BIOLOGY OF REPRODUCTION 57, 1183-1192 (1997)
Ovarian Morphology and Endocrine Characteristics of Female Sheep Fetuses That
Are Heterozygous or Homozygous for the Inverdale Prolificacy Gene (fecX'
P. Smith,2 W-S. 0,3 K.A. Corrigan,2 T. Smith, 2 T. Lundy,2 G.H. Davis, 4 and K.P. McNatty' 2
Wallaceville Animal Research Centre, 2 AgResearch, Upper Hutt, New Zealand
Department of Anatomy,3 University of Hong Kong, Hong Kong, Peoples Republic of China
Invermay Agricultural Centre, 4 AgResearch, Private Bag, Mosgiel, New Zealand
ABSTRACT
The Inverdale gene (fecX), located on the X chromosome, is
a major gene affecting the ovulation rate of sheep. At each ovulation, ewes heterozygous (1+) for the fecX' gene ovulate, on
average, one more egg than noncarriers (++), whereas ewes
that are homozygous (II) for this gene are infertile and have
"streak" ovaries. Since formation of the ovary occurs in fetal
life, it is possible that the fecX' gene influences ovarian development before birth. The aims of this study were to examine the
effects of the fecX' gene on germ cell development, follicular
formation and growth, and plasma gonadotropin concentrations
at 5 different days of gestation (i.e., Days 40, 90, 105, 120, and
135) and also in adult life.
The results suggest that one copy of the X-linked mutation in
female fetuses leads to a retardation of germ cell development
at Days 40 and 90 of gestation. However, from Day 105 of gestation, follicular formation and growth appear normal. By contrast, in females with two copies of the X-linked mutation, germ
cell development and follicular formation appear normal, but
thereafter follicular growth from the primary stage of development is impaired. During fetal life the plasma concentrations of
FSH and LH, although not measurable at Day 40, were similar
between all the genotypes at Day 105, 120, and 135 of gestation. The only exception was for LH at Day 90 in the + and II
animals: in ewes with these genotypes the plasma concentrations of LH were similar but significantly lower (p < 0.01) than
in the + + genotype. In adult animals the plasma concentrations
of FSH and LH were not different between the ++ and I+ genotypes, reflecting similar levels of ovarian follicular activity.
However, in adult II animals, the plasma concentrations of FSH
and LH were significantly higher (both p < 0.01) than in the
++ and I+ genotypes, reflecting the absence of normal secondary and antral follicles.
In summary, these data show that the fecX' gene affects ovarian development before birth and that the nature of the effect
is influenced by whether the female fetus is a homozygous or
heterozygous carrier of the X-linked mutation.
INTRODUCTION
The Inverdale gene (fecX), located on the X chromosome, is a major gene affecting the ovulation rate of Romney sheep [1]. At each ovulation, ewes heterozygous (I+)
for the fecXI gene ovulate, on average, one more egg than
noncarriers (++), whereas ewes that are homozygous (II)
for this gene are infertile and have "streak" ovaries [2]. In
a previous study [3], the streak ovaries of II ewes were
shown to have no normal follicles beyond the primary stage
of development. Moreover, the streak ovaries displayed no
evidence of functional activity, since the plasma concentraAccepted July 7, 1997.
Received February 19, 1997.
'Correspondence: K.P. McNatty, Wallaceville Animal Research Centre,
PO Box 44063, Upper Hutt, New Zealand. FAX: 00-64-4-528-1380;
e-mail: [email protected]
1183
tions of estradiol and immunoreactive inhibin were undetectable and those of FSH and LH were comparable to
those in ovariectomized ewes. In contrast, in I+ ewes, the
plasma concentrations of inhibin, estradiol, and gonadotropins were similar to those of control ewes [4].
Gonadal dysgenesis associated with the X chromosome
has previously been reported in humans with Turner's syndrome (XO) [5]. Likewise there are reports of a premature
ovarian failure syndrome in XO mice and humans that is
transmitted in an X-linked manner [6, 7]. However, to our
knowledge, there are no reports other than for Inverdale
sheep of an X-linked gene that is expressed as an increase
in ovulation rate in female heterozygotes but as gonadal
dysgenesis in homozygotes.
