BIOLOGY OF REPRODUCTION 62, 103–107 (2000)
Ovarian Follicle Apoptosis in Bovine Growth Hormone Transgenic Mice 1
Natalia A. Danilovich,3 Andrzej Bartke,2,3 and Todd A. Winters4
Department of Physiology,3 Southern Illinois University School of Medicine, Carbondale, Illinois 62901-6512
Department of Animal Science, Food and Nutrition,4 Southern Illinois University, Carbondale, Illinois 62901-4417
ABSTRACT
ovaries of transgenic metallothionein (MT) bovine GH
(bGH) mice [7] suggests that GH, directly or indirectly,
may interfere with mechanisms responsible for atresia and
thus increase the number of ova shed during each ovulation
[6].
In contrast to this evidence for stimulatory effects of GH
on reproductive function, acromegaly, which is accompanied by pathological GH excess, is often associated with
distinct reproductive dysfunction such as amenorrhea, menstrual irregularities, and impotence [8]. Female human GH
transgenic mice are sterile [9]. Severe reproductive defects
have been reported in transgenic pigs and sheep overexpressing GH [10,11]. In addition, GH administration to lactating goats had no positive effects on the ovaries [12].
Recent studies have suggested that apoptosis is the underlying mechanism of ovarian follicle degeneration during
atresia [13–15]. Several hormones and factors have been
reported to play key roles in GC apoptosis, with gonadotropins and estrogens preventing apoptosis and androgens
antagonizing the effect of estrogens [16,17]. In the rat,
treatment with gonadotropins or IGF-I suppressed apoptotic
DNA fragmentation of preovulatory follicles in a dose-dependent manner [16]. In addition, GH inhibited preovulatory follicle apoptosis by stimulating the production of endogenous IGF-I [18].
We decided to use phosphoenolpyruvate carboxykinase
(PEPCK) bovine GH (bGH) transgenic mice overexpressing GH to study the effects of prolonged elevation of peripheral GH levels on levels of apoptosis in ovaries. The
presence of apoptosis was assessed by terminal deoxynucleotidyl transferase (TdT) dUTP nick end-label (TUNEL)
staining, an in situ end-labeling method.
Growth hormone directly or via insulin like-growth factor-I
has been shown to inhibit preovulatory follicle apoptosis, which
is the underlying mechanism of follicular atresia. We studied the
levels of apoptosis in the ovaries of transgenic mice expressing
bovine growth hormone. Female bovine growth hormone transgenic mice (n 5 10) and nontransgenic litter mates (n 5 8) were
killed at early proestrus. Ovaries were collected, sectioned, and
processed using a nonradioactive in situ method for apoptosis
detection. Follicles were classified and counted on the basis of
size and level of apoptosis. Our results demonstrate that the
percentage of ovarian follicles containing apoptotic cells was
lower in transgenic versus normal mice (30% vs. 46%; P ,
0.05). The percentage of follicles undergoing heavy apoptosis
was lower (P , 0.05) in transgenic versus control animals in
preovulatory and early antral follicles, but it was not different
in preantral follicles. The percentage of healthy preovulatory follicles was also higher in transgenic versus normal mice (7.4%
vs. 4.3%; P , 0.05). These results indicate that growth hormone
overexpression in transgenic mice significantly decreases follicle
apoptosis, and thus atresia in the mouse ovary, therefore leading
to increased propensity for ovulation in these animals.
INTRODUCTION
Although the involvement of growth hormone (GH) in
the control of ovarian function has been well documented,
the mechanisms of GH action on reproductive organs are
poorly understood. It has been suggested that at least some
of the actions of GH on the granulosa cells (GCs) may be
mediated by liver-derived or locally produced insulin-like
growth factor-I (IGF-I) and that GH also has direct, nonIGF-mediated effects on GC differentiation [1]. IGF-I synergizes with gonadotropins in the stimulation of a variety
of GC functions, including estrogen and progesterone production [2], as well as the formation of LH and hCG receptors [3]. Studies in IGF-I null mutant mice have revealed
significantly reduced serum concentrations of estradiol,
which is accompanied by primary gonadal failure in females due to gonadotropin resistance at the level of the GC
[4,5].
