J. Embryol. exp. Morph. 78, 33-42 (1983)
Printed in Great Britain © The Company of Biologists Limited 1983
33
Effect of exogenous sex steroids upon the number of
germ cells and the growth of foetal ovaries grafted
under the kidney capsule of adult ovariectomized
hamsters
By J. ARRAU 1 , L. ROBLERO, M. CURY 2 AND
R. GONZALEZ
From the Laboratorio de Endocrinologia, Instituto de Ciencias Biologicas,
Pontificia Universidad Catolica de Chile and the Departamento de Medicina
Experimental, Universidad de Chile (Sede Oriente)
SUMMARY
The effect of estradiol E, progesterone P and testosterone propionate T upon growth and
germ cell population of foetal ovaries transplanted under the kidney capsule of adult
ovariectomized hamsters was examined. Foetal ovaries were obtained 15 dayspostcoitum and
the host received daily subcutaneous injections of E (1 |Ug/day), P (5 mg/day), T (500 jug/day)
or E + P in 0-1 ml oil for 25 days beginning 5 days before grafting. One day after the last
injection, the size of the ovary, the germ cell population and plasma steroid levels were
assessed. The results of hormone assays indicate that daily steroid administration was able to
maintain continuously elevated plasma levels of the corresponding hormone. Interconversions or steroid secretion by the graft, if any, were not reflected in the peripheral circulation.
The growth of the graft was stimulated by T and inhibited by P, in comparison with E and oiltreated controls. All germ cells were at the stage of primary oocyte forming part of a primordial or growing follicle. Their absolute number was significantly increased by T and E and
significantly decreased by P and E + P. The number of oocytes per mm3 of ovary was increased
80 %, 48 % and 40 % with E, T and E + P, respectively.
It is concluded that, in the hamster, exogenous sex steroids given to the host can exert
specific effects upon the growth and oocyte population of a grafted foetal ovary. Whether or
not the action of steroids upon the graft is a direct one and whether they influenced oogonial
mitosis, the evolution of the meiotic prophase or atresia of primary oocytes remains to be
determined.
INTRODUCTION
In the majority of mammals studied so far, the onset of oogenesis is restricted
to the foetal period: human (Baker, 1963), rat (Beaumont & Mandl, 1962),
sheep (Mauleon, 1973), hamster (Arrau, Roblero & Cury, 1981) and only in
1
Author's address: Laboratorio de Endocrinologia, I.C.B., Pontificia Universidad Catdlica
de Chile, Casilla 114-D, Santiago, Chile.
2
Author's address: Departamento de Medicina Experimental, Universidad de Chile (Sede
Oriente), Santiago, Chile.
34
J. ARRAU, L. ROBLERO, M. CURY AND R. GONZALEZ
lemuridae primates (galagos, lorises) does it occur throughout adult life (Anand
Kumar, 1974). The control mechanisms involved in either one of these
modalities are incompletely understood and are probably multifactorial.
Another common feature of oogenesis is that the transformation of oogonia
into primary oocytes is associated with a dramatic reduction in the number of
germ cells. The factors governing this loss of germ cells are poorly understood.
Foetal ovaries of various species can accomplish in vitro several steps of the
steroid biosynthetic pathway: human (George & Wilson, 1978; Payne & Jaffe,
1974), rabbit (Milewich, George & Wilson, 1977; George, Simpson, Milewich
& Wilson, 1979), sheep (Mauleon, Bezard & Terqui, 1977), cow (Roberts &
Warren, 1964), and are able to secrete oestrogens and androgens in vivo: rabbit
(George, Milewich & Wilson, 1978), guinea pig (Rigaudiere, 1977). In addition,
oestradiol synthesis precedes or is concomitant with the meiotic prophase:
human (Baker, 1963; Payne & Jaffe, 1974), sheep (Mauleon, 1973; Mauleon et
al. 1977), a chronologic relationship that is compatible with the concept that
these steroids could play a role in regulating the onset of meiosis and that their
level could influence the size of the germinal cell population. This idea is supported by the observations of Anand Kumar (1968,1974) in lemuridae which show
that exogenous steroids can modify the ratio oogonia: primary oocytes. In
contrast with the above concept, Prepin, Vigier & Jost (1979) observed no
changes in the evolution of the meiotic prophase or in gonadal differentiation
when rat gonads were cultured in vitro in the presence of oestradiol, progesterone or the antioestrogen CI-128. Alternative interpretations of these results
are that the conditions attained in vitro were not suitable to demonstrate the
effect of steroids upon oogenesis or that their influence upon the gonad is not a
direct one.