In sheep, gonadal sex differentiation begins after Day 30
of gestation, and by Day 40 the ovary is easily distinguished from the testis [8, 9]. At Day 40, all the germ cells
are present in the ovary as mitotically active oogonia, cells
from the mesonephros are migrating into the developing
ovary, and the plasma and pituitary concentrations of FSH
and LH are undetectable [9, 10]. By Day 90 of gestation,
the fetal ovary contains germ cells that are present either
as oogonia, degenerating germ cells, isolated oocytes, or
primordial follicles [9]. Around Day 100, the first primary
follicles (i.e., with one complete layer of cuboidal granulosa
cells) can be observed, and by Day 120, approximately
19% of the germ cell population may be present in follicles
with up to three concentric layers of granulosa cells. By
Day 135 of gestation, about 90% of the germ cells are in
primordial follicles and 4% in growing follicles that may
develop to between 0.25 and 0.80 mm in diameter, while
the remainder are present as isolated oocytes [9, 11]. Plasma concentrations of LH and FSH in fetal life increase after
Day 55 to peak values around Days 90-100; thereafter, they
decline to low values before birth [11]. Although concentrations of LH and FSH are high when the growth of primordial follicles is initiated, the role of gonadotropins in
early follicular development remains obscure [12]. Nevertheless, since formation of the ovary occurs in fetal life, it
is reasonable to suppose that the Inverdale X-linked mutation may affect some aspect of ovarian development and/
or function before birth. Since gonadotropin concentrations
in the adult II genotype differ from those in I+ and + +
ewes, it is possible that such differences might also occur
in fetal life.
The aims of the present study were to examine germ cell
and follicular development as well as pituitary hormone
concentrations in putative II/I+ carriers of the Inverdale
gene at Days 40, 90, 105, 120, and 135 of gestation and to
compare these findings with those for the controls (i.e., + +
and the known I+ genotype at the same gestational ages).
The results obtained for fetuses were also compared with
those for adult ewes.
1184
SMITH ET AL.
MATERIALS AND METHODS
The experimental procedures reported in this study were
carried out in accordance with the 1987 Animal Protection
(Codes of Ethical Conduct) Regulations of New Zealand
after approval was granted by the AgResearch Animal
Ethics Committees of Invermay and Wallaceville.
Animals and Procedures
II or I+ ewes were each obtained from a flock in which
progeny-tested rams had been mated with I+ or + + daughters of progeny-tested rams [2]. Adult nonpregnant ewes
(aged 1.5 yr) of the II (n = 5) or I+ genotype (n = 6) as
identified by laparoscopy [1, 2], together with nonpregnant
age-matched control ewes (n = 5), were slaughtered. The
ovaries were recovered and dissected free of extraneous
tissue. Immediately before slaughter, blood samples were
collected and centrifuged at 4000 x g for 10 min at 20 0C,
and the plasma was stored at -15°C until assayed for FSH
and LH by RIA. Both ovaries were fixed in Bouin's fluid,
processed, and serially sectioned at 5 Im for morphometric
and morphological studies.
Female fetuses were recovered from pregnant ewes at
Days 40, 90, 105, 120, and 135 postmating. Matings of
noncarrier (++) rams (n = 5) with ++ ewes (n = 54)
gave female fetuses (n = 63) that were noncarriers (++)
of the (fecX) gene. Matings of carrier (I) rams (n = 3) with
+ + ewes (n = 44) gave I+ female fetuses (n = 48). These
fetuses were recovered after a barbiturate overdose (i.e.,
sodium pentobarbitone; 20 ml i.v. of a 500 mg/ml solution
w:v) to the mother. Matings of I rams (n = 5) to I+ ewes
(n = 49) produced II or I+ female fetuses (n = 69), and
these were recovered by cesarian section or after a barbiturate overdose. At the time of this study, biochemical or
DNA markers to distinguish the putative (p) II from pI+
fetuses were not available.
Fetuses obtained by the two procedures described were
blood-sampled by cardiac puncture; the blood was centrifuged at 40C and the plasma stored at - 15°C until assayed
for FSH and LH by RIA. The fetal weights and crownrump lengths were measured, and the ovaries, mesonephros
(at Day 40 only), adrenals, and pituitaries were dissected
and weighed. The left ovaries from 40- and 90-day-old fetuses and from adults were fixed in Bouin's fluid and embedded in paraffin. Fetal ovaries were serially sectioned at
a thickness of 4 im, whereas those from the adults were
sectioned at a thickness of 5 im. The left ovaries from the
105-, 120-, and 135-day-old fetuses were embedded in plastic (Technovit 7100; Kulzer & Co., GmBH, Wehrheim,
Germany) and serially sectioned at a thickness of 30 pm.