The numbers of implantation sites and corpora lutea in
transgenic mice overexpressing GH and in normal mice injected with GH were significantly greater than the corresponding values in normal animals [6]. Significant reduction in the number of follicles with signs of atresia in the
MATERIAL AND METHODS
Animals
Transgenic mice used in the present study were derived
from a single male founder produced by microinjection of
hybrid genes consisting of a promoter region of rat PEPCK
gene and the coding region of the bGH gene into male
pronuclei of recently fertilized mouse eggs [19]. The line
was propagated by mating hemizygous transgenic males to
normal C57 BL/63C3H F1 hybrid females purchased from
the Jackson Laboratory (Bar Harbor, ME).
Design of the Experiment
Adult (2–4 mo of age) females were housed in litter
mate groups (3–5 animals per cage) under standard laboratory conditions with controlled illumination (12L:12D)
and temperature (228C) and had unrestricted access to food
and water. Transgenic (Tg; n 5 10) and normal, nontransgenic (N; n 5 8) litter mates were distinguished on the basis
of body weight, shape of the head, body proportions, and
pelage characteristics. The estrous cycle pattern was assessed by daily examination of cellular composition of vaginal washings for at least five cycles. Only females having
This work was supported by Illinois Council for Food and Agricultural
Research, the SIU-C University Priorities and Interdisciplinary Initiative
Program, and the National Institute of Child Health and Human Development (HD 20033 and HD 20001)
2
Correspondence. FAX: 618 453 1517; e-mail: [email protected]
1
Received: 11 September 1998.
First decision: 1 October 1998.
Accepted: 3 September 1999.
Q 2000 by the Society for the Study of Reproduction, Inc.
ISSN: 0006-3363. http://www.biolreprod.org
103
104
DANILOVICH ET AL.
regular 4- to 6-day estrous cycles were included in the experiment. Mice were killed at early proestrus, and ovaries
were collected. One ovary of each animal was processed
for detection of apoptosis in situ.
To evaluate atretic follicles, several histological characteristics were used: disruption of the granulosa cell layer,
pyknotic GCs, and thinning of the GC layer with some
degree of hypertrophy of the theca cell layer [21].
Detection of Apoptosis by TUNEL
Statistical Evaluation of Data
Histological tissue sections from both Tg and N ovaries
were stained for apoptotic DNA (Oncor, Gaithersburg,
MD). Each ovary was fixed in 4% buffered formaldehyde
overnight, dehydrated, and embedded in paraffin. The ovaries embedded in paraffin were cut into 5-mm sections and
mounted on silane-coated glass slides. Each seventh section
was used for TUNEL staining. After deparaffinization in
xylene, tissue sections were incubated with proteinase K
(20 mg/ml) and then treated with 2% hydrogen peroxide
(H2O2) in methanol to block endogenous peroxidase activity. The slides were subsequently pretreated with TdT buffer, incubated with anti-digoxigenin antibody conjugated to
peroxidase, stained in a solution of diaminobenzidine, the
substrate for peroxidase, and counterstained in methyl
green (Oncor). After counterstaining, the slides were
washed in 100% butanol, cleared in 3 changes of xylene,
and mounted with Permount (Fisher Scientific Company,
Fair Lawn, NJ). Negative controls were processed in an
identical manner in each experimental run. For negative
controls, the TdT buffer was substituted with the same volume of distilled water. Biological positive control slides
were included with the kit. These control slides contained
rat mammary glands obtained on the fourth day after weaning when extensive apoptosis occurs.
In the present study, in order to distinguish between the
specific and nonspecific staining of nuclei, the volume of
terminal transferase enzyme was reduced. The diminished
terminal transferase content, 11 ml instead of 16 ml, of the
working strength TdT was used. The omitted volume of
TdT was substituted by an equal volume of water. This
resulted in slightly decreased staining of cells containing
apoptotic bodies, whereas nonspecific staining of nuclei
disappeared.
The data were analyzed by one way ANOVA and a Fischer’s LSD post-hoc test using P , 0.05 as the level of
significance. All results are presented as mean 6 SEM.
