,
BAX is found throughout gestation displaying a strong and
stable pattern of immunostaining in germ cells. It is clear that
BAX is a central player in determining germ cell death in the
human ovary. The widespread expression pattern found for
apoptosis-inducing BAX in the developing human ovary is
commensurate with the levels of massive germ cell degeneration occurring throughout ovarian genesis, affecting 85% of
the original pool of oogonia by the end of gestation (Baker,
1963; Gondos et al., 1986; Forabosco et al., 1991). Nevertheless, detection of BAX by immunohistochemistry not necessarily indicates that apoptosis is active since BAX activity is
dependent on intracellular localization (Hsu et al., 1997;
Zamzami et al., 1998), and other anti-apoptotic Bcl-2
members, like Bcl-2 or Mcl-1 (Hsu et al., 1998), may be interacting with Bax to inhibiting its functional proapoptotic effect.
Studies performed mainly in rat and mice have established that
BCL-2 family members are essential players in activation or
repression of apoptosis in the mammalian ovary (Kim and
Tilly, 2004). In particular, BAX and BCL-2 genes function as
a rheostat determining cell death or survival depending on
the levels of each protein (Tilly, 2001; Tilly et al., 1997). In
fact, proapoptotic BAX seems to be up-regulated by default
in the mammalian ovary, being responsible for massive germ
cell degeneration by attrition of fetal germ cells and apoptosisdependent follicular atresia in the adult (Kim and Tilly, 2004).
Experimental studies in the rat have shown that reducing the
1899
Albamonte et al.
level of BAX expression increases granulosa cell survival and
diminishes follicular atresia, even in the absence of any change
in BCL-2 level (Tilly et al., 1995). On the other hand, increased
BAX expression is responsible for granulosa cell demise and
follicular atresia in the rat (Tilly et al., 1995), human (Kugu
et al., 1998), monkey (Uma et al., 2003) and quail (Van
Nassauw et al., 1999) ovary. Recently, it has been shown in
a caviomorph rodent that natural over-expression of BCL-2
in the ovary, accompanied by a down-regulation of BAX, promotes continuous folliculogenesis with suppressed apoptosisdependent follicular atresia and reduced degeneration of
germ cell in fetal life (Jensen et al., 2005, 2006). The results
presented here and previous reports (Quenby et al., 1999;
Vaskivuo et al., 2001; Abir et al., 2002; Hartley et al., 2002;
Fulton et al., 2005) indicate that Bcl-2 family members are
involved in oocyte degeneration in the human developing
ovary acting in a concerted way during mitotic proliferation,
entrance to meiosis and primordial follicle formation, with
Bax being a central player.
The levels of apoptosis revealed by TUNEL did not correlate
with the high incidence of germ cell attrition reported in classical studies (Baker, 1963; Gondos et al., 1986; Forabosco
et al., 1991). The prevalence of TUNEL positive cells from
week 12 to term was low, ,10%, except for samples belonging
to gestation weeks 18– 20 in which we recorded around 20%
apoptotic nuclei. The increase in apoptosis level at 18– 20
weeks was coincident with the lack of detectable anti-apoptotic
BCL-2. This may well reflect degeneration of oocytes in which
apoptosis started earlier in oogenesis since it is not possible to
relate the positive stain to the actual time it takes to disappear a
dying cell from the tissue, and DNA degradation is a late event
in the sequence of cell death (Collins et al., 1997). Similarly, an
increased incidence of apoptosis reaching 14– 17% has been
shown by others to take place between weeks 14 and 20, and
,10% apoptotic cells are found at week 13 and .22 gestation
weeks (Vaskivuo et al., 2001). In other reports, the levels of
apoptosis revealed by TUNEL were ,5% for gestation
periods spanning 19– 33 weeks (Abir et al., 2002) or 13 –18
weeks (Hartley et al., 2002). In agreement with our results,
an analysis of cleaved caspase-3 immunodetection has shown
that an increase in apoptosis takes place preceding primordial
follicle formation (Fulton et al., 2005). Thus, it seems that a
critical increase in apoptosis occurs before the developing
ovary reaches the primordial follicle formation stage. Classical
studies have shown that progression from mitosis to meiotic
prophase I spans from weeks 11 to 20 as overlapped processes
in the human ovary (Kurilo, 1981), and it is believed that cell
death increases from these periods to reach a final 85%
oocyte degeneration at the time of birth (Baker, 1963;
Gondos et al., 1986). Nevertheless, the apoptosis incidence
reported in this study and by others is far too low to explain
this intensive level of cell death reported in classical studies.