The purpose of using these thick sections was to test whether the follicular-like structures present at Days 105, 120,
and 135 of gestation were distinct entities and separate from
the adjacent interstitium and intraovarian rete via a basement membrane. The adrenals and pituitaries were frozen
and used in an unrelated study.
Morphometric Studies
Ovarian volume was estimated by the Cavalieri principal
described by Smith et al. [9]. Briefly, the area of tissue was
estimated by point counting on every 10th (i.e., fetal ovaries) or 100th (i.e., adult ovaries) section, and the sum of
areas was multiplied by the distance between the sections
to yield the ovarian volume. These volumes were adjusted
to account for differences in shrinkage between those ova-
ries embedded in paraffin and those embedded in plastic.
From parallel studies on fetal ovaries of the same weights
and age, the degree of shrinkage in paraffin tissue was 1.96
times greater than that of equivalent ovaries embedded in
plastic. The ovarian volumes for the 40- and 90-day-old
fetuses and adults were adjusted to the shrinkage values in
plastic to facilitate comparisons between all ages. Germ cell
numbers were counted from about 10 pairs of serial sections for the 40- and 90-day-old fetal and adult ovaries
using the nuclear dissector method [13] as described by
Smith et al. [9]. The nucleus or chromatin was used as a
means of identification of oocytes. At Days 105, 120, and
135, germ cell numbers were counted using the optical dissector as described elsewhere [11]. The nuclear and optical
dissector methodologies are based on similar principles,
namely a 3-dimensional approach to cell quantification and
a systematic sampling procedure that begins from a random
start position. The nuclear dissector utilizes a number of
pairs of adjacent serial sections, namely the 4-m or 5-Lm
paraffin sections in this study. A germ cell nucleus or chromatin was counted if it was apparent in the second of the
adjacent sections but not in the first section, providing that
the nucleus or chromatin was within an unbiased counting
frame. The resulting number (designated Q) is the number
of germ cells in the volume equal to the area of the counting frame multiplied by the height of the dissector (i.e., 4
or 5 m). The optical dissector was used on the 30-pm
plastic sections. By means of focusing within the section
through a known depth, generally between 10 and 20 Vpm,
the germ cell nucleus or chromatin is counted as it comes
into focus within the unbiased counting frame. The resulting number of germ cells (Q) represents the number in the
volume derived from the area of the counting frame multiplied by the depth of the dissector (i.e., 10-20 Im). In
this context, the optical dissector when focusing through 16
tm within a section is effectively taking a series of four
4-Vpm nuclear dissectors. A comparison of the two methods
for estimating germ cell numbers in fetal sheep has been
performed using contralateral ovaries. From these studies,
the overall mean values at each gestational age differed
between the two methods by < 6%. During the estimation
of total numbers, the germ cells were classified into oogonia, oocytes (i.e., germ cell in meiosis but not surrounded
by follicular cells), primordial follicles (i.e., < 1 complete
layer of cuboidal granulosa cells), primary follicles (i.e.,
between 1 and < 2 complete layers of granulosa cells), and
larger follicles (i.e., secondary to antral follicles). The diameters of the germ cells and primordial follicles were also
measured. The oogonia, oocytes, and follicles to be measured were chosen randomly with a minimum of 20 counted
from each animal. For oogonia, oocytes, and follicles, the
sections containing the germ cell nucleolus were measured;
for meiotic oocytes, the sections containing the nuclear
chromatin were selected for measurement. Two measurements of each diameter were made at right angles to each
other and averaged using the microscale TC image analysis
system (Digihurst Ltd, Royston, Herts., UK).
For studies on the relationships between the number of
granulosa cells and oocyte diameter, the following procedures were used. From plastic sections (i.e., 30 m thick)
of fetal ovaries at Days 120 and 135 of gestation, oocytes
were visualized in the optical plane containing the nucleolus, and their diameters were recorded as described above,
as was the total number of granulosa cells around the circumference in the same optical plane. A minimum of 8
granulosa cell-oocyte complexes were measured for each
OVARY AND ENDOCRINOLOGY OF INVERDALE FETUSES
1185
TABLE 1. Effect of the Inverdale gene on crown-rump length, mesonephros weight, ovarian volume, and number and diameter of germ cells per ovary
in fetal sheep at Day 40 of gestation.