Histological Assessment of Ovarian Follicles
For classification of apoptosis in the follicles (details below), preantral, early antral, and preovulatory follicles were
used, while primordial follicles were ignored [20]. We determined the number of preantral follicles (characterized by
at least two layers of GCs), the number of early antral follicles with a confluent antrum, and the number of preovulatory follicles (characterized by a bigger antral cavity and
displacement of the oocyte to eccentric position). Each follicle category was subdivided into three stages of apoptosis:
1) healthy follicles with undetected level of apoptosis (A0);
2) follicles containing less than 20 apoptotic cells in the
granulosa compartment, classified as follicles with slight
level of apoptosis (A1); and 3) follicles with 20 or more
apoptotic cells in the granulosa compartment, classified as
follicles with marked level of apoptosis (A2).
Follicles in each seventh section were counted. Thus, a
total of 7 of 50 sections per ovary were evaluated. Follicles
were only counted when they contained the nucleus of the
oocyte. In each ovary, the follicles were counted three
times, on three different days, to avoid mistakes. The results
were expressed as the percentages of preantral, early antral,
and preovulatory follicles with different (A0, A1, and A2)
stages of apoptosis of total counted follicles (sum of preantral, early antral, and antral).
RESULTS
The total number of follicles (preantral and antral) per
mouse was not different between Tg (22 6 0.8) and N (21
6 1.2) groups of mice. More early antral and preovulatory
follicles were present in the ovaries of Tg females in comparison to ovaries from N females, whereas the number of
preantral follicles was greater in N mice compared to Tg
animals. However, the differences were not significant since
the values of each follicular category displayed high ranges.
As shown in Figure 1B, positive staining was observed
most abundantly in large follicles. Mouse ovaries in both
groups displayed staining of unhealthy atretic follicles (Fig.
1C). Positive staining was also noted in early antral follicles
of N and Tg animals. No labeling was noted among primordial follicles or in the interstitial tissue in either group.
As shown in Figure 1G, in Tg mice there was negligible
incorporation of digoxigenin-deoxy-UTP in any of the follicles with small antra. In N mice, there were early antral
follicles in advanced atresia with heavy staining suggestive
of high level of apoptosis (Fig. 1H), and many antral follicles were present in different stages of atresia. In both
groups studied, there was minimal staining of preantral follicles of any size. In contrast, early antral follicles in N
mice (29.8%) as well as in Tg mice (17.1%) demonstrated
maximal positive staining compared to preantral and preovulatory follicles in both groups (Fig. 1H). Healthy preantral or antral follicles were consistently negative (Fig. 1, D,
A, and F). Although the nuclear staining involved mainly
GCs in the positive follicles, scattered cells in the theca
interna layer were also involved (Fig. 1E). No staining was
seen in the negative control group when terminal transferase was deleted from the reaction (data not shown). In contrast, the biological positive control (rat mammary glands
obtained on the fourth day after weaning) demonstrated
positive staining cells containing DNA fragmentation (data
not shown).
Histological analysis of ovarian sections showed that
some antral follicles (sum of early antral and preovulatory)
in Tg (29.8%) and in N (39.4%) animals were atretic, with
pyknotic nuclei formation, cell shrinkage, detached or thin
layers of GCs, and a hypertrophy of the theca interna (Fig.
1, B and C).
The percentage of follicles demonstrating apoptosis in
the ovaries from the Tg and N mice was 30% and 46%,
respectively (P , 0.05). The percentage of preantral follicles with an undetected level of apoptosis (A0) was not
different between two groups of mice. A slight level of
apoptosis (A1) was observed in , 0.6% of preantral follicles in the Tg mice compared to the ovaries of N mice that
demonstrated 6.2% (P , 0.05) (Fig. 2). In contrast, preantral follicles did not show heavy staining for apoptosis (A2)
in either group of mice. The percentage of early antral follicles with undetected level of apoptosis (A0) was significantly larger in Tg (29%) compared with the N mice
(15.9%) (P , 0.05) (Fig. 2). The percentages of early antral
GH AND OVARIAN APOPTOSIS
105
FIG. 1. Immunohistochemical detection of DNA fragmentation by the TUNEL method in ovaries of PEPCK-bGH transgenic and normal mice: preovulatory follicle with A) undetected level of apoptosis (Ao), B) slight level of apoptosis (A1), and C) marked level of apoptosis (A2); preantral follicle with
D) undetected level of apoptosis (Ao) and E) slight level of apoptosis (A1); early antral follicle with F) undetected level of apoptosis (Ao), G) slight level
of apoptosis (A1), and H) marked level of apoptosis (A2).