The purpose of the present investigation was to examine the effects of sex
steroids upon the germ cell population in an in vivo situation utilizing an experimental model previously reported (Arrau & Roblero, 1979). In this model,
foetal hamster ovaries taken at the onset of meiosis (Arrau et al. 1981) are grafted
under the kidney capsule of an ovariectomized adult host. Although there is a
great loss of germ cells in the graft, those remaining attain the stage of primary
oocyte at the normal time. This report describes the effect of exogenous
oestradiol, progesterone or testosterone given to the host on the oocyte population and overall growth of the graft after 20 days of exposure to each hormone.
MATERIALS AND METHODS
Animals and treatment
Female hamsters were utilized at four months of age. They were kept under
controlled illumination (14 h light and 10 h darkness) and temperature (24 °C)
and with free access to water and pelleted food. Females were caged with males
Sex steroids and oogenesis
35
on the evening of pro-oestrous and the following day was considered day one of
pregnancy, if sperm were found in the vaginal smear. On the 15th day of pregnancy, the animals were killed to remove and graft the foetal ovaries, as
previously described (Arrau & Roblero, 1979).
Adult hamsters ovariectomized two months previously were used as recipients
of foetal ovaries. Under anaesthesia with Avertine TM, the left kidney of each
host was exposed and one foetal ovary was inserted under the kidney capsule.
Hormonal treatment of the hosts was initiated five days before grafting and
continued until the day before they were killed.
The animals receiving grafts were divided into five groups according to treatment, which consisted of a daily subcutaneous injection of 0-1 ml olive oil alone
or containing a steroid as follows:
-Group I (Controls); 17 females treated with the vehicle alone.
- Group II; 10 females treated with 1 /ig of oestradiol 17 beta (E), daily for 25
days.
- Group III; 11 females treated with 5 mg of progesterone (P) daily for 25 days.
-Group IV; 10 females treated with 500jug of testosterone propionate (T)
daily for 25 days.
- Group V; 10 females treated with 1 fig of E + 5 mg of P daily for 25 days.
In a previous study, it was established that all germ cells become arrested in
the dictyate stage and enclosed in a primordial follicle between 10 and 20 days
after grafting (Arrau & Roblero, 1979). Accordingly, day 20 after grafting was
chosen to examine the grafts. Each graft was processed for histologic study as
previously described (Arrau & Roblero, 1979) and was serially sectioned at
7/im and stained with the periodic acid Schiff (P.A.S.) Harris haematoxylin.
Germ cell count and estimation of graft volume
Germ cells were easily identifiable by their large size, neat round shape and
large vesicular nucleus. The mean ± S.E.M. diameter of the germ cell nucleus was
found to be 19 ± 0-6 fim (n = 300). All germ cell nuclei present in each 7 pim
section were counted, therefore, each nucleus was counted more than once. This
error was cancelled by dividing the total number of nuclei by 2-71 (this figure
results from dividing the mean diameter of nuclei by the section width).
In order to estimate the graft volume, the largest diameter of the section and
a second diameter at right angle with the first were measured in every fifth serial
section. These two diameters were multiplied by the thickness of the section and
the addition of these figures was multiplied by 5. Graft volume was expressed in
mm3.