An incidence of 10– 20% TUNEL positive cells is close to
the level of oocyte attrition at week 20 recorded in classical
studies (Baker, 1963; Baker and Franchi, 1967), but from this
time-point to term no increase in apoptosis was detected. It is
worthwhile to consider that other mechanisms may be at
work in the developing human ovary. For instance, it has
1900
been demonstrated that germ cell exfoliation from the
ovarian surface contributes to eliminate PGC as well as
oogonia and oocytes (Motta et al., 1997).
In conclusion, these data demonstrate that Bcl-2 family plays
an essential role in regulating germ cell number in the human
developing ovary through the concerted temporal expression
of both pro- and anti-apoptotic members at the different cell
compartments during oogenesis. It is worthwhile to note that
although some data have accumulated on apoptosis in the
human developing ovary, we are still far from having a detailed
knowledge on the regulation of germ cell number that is drastically reduced from 7 106 cells at the fifth month postconception to around 750 000 oocytes at birth (Baker, 1963;
Forabosco et al., 1991). Future analysis spanning the entire
ovarian organogenesis and including more members of the
Bcl-2 family and other apoptosis-related proteins will help to
conduct a meta-analysis on fetal germ cell degeneration
which is still pending.
Acknowledgement
We specially thank Dr Silvia Galeano from Merlo Hospital for kindly
providing the ovarian samples. We are especially indebted to Dr Noce
from the Mitsubishi Kagaku Institute of Life Sciences, Tokyo, Japan,
for his generous supply of Mvh (VASA) antibodies.
Funding
MSA was supported through a research fellowship from Serono
Argentina SA. This research was funded by the Ministry of
Public Health through a Ramón Carrillo-Arturo Oñativia
fellowship awarded to ADV, the National Research Council
of Argentina, CONICET (PIP 6504/06), and Universidad
Maimónides, Argentina.
References
Abir R, Orviet R, Dicker D, Zukerman Z, Barnett M, Fisch B. Preliminary
studies on apoptosis in human fetal ovaries. Fertil Steril 2002;78:259–262.
Albamonte MS, Elizondo K, Albamonte MI, Willis MA, Espinosa MB, Vitullo
AD. Expresión de BCL-2/BAX y apoptosis en ovario humano adulto.
Medicina Bs As 2005;65(Suppl II):126.
Balbiani EG. Sur la constitution du germe dans ĺoeuf animal avant la
fédcondation. Compt Rend Acad Sci 1864;58:584– 588.
Baker TG. A quantitative and cytological study of germ cells in human ovaries.
Proc R Soc Lond B 1963;158:417– 433.
Baker TG, Franchi LL. The fine structure of oogonia and oocytes in human
ovaries. J Cell Biol 1967;2:213– 224.
Billig H, Chun SY, Eisenhauser K, Hsueh AJW. Gonadal cell apoptosis:
hormone-regulated cell demise. Hum Reprod 1996;2:103–117.
Boise LH, González-Garcia M, Postema CE, Ding L, Lindsten T, Turka LA,
Mao X, Núñez G, Thompson CB. bcl-x, a bcl-2 related gene that functions
as a dominant regulator of apoptotic cell death. Cell 1993;4:597–608.