106
DANILOVICH ET AL.
FIG. 2. Percentage of ovarian preantral (Pa), early antral (Ea); and preovulatory (Po) follicles with undetected (Ao), slight (A1), and marked (A2)
levels of apoptosis in the adult normal and PEPCK-bGH transgenic mice.
Values are expressed as mean 6 SEM. Asterisks denote statistical significance between groups (P , 0.05).
follicles with a slight level of apoptosis (A1) were not different between groups. However, the percentage of early
antral follicles with a marked level of apoptosis (A2) appeared to be reduced in the Tg animals (1.5%) compared
with N mice (3.6%) (Fig. 2).
The percentages of preovulatory follicles with undetected and slight levels of apoptosis did not differ between the
two groups. The apparent difference in the proportion of
preovulatory follicles with a marked level of apoptosis (A2)
between Tg and the N group of animals, 3.3% and 6.6%,
respectively, was not statistically significant. (Fig. 2).
DISCUSSION
There is a plethora of evidence that GH, directly or via
IGF-I, regulates the reproductive system in females and influences several ovarian functions including stimulation of
folliculogenesis, increase in ovulation rate, and enhancement of steroidogenesis, including estrogen and progesterone production. In the present study we investigated the
hypothesis that GH (or IGF-I) inhibits apoptotic cell death
in the ovary of Tg mice with chronic exposure to excessive
bGH levels.
Histologically, the present study demonstrates that apoptosis in the mouse ovary is associated with atretic follicles.
These results are consistent with findings in rats [13,15,22].
The cells involved are for the most part GCs; however,
occasional theca cells also undergo apoptosis. The results
presented here are in agreement with reports that apoptosis
is a mechanism responsible for the demise of GCs during
follicular atresia [13,14,21].
The total numbers of preantral as well as antral follicles
were not different between groups, which is in agreement
with the report that GH does not increase the number of
follicles in Tg mice overexpressing GH [7].
Our observation that GH significantly reduces the number of apoptotic cells in preantral follicles in Tg mice compared to N animals is in agreement with reports of a reduction of the percentages of preantral follicles with signs
of atresia in MT-bGH Tg mice [7]. Moreover, preantral follicles did not show heavy staining for apoptosis (A2) in
either group of mice. This finding is correlated with reports
that apoptosis is minimal in preantral follicles [23]. In addition, previous histological investigations revealed mini-
mal atresia or apoptosis of granulosa cells within preantral
follicles in vivo [21].
Results obtained in the present study indicate that more
early antral follicles display stained nuclei compared to
preantral and preovulatory follicles in both groups of mice.
Such observations are consistent with the conclusion that
the majority of follicles undergo degeneration at the early
antral follicle stage [23,24]. In the present study, overexpression of GH was unable to completely suppress apoptosis in early antral follicles. However, the Tg mice showed
significantly reduced apoptosis of these follicles compared
to the N animals. It has been previously shown that GH
markedly increases levels of IGF-I mRNA transcripts in
early antral follicles, but the production of local IGF-I induced by GH is insufficient to suppress apoptosis in follicles at this stage of development [25].
Our findings suggesting that apoptosis of preovulatory
follicles was also reduced in Tg mice compared with the
N animals are consistent with the previous observations that
GH inhibits apoptosis of preovulatory follicles via the production of endogenous IGF-I [18]. In addition, it has been
shown that apoptosis of preovulatory follicles is prevented
by treatment with IGF-I [16]. The levels of IGF-I receptor
are higher in GCs from large rather than small antral follicles in rat, human, and bovine ovaries and are up-regulated by gonadotropins [26,27]. Several locally produced
intraovarian regulators such as growth factors including
IGF-I, epidermal growth factor (EGF), and basic fibroblast
growth factor (bFGF) have been shown to suppress follicle
apoptosis in cultured preovulatory follicles [16,22]. The
number of EGF receptors in preovulatory follicles is much
higher than in early antral follicles [22], which also may
explain the higher level of suppression of apoptosis in preovulatory as compared with early antral follicles (13% vs.
26.2% in N animals; 12.6% vs. 17% in Tg mice).