Steroid levels in plasma
It was considered convenient to distinguish the circulating levels of sex steroids
attributable to the injected hormone from those due to endogenous production
or interconversions. For this purpose all grafted animals were bled by cardiac
36
J. ARRAU, L. ROBLERO, M. CURY AND R. GONZALEZ
puncture at the time of sacrifice (24h after the last injection). In addition five
groups, each of six ovariectomized females which received no graft, were given
daily injections as indicated for groups I through V. Half of the animals from
each group were sacrificed at 45 min and the other half at 24 h after the last
injection to collect blood for the assay of steroids.
Radioimmunoassay of steroids
Immediately after blood collection, the plasma was separated by centrifugation and stored at — 20°C until assayed. For the assay of testosterone, 50 to
500 /il of plasma were extracted with five volumes of ether after adding a tracer
amount of [3H]testosterone to correct for procedural losses. A specific antiserum
raised in sheep against testosterone-carboximethyloxime coupled to bovine
serum albumin, was used. Cross reaction was 14 % with 5 alpha-dihydrotestosterone and negligible with other androgens. The sensitivity of the assay was
below 2-4 pg per tube at a final dilution of the antiserum of 1: 210000. Interassay
coefficients of variation of middle (3-6ng/ml) and high (6-6ng/ml) pools were
28-6% and 32-2%, respectively. The radioimmunoassay of oestradiol and
progesterone have been previously described (Fuentealba, Vera & Nieto, 1982;
Villaldn ef a/. 1982).
Statistical analysis
All values were expressed as Mean ± S.E.M. The statistical significance of the
differences between the mean number of cells and mean ovarian volume was
evaluated by the test of Tukey (Glass & Stanley, 1974). Mean steroid concentration in plasma were analysed by Student's 't' test. Probabilities below 0-05 were
considered significant.
RESULTS
Two grafts in the control group and one in the progesterone-treated group did
not take and these were excluded from further analysis. The general appearance
of the grafts and some details of the morphology of the germ cells, follicular
development and stromal tissue were shown in Figs 1 and 2. Sections of normal
ovarian tissue of the same postcoital age are also shown for comparison. The
grafts appeared healthy, as judged from the degree of vascularization and the
presence of abundant mitosis among follicular cells. A greater density of follicles
characterized the sections obtained from animals treated with testosterone or
oestradiol, as compared to the other groups. Yet, in all grafted ovaries, the
number of follicles appeared greatly diminished, in comparison with non grafted
ovaries of the same age.
All germ cells observed corresponded to primary oocytes forming part of
a primordial or growing follicle. Sections from progesterone-treated animals
Sex steroids and oogenesis
37
Figures 1-3
Sections of hamster ovaries at the age 25 days postpartum (1); or its equivalent
following grafting under the kidney capsule of ovariectomized recipients (2); or
ovariectomized recipients treated with 1 jUg of estradiol 17 Beta per day for 25 days
(3). o: ovary; k: kidney. Bar = 200jum.
exhibited abundant stroma with the appearance of dense connective tissue.
Table I shows the effect of treatment with steroids upon the grafted foetal
ovary. In comparison with the vehicle-injected controls, the total number of
germ cells, was significantly increased in the oestradiol- and testosterone-treated
38
J. ARRAU, L. ROBLERO, M. CURY AND R. GONZALEZ
Table 1. Influence of sex steroids upon the number ofoocytes and volume of foetal
ovaries grafted under the kidney capsule of adult ovariectomized female hamster
Treatment
Control
(olive oil)
Oestradiol
1 /zg/day
Progesterone
5 mg/day
Testosterone
Propionate
500/ig/day
Oestradiol +
progesterone
1 jug + 5 mg/day
No of
animals
Total no. of
oocytes per ovary
(X±S.E.M.)
Volume of grafted
ovaries (mm3)
(X±S.E.M.)