Castrillon DH, Quade BJ, Wang TY, Quigley C, Crum CP. The human VASA
gene is specifically expressed in the germ cell lineage. Proc Natl Acad Sci
USA 2000;97:9585–9590.
Collins JA, Schandl CA, Young KK, Vesely J, Willingham MC. Major DNA
fragmentation is a late event in apoptosis. J Histochem Cytochem
1997;45:923–934.
De Pol A, Vaccina F, Forabosco A, Cavazzuti E, Marzona L. Apoptosis of germ
cells during human prenatal oogenesis. Hum Reprod 1997;12:2235– 2241.
De Pol A, Marzona L, Vaccina F, Negro R, Sena P, Forabosco A. Apoptosis in
different stages of human oogenesis. Anticancer Res 1998;18:3457–3462.
Apoptosis in the human fetal ovary
Flaws JA, Hirshfield AN, Hewitt JA, Babus JK, Furth PA. Effect of Bcl-2 on
the primordial follicle endowment in the mouse ovary. Biol Reprod
2001;64:1153–1159.
Forabosco A, Sforza C, De Pol A, Vizzotto L, Marzona L, Ferrario VL.
Morphometric study of the human neonatal ovary. Anat Rec
1991;231:201–208.
Fulton N, Martins da Silva RA, Bayne RAL, Anderson RA. Germ cell
proliferation and apoptosis in the developing human ovary. J Clin
Endocrinol Metab 2005;90:4664–4670.
Gondos B, Westergaard L, Byskov AG. Initiation of oogenesis in the human
fetal ovary: ultrastructural and squash preparation study. Am J Obstet
Gynecol 1986;155:189–195.
Guigon CJ, Magre S. Contribution of germ cells to differentiation and
maturation of the ovary: insights from models of germ cell depletion. Biol
Reprod 2006;74:450–458.
Hartley PS, Bayne RAL, Robinson LLL, Fulton N, Anderson RA.
Developmental changes in expression of myeloid cell leukaemia-1 in
human germ cells during oogenesis and early folliculogenesis. J Clin
Endocrinol Metab 2002;87:3417–3427.
Hertig AT, Adams EC. Studies on the human oocyte and its follicle. I
Ultrastructural and histochemical observations on the primordial follicle
stage. J Cell Biol 1967;34:647– 675.
Hirschfield AN. Development of follicles in the mammalian ovary. Int Rev
Cytol 1991;121:43– 101.
Hsu SY, Kaipia A, McGee E, Lomeli M, Hsueh AJ. Bok is a pro-apoptotic
Bcl-2 protein with restricted expression in reproductive tissues and
heterodimerizes with selective anti-apoptotic Bcl-2 family members. Proc
Natl Acad Sci USA 1997;94:12401–12406.
Hsu SY, Lin P, Hsueh AJ. BOD (Bcl-2-related ovarian death gene) is an
ovarian BH3 domain-containing proapoptotic Bcl-2 protein capable of
dimerization with diverse antiapoptotic Bcl-2 members. Mol Endocrinol
1998;12:1432–1440.
Hsueh AJW, Eisenhauer K, Chun SY, Hsu SY, Billig H. Gonadal cell
apoptosis. Recent Prog Horm Res 1996;51:433–455.
Hurst PR, Mora JM, Fenwick MA. Caspase-3, TUNEL and ultrastructural
studies of small follicles in adult human ovarian biopsies. Hum Reprod
2006;21:1974–1980.
Jensen F, Leopardo L, Espinosa MB, Vitullo AD. Inhibición de la apoptosis en
el ovario fetal de Lagostomus maximus. Medicina Bs As 2005;65(Suppl
II):125.
Jensen F, Willis MA, Albamonte MS, Espinosa MB, Vitullo AD. Naturally
suppressed apoptosis prevents follicular atresia and oocyte reserve decline
in the adult ovary of Lagostomus maximus (Rodentia, Caviomorpha).
Reproduction 2006;132:301– 308.
Kaipia A, Hsueh AJW. Regulation of ovarian follicle atresia. Annu Rev Physiol
1997;59:349–363.