The mechanism by which GH leads to an increase in
cell survival is unknown. Some investigators have reported
that GH exerts its action directly on granulosa and thecainterstitial cells in the ovary independent of IGF-I [1,28].
On the other hand, there is a growing body of evidence that
GH may not act directly on follicles but acts via enhanced
production of IGF-I in the ovary [5,29] or via increased
concentrations of IGF-I/insulin in the serum [30] that is due
to GH positive regulation of the IGF synthesis by the liver
and other tissues [31].
In summary, the following conclusions can be made
from the present study: 1) apoptosis is the mechanism responsible for follicular atresia in the mouse ovary, and the
GCs are a major cell type undergoing apoptosis; 2) early
antral follicles are most vulnerable to undergo atretic degeneration, whereas apoptosis of preantral follicles is minimal; 3) the level of apoptosis was significantly reduced in
Tg mice overexpressing GH compared to levels in N mice.
The overall results of the present investigation indicate that
overexpression of bGH does restrain, directly or via IGFI, follicle apoptosis in mouse ovaries, thus increasing the
propensity for ovulation in these animals.
ACKNOWLEDGMENTS
The technical assistance of Dr. Wenguang Cao and Maureen Doran is
gratefully acknowledged. We thank Mae Hilt for typing.
REFERENCES
1. Hutchinson LA, Findlay JK, Herington AC. Growth hormone and
insulin-like growth factor-I accelerate PMSG-induced differentiation
of granulosa cells. Mol Cell Endocrinol 1988; 55:61–69.
GH AND OVARIAN APOPTOSIS
2. Adashi EY, Resnick CE, D’Ercole AJ, Svoboda ME, Van Wyk JJ.
Insulin-like growth factors as intraovarian regulators of granulosa cell
growth and function. Endocr Rev 1985; 6:400–420.
3. Davoren JB, Hsueh AJW, Li CH. Somatomedin C augments follicle
stimulating hormone induced differentiation of cultured rat granulosa
cells. Am J Physiol 1986; 249:E26–E33.
4. Baker J, Hardy MP, Zhou J, Bondy C, Lupu F, Bellve AR, Efstratiadis
A. Effects of an IGF1 gene null mutation on mouse reproduction. Mol
Endocrinol 1996; 10:903–918.
5. Zhou J, Kumar TR, Matzuk MM, Bondy C. Insulin-like growth factor
I regulates gonadotropin responsiveness in the murine ovary. Mol Endocrinol 1997; 11:1924–1933.
6. Cecim M, Kerr J, Bartke A. Effects of bovine growth hormone (bGH)
transgene expression or bGH treatment on reproductive functions in
female mice. Biol Reprod 1995; 52:1144–1148.
7. Mayerhofer A, Weis J, Bartke A, Yun JS, Wagner TE. Effects of
transgenes for human and bovine growth hormones on age-related
changes in ovarian morphology in mice. Anat Rec 1990; 227:175–
186.
8. Jadresic A, Banks LM, Child DF, Diamant L, Doyle FH, Fraser TR,
Joplin GF. The acromegaly syndrome. Q J Med 1982; 202:189–204.
9. Bartke A, Steger RS, Hodges SL, Parkening TA, Collins TJ, Yun JS,
Wagner TE. Infertility in transgenic female mice with human growth
hormone expression: evidence for luteal failure. J Exp Zool 1988; 248:
121–124.
10. Rexroad CE Jr, Hammer RE, Bolt DJ, Mayo KE, Frohman LA, Palmiter RD, Brinster RL. Production of transgenic sheep with growthregulating genes. Mol Reprod Dev 1989; 1:164–169.
11. Pursel VG, Bolt DJ, Miller KF, Pinkert CA, Hammer RE, Palmiter
RD, Brinster RL. Expression and performance in transgenic pigs. J
Reprod Fertil Suppl 1990; 40:235–243.
12. Driancourt MA, Disenhaus C. Lack of effects of growth hormone
administration on ovarian function of lactating goats. Anim Reprod
Sci 1997; 46:123–132.
13. Hughes FM Jr, Gorospe WC. Biochemical identification of apoptosis
(programmed cell death) in granulosa cells: evidence for a potential
mechanism underlying follicular atresia. Endocrinology 1991; 129:
2415–2422.