Number of oocytes
per mm3 of ovarian
tissue ( X ± S . E . M . )
15
1265 ± 77
0-95 ± 0-03
1233 ± 63
10
2080 ± 100*f
1-00 ±0-08
2256±251*t
10
667 ±81*
0-66 ±009*
1195 ±212
10
3314 ± 137* t±§
l-91±0-04*tt§
1857 ± 176
10
339 ± 82*f±§
0-21 ± 0-04*n§
1754 ± 334
* Significantly different from control group.
fSignificantly different from progesterone.
^Significantly different from oestradiol.
§Significantly different from oestradiol + progesterone.
groups, and was significantly decreased in groups treated with progesterone and
oestradiol plus progesterone.
The ovarian graft volume was increased by testosterone treatment, it was
reduced by progesterone and by oestradiol plus progesterone treatment and it
was not changed by treatment with oestrogen alone.
Only in the control group was there a statistically significant positive correlation between the number of germ cells per graft and the size of the graft
(r = 0-635).
When the number of germ cells per unit volume of ovarian tissue was considered, it was found that oestradiol, testosterone and oestradiol plus progesterone increased their relative number by 80,48 and 40 %, respectively, over the
control group.
Grafted animals which received oil injections had very low levels of steroids
in plasma (Mean±S.E.M. = E: l-7±0-7pg/ml; P: 2-8±0-3ng/ml; T:
0-06 ± 0-0 ng/ml) and these were not higher than those of non-grafted oilinjected controls. The plasma level of the corresponding steroid was found
elevated 24 h after the last injection in hormone-treated animals. E, P and T
levels were elevated 100,2 and 10 times respectively over the values observed in
the oil-injected controls. In comparison with reported values for intact animals
(Saidapur & Greenwald, 1978) oestradiol was slightly increased, progesterone
was in the lower part of the normal range and testosterone was three times above
Sex steroids and oogenesis
39
faV. &>,
r.^: r^-.
Figures 4-7
Sections of hamster ovaries grafted under the kidney capsule of ovariectomized
recipients treated for 25 days with 5 fig of testosterone (4 & 5) or with 5 mg of
progesterone (6 & 7). Note the different density of follicle-enclosed oocytes (arrow)
and stromal tissue with the two treatments. Dense connective tissue bands (*) are
present in the stroma in the progesterone-treated group, o: ovary; kidney. Figs 4 &
6 Bar = 200 jum. Figs 5 & 7 Bar =
the upper limit. The mean levels at 45 min were 1-5- to 2-fold higher than at 24 h
following the last injection.
DISCUSSION
The data presented clearly indicate that sex steroids can exert a profound
influence in vivo upon the number of germ cells and the growth of foetal ovaries
40
J. ARRAU, L. ROBLERO, M. CURY AND R. GONZALEZ
and that these two parameters can be dissociated by shifting the balance between
these hormones. Each steroid exerted a specific effect upon the ovary.
Oestradiol increased the number of germ cells without changing the growth of
the ovary. Progesterone decreased both parameters while testosterone had the
opposite effect. The effects of testosterone may be partially explained by
aromatization taking place in the graft itself or in the host.
The results of the oocyte counts confirmed the previously reported enormous
loss of germ cells that occurs in the grafts (Arrau & Roblero, 1979). It also shows
that by and large all steroid treatments tested failed to prevent this loss of germ
cells. Temporary ischaemia and/or immunological rejection may be responsible
to a large extent for these losses and one could think that differential effects of
steroids on angiogenesis around or within the graft or on immunosuppression
may account for the significant differences in the number of oocytes encountered
with the various treatments. However, possible effects of steroids on
angiogenesis or on immunosuppression can hardly account for differential effects of steroids on germ cells and stroma. Therefore, we believe that the effects
observed represent the response of the graft components which survived the
ischaemia. The stages of oogenesis influenced by the treatments need further
investigation because changes in the number of oocytes can be attained by influencing multiplication of oogonia or the rate of atresia of oocytes.