Kim MR, Tilly JL. Current concepts in Bcl-2 family member regulation of
female germ cell development and survival. Biochem Biophys Acta
2004;1644:205– 210.
Kugu K, Ratts VS, Piquette GN, Tilly KI, Tao XJ, Martimbeau S, Aberdeen
GW, Krajewski S, Reed JC, Pepe GJ et al. Analysis of apoptosis and
expression of the bcl-2 gene family members in the human and baboon
ovary. Cell Death Differ 1998;5:67– 76.
Kurilo LF. Oogenesis in antenatal development in man. Hum Genet
1981;57:86–92.
Modi DN, Sane S, Bhartiya D. Accelerated germ cell apoptosis in sex
chromosome aneuploid fetal human gonads. Mol Hum Reprod
2003;9:219–225.
Motta PM, Makabe S, Nottola SA. The ultrastructure of human
reproduction. I. The natural history of the female germ cell: origin,
migration and differentiation inside the developing ovary. Hum Reprod
Update 1997;3:281–295.
Oltvai ZN, Milliman CL, Korsmeyer SJ. Bcl-2 heterodimerizes in vivo with a
conserved homolog, Bax, that accelerates programmed cell death. Cell
1993;74:609– 619.
Pepling ME, Spradling AC. Mouse ovarian germ cell cysts undergo programmed
breakdown to form primordial follicles. Dev Biol 2001;234:339–351.
Quenby SM, Gazvani MR, Brazeau C, Neilson J, Lewis-Jones DI, Vince G.
Oncogenes and tumor suppressor genes in the first trimester human fetal
gonadal development. Mol Hum Reprod 1999;5:737–741.
Stoop H, Honecker F, Cools M, de Krijger R, Bokemeyer C, Looijenga LHJ.
Differentiation and development of human female germ cells during
prenatal gonadogenesis: an immunohistochemical study. Hum Reprod
2005;20:1466–1476.
Tilly JL. Apoptosis and ovarian function. Rev Reprod 1996;1:162– 172.
Tilly JL. Commuting the death sentence: How oocytes strive to survive. Nat
Rev 2001;2:838–848.
Tilly JL, Tilly KL, Kenton ML, Johnson AI. Expression of members of the bcl2 gene family in the immature rat ovary: equine chorionic gonadotropin–
mediated inhibition of granulosa cell apoptosis is associated with
decreased bax and constitutive bcl-2 and bcl-xlong messenger RNA levels.
Endocrinology 1995;136:232–241.
Tilly JL, Tilly KI, Perez GI. The genes of cell death and cellular susceptibility
to apoptosis in the ovary: A hypothesis. Cell Death Differ 1997;4:
180– 187.
Uma J, Muraly P, Verma-Kumar S, Medhamurthy R. Determination of onset of
apoptosis in granulosa cells of the preovulatory follicles in the bonnet
monkey (Macaca radiat): correlation with mitogen-activated protein
kinase activities. Biol Reprod 2003;69:1379– 1387.
Van Nassauw L, Tao L, Harrison F. Distribution of apoptosis-related proteins
in the quail ovary during folliculogenesis: BCL-2, BAX and CPP32. Acta
Histochem 1999;101:103– 112.
Vaskivuo TE, Anttonen M, Herva R, Billig H, Dorland M, te Velde ER,
Stenbäck F, Heikinheimo M, Tapanainen JS. Survival of human ovarian
follicles from fetal to adult life: Apoptosis, apoptosis-related proteins, and
transcription factor GATA-4. J Clin Endocrinol Metab 2001;86:3421–3429.
Zamzami N, Brenner C, Marzo I, Susin SA, Kroemer G. Subcellular and
submitochondrial mode of action of Bcl-2-like oncoproteins. Oncogene
1998;16:2265–2282.
Submitted on October 19, 2007; resubmitted on April 18, 2008; accepted on
April 22, 2008
1901