14. Tilly JL, Kowalski KI, Johnson AL, Hsueh AJW. Involvement of apoptosis in ovarian follicular atresia and postovulatory regression. Endocrinology 1991; 129:2799–2801.
15. Hsueh AJW, Billig H, Tsafriri A. Ovarian follicle atresia: a hormonally
controlled apoptotic process. Endocr Rev 1994; 15:707–724.
16. Chun SY, Billig H, Tilly JL, Furuta I, Tsafriri A, Hsueh AJW. Gonadotropin suppression of apoptosis in cultured preovulatory follicles:
mediatory role of endogenous insulin-like growth factor-I. Endocrinology 1994; 135:1845–1853.
17. Billig H, Furuta I, Rivier C, Tapanainen J, Parvinen M, Hsueh AJW.
Apoptosis in testes germ cells: developmental changes in gonadotro-
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
107
pin dependence and localization to selective tubule stages. Endocrinology 1995; 136:5–12.
Eisenhauer KM, Chun SY, Billig H, Hsueh AJW. Growth hormone
suppression of apoptosis in preovulatory rat follicles and partial neutralization by insulin-like growth factor binding protein. Biol Reprod
1995; 53:13–20.
McGrane MM, deVente J, Yun J, Bloom J, Park E, Wynshaw A,
Wagner T, Rottman FM, Hanson RW. Tissue specific expression and
dietary regulation of a chimeric phosphoenolpyruvate carboxykinase/
bovine growth hormone gene in transgenic mice. J Biol Chem 1988;
263:11443–11451.
Bloom W, Fawcett DW. Textbook of Histology. Philadelphia: WB
Saunders; 1975: 858–906.
Palumbo A, Yeh J. In situ localization of apoptosis in the rat ovary
during follicular atresia. Biol Reprod 1994; 51:888–895.
Tilly JL, Billig H, Kowalski KI, Hsueh AJW. Epidermal growth factor
and basic fibroblast growth factor suppress the spontaneous onset of
apoptosis in cultured rat granulosa cells and follicles by a tyrosine
kinase-dependent mechanism. Mol Endocrinol 1992; 6:1942–1950.
McGee EA, Spears N, Minami S, Hsu SY, Chun SY, Billig H, Hsueh
AJW. Preantral ovarian follicles in serum-free culture: suppression of
apoptosis after activation of the cyclic guanosine 39,59-monophosphate
pathway and stimulation of growth and differentiation by follicle-stimulating hormone. Endocrinology 1997; 138:2417–2424.
Hirshfield AN. Size-frequency analysis of atresia in cycling rats. Biol
Reprod 1997; 38:1181–1188.
Chun SY, Eisenhauer KM, Minami S, Billig H, Perlas E, Hsueh AJW.
Hormonal regulation of apoptosis in early antral follicles: folliclestimulating hormone as a major survival factor. Endocrinology 1996;
137:1447–1456.
El-Roeiy A, Chen X, Roberts VJ, LeRoith D, Roberts CT Jr, Yen
SSC. Expression of insulin-like growth factor-I (IGF-I) and IGF-II and
the IGF-I, IGF-II, and insulin receptor genes and localization of the
gene products in the human ovary. J Clin Endocrinol Metab 1993;
77:1411–1418.
Spicer LJ, Alpizar E, Vernon RK. Insulin-like growth factor-I receptors in ovarian granulosa cells: effect of follicles size and hormones.
Mol Cell Endocrinol 1994; 102:69–76.
Jia X-C, Kalmijin J, Hsueh AJW. Growth hormone enhances folliclestimulating hormone-induced differentiation of cultured rat granulosa
cells. Endocrinology 1986; 118:1401–1409.
Davoren JB, Hsueh AJW. Growth hormone increases ovarian levels
of immunoreactive somatomedin C/insulin-like growth factor I in
vivo. Endocrinology 1986; 118:888–890.
Gong JG, Baxter G, Bramley TA, Webb R. Enhancement of ovarian
follicle development in heifers by treatment with recombinant bovine
somatotrophin: a dose-response study. J Reprod Fertil 1997; 110:91–
97.
Lamson G, Giudice LC, Rosenfeld RG. Insulin-like growth factor
binding proteins: structure and molecular relationships. Growth Factors 1991; 5:19–28.