A comparison of the observations by Anand Kumar (1968) in lemuridea with
the present observations in hamster can lead to a common explanation. In the
prosimian primate, the ovary contains normally oogonia and oocytes in a 1:1
ratio, therefore, it afforded the possibility of examining also the effects upon the
oogonia population. In this primate, oestradiol and progesterone did not change
and testosterone increased the number of oogonia in comparison with the oiltreated group. Regarding the number of oocytes, oestradiol and testosterone
increased and progesterone decreased their number in both species. A common
interpretation for these findings is that oestradiol prevents atresia, progesterone
facilitates atresia and testosterone stimulates oogonial mitosis and prevents
atresia.
The effects reported here may be the results of direct as well as indirect actions
of the steroids upon the ovarian graft. Prepin etal. (1979), could not find a direct
action of steroids upon the rat foetal ovary in vitro and Stumpf, Narbaits & Sar
(1980) reported the absence of oestrogen receptors in the foetal ovary of mice.
Indirect effects mediated through inhibition of pituitary gonadotrophic
secretion seem probable because the exogenous steroids reached levels in plasma
which would prevent the increase in gonadotrophic secretion that follows castration. This possibility cannot be excluded, since gonadal growth was maximally
inhibited by the combined treatment with oestradiol and progesterone, which is
known to be a potent inhibitor of gonadotrophin secretion. A direct effect of
pituitary gonadotrophins upon oogenesis in foetal gonads is not supported by
their lack of effect in vitro (human: Baker & Neal, 1974; hamster: Challoner,
Sex steroids and oogenesis
41
1975) and by the normal evolution of oogenesis in hypophysectomized sheep
foetuses (Mauleon, 1973) in anencephalic human foetuses (Baker & Scrimgeour,
1981) and in rats with hypophyseal agenesis caused by triparanol or AY 9944
(Giroud, Roux, Dupuis & Howath, 1980). However, taking into account possible interactions between steroids and gonadotrophins at the ovarian level
further experiments in hypophysectomized-ovariectomized hamster will be
required to clarify this issue.
Although the mechanism remains obscure, a clear cut effect of oestradiol,
progesterone and testosterone upon the germ cell population and growth of the
grafted foetal ovary was demonstrated in the hamster.
We wish to thank Dr H. B. Croxatto for helpful discussions; Mrs A. Brandeis and E. Nunez
for the steroid radioimmunoassays and Mrs F. Guzman and M. L. Alcoholado for skilful
technical assistance in preparing and assessing histological sections. This work was supported
by DIUC No. 82/81 and The Rockefeller Foundation Grant RF 80007.
REFERENCES
T. C. (1968). The effects of gonadal hormones on primordial germ cell
population in the slender loris. J. Reprod. Fertil. 16, 322-323.
ANAND KUMAR, T. C. (1974). Oogenesis in adult prosimian primates. In Contribution to
Primatology 3 (ed. W. P. Luckett), pp. 82-96. Karger, Basel.
ARRAU, J. & ROBLERO, L. (1979). A method for the in vivo experimental study of meiotic
prophase in the golden hamster. Archs Anat. microsc. Morph. exp. 68, (2) 73-84.
ARRAU, J., ROBLERO, L. & CURY, M. (1981). New observations on the onset and duration of
the meiotic prophase in the female golden hamster (Mesocricetus auratus). J. Anat. 132,
627-633.
BAKER, T. G. (1963). A quantitative and cytological study of germ cells in human ovaries.
Proc. R. Soc. Lond. B. 158, 417-433.
BAKER, T. G. & NEAL, P. (1974). Oogenesis in human fetal ovaries maintained in organ
culture. /. Anat. 117, 591-604.
BAKER, T. G. & SCRIMGEOUR, J. B. (1980). Development of the gonad in normal and
anencephalic human fetuses. /. Reprod. Fertil. 60, 193-199.
BEAUMONT, M. H. & MANDL, A. M. (1962). A quantitative and cytological study of oogonia
and oocytes in the fetal and neonatal rat. Proc. R. Soc. Lond. B 155, 557-559.
CHALLONER, S. (1975). Studies of oogenesis and follicular development in the golden hamster.
2. Initiation and control of meiosis in vitro. J. Anat. 119, 146-156.
FUENTEALBA, B., VERA, R. & NIETO, M. (1982). Changes in nuclear estrogen receptor level
in the rat oviduct during ovum transport. Biol. Reprod. 27, 12-16.
GEORGE, F. W. & WILSON, J. D. (1978). Conversion of androgens to estrogens by human fetal
ovary. J. clin. Endocr. Metab. 47, 550-555.
GEORGE, F. W., MILEWICH, L. & WILSON, J. D. (1978). Oestrogen content of the embryonic
rabbit ovary. Nature 274, 172-173.
GEORGE, F. W., SIMPSON, E. R., MILEWICH, L. & WILSON, J. D. (1979). Studies on the
regulation of the onset of steroid hormone biosynthesis in fetal rabbit gonads. Endocrinology 105, 1100-1106.
GIROUD, A., Roux, C., DUPUIS, R. & HOWATH, C. (1980). Ovarian meiotic prophase and the
fetal pituitary in the rat. Reprod. Nutr. Develop. 20, 681-685.
GLASS, G. V. & STANLEY, J. C. (1974). M6todos estadisticos aplicados a ciencias sociales.
Englewood Cliffs, N.Y.: Prentice-Hall International.
ANAND KUMAR,
42
J. ARRAU, L. ROBLERO, M. CURY AND R. GONZALEZ
P. (1973). Modification exp6rimentale de l'apparition et de 1'evolution de la
prophase m6iotique dans l'ovarie d'embryon de brebis. Ann. Biol. Anim. Bioch. Biophys.
13, 89-101.
MAULEON, P., BEZARD, J. & TERQUI, M. (1977). Very early and transient 17 beta estradiol
secretion by fetal sheep ovary: In vitro study. Ann. Biol. Anim. Bioch. Biophys. 17,
399-401.
MILEWICH, L., GEORGE, F. W. & WILSON, J. D. (1977). Estrogen formation by the ovary of
the rabbit embryo. Endocrinology 100, 187-196.
PAYNE, A. H. & JAFFE, R. B. (1974). Androgen formation from pregnenolona sulfate by the
human fetal ovary. /. clin. Endocr. Metab. 39, 300-304.
PREPIN, J., VIGIER, B. & JOST, A. (1979). Meiosis in fetal freemartin gonads and in rat fetal
ovaries in vitro. Ann. Biol. Anim. Bioch. Biophys. 19, 1263-1271.
RIGAUDIERE, N. (1977). Evolution des teneurs en testost6rone et dehydrostestoste"rone dans
le plasma, le testicule et l'ovarie chez le cobaye ou cours de la vie fetale. C. r. Acad. Sci.
(Paris) 285, 989-992.
ROBERTS, J. D. & WARREN, J. C. (1964). Steroid biosynthesis in the fetal ovary. Endocrinology
74, 846-852.
SAIDAPUR, K. S. & GREENWALD, G. S. (1978). Perispheral blood and ovarian levels of sex
steroids in the cyclic hamster. Biol. Reprod. 18, 401-408.
STUMPF, W. E., NARBAITS, R. & SAR, M. (1980). Estrogen receptors in the fetal mouse. /.
Steroid. Bioch. 12, 55-64.
VILLAL6N, M., ORTIZ,M. E., AGUAYO, C,MUNOZ, J. &CROXATTO,H. B. (1982). Differential
transport of fertilized and unfertilized ova in rat. Biol. Reprod. 26, 337-341.
MAULEON,
{Accepted 1 August 1983)