Fetal
characteristic
Crown-rump length (cm)
Mesonephros weight (mg)
Ovarian volume (mm')
Total number of
germ cells x10 3
Diameter of germ cells
(oogonia, m)
Number of female fetuses
Inverdale genotype*
++
I+
pl+
pll
3.8a
(3.5, 4.1)
17.0'
(15.0, 18.0)
1.56'
(1.22, 1.96)
4.0'
(3.8, 4.2)
18.0 a
(16.0, 19.0)
1.28 a
(1.11, 1.46)
3.8a
(3.4, 4.2)
14.0'
(10.0, 17.0)
1.32'
(1.12, 1.54)
4.1 a
(3.4, 4.9)
16.0 a
(14.0, 18.0)
1.32a
(1.14, 1.51)
47a
24b
27
(38, 58)
14.4'
(13.6, 15.3)
8
(19, 31)
14.9'
(13.3, 16.7)
11
b
(20, 36)
14.4a
(13.5, 15.4)
4
45a
(37, 55)
14.4a
(13.6, 15.3)
5
* Values are geometric means (and 95% confidence limits). pl+, pll refer to putative I+ and putative II, respectively.
1,bFor each row, values with different alphabetic superscripts are significantly different. a vs. b = p < 0.05 (Duncan's Multiple Range test).
ovary, and the results for each genotype (i.e., + +, I+, pI+,
and pII) were analyzed separately.
Hormone Assays
FSH. The FSH RIA kit was supplied by the National
Institute of Arthritis, Metabolism and Digestive Disease
(NIAMDD), Bethesda, MD. The ovine (o) FSH for iodination was NIAMDD-oFSH-I-1; the oFSH reference preparation was NIAMDD-oFSH-RP1 (biopotency 75 x NIHFSH-S1); and the oFSH antiserum was NIAMDD-antioFSH-1 (AFP-C5288113). The volume of plasma used in
the assay was 0.1 ml, and each dilution was assayed in
duplicate. The internal standards, standard curve samples,
and serially diluted extracts were prepared in FSH-free hypophysectomized ewe plasma, and the minimum detectable
concentration was 0.2 ng/ml. The intra- and interassay coefficients of variation (CV) for the internal standards were
both < 8%.
LH. The LH RIA was identical to that described previously by McNatty et al. [14]. Briefly, the iodination standard was NIDDK-oLH-I-3 (AFP-9598B); the LH antiserum
was raised at Wallaceville and is described elsewhere [14].
The oLH reference preparation was NIAMDD-oLH-23
(biopotency 2.3 x NIH-LH-S1). The volume of plasma that
was assayed in duplicate was 0.1 ml. The internal standards, standard curve samples, and serially diluted extracts
were prepared in LH-free hypophysectomized ewe plasma.
The overall intra- and interassay CV for the internal standards that were included with every standard curve estimation were both < 10%. The minimum detectable concentration of LH was 0.2 ng/ml.
Statistical Procedures
Data from individual adult ewes or fetuses were logtransformed and subjected to analyses of variance to test
for differences between genotypes. Fetuses tentatively classified as pI+ were analyzed separately from the known I+,
pII, and ++ animals.
Partial correlation analyses controlled for genotype were
performed to determine the relationships between the number of germ cells and the diameters of the oogonia or oocytes.
The data for oocyte diameter and number of granulosa
cells were curve-fitted using the SPSS (Chicago, IL) statistical package. The data for each genotype separately could
be described by the exponential equation Y = Aecx, where
Y = the number of granulosa cells in the largest cross sec-
tion, X = the diameter of the oocyte, A = the intercept,
and C = the relative rate of increase of the number of
granulosa cells compared to that of the diameter of the oocyte. The constants A and C were subjected to standard ttests to examine for differences between genotypes.
RESULTS
Morphology and Morphometry of Fetal Ovaries
At Day 40 of gestation. It was not possible to distinguish
the ++, known I+, or pII/I+ genotypes by crown-rump
length, mesonephros weight, adrenal weight (data not
shown), or ovarian volume (Table 1). Irrespective of genotype, the fetal ovaries at Day 40 and thereafter had welldefined cortical and medullary regions. At Day 40, the
germ cells were mainly localized in cords in the cortical
region. Ovaries from all genotypes contained germ cells at
the oogonia stage and were similar morphologically.
The mean number of germ cells per ovary in the + +
genotype was significantly greater than in the authentic I+
animals (++ > I+, p < 0.05). From the II/I+ group, in
five of the ovaries the numbers of germ cells per ovary
were similar to those for the + + group, while in the other
four ovaries the numbers of germ cells were similar to those
for known I+ genotype (Table 1). Thus the former were
tentatively assigned as pII and the latter as pI+ (i.e., those
similar to the known I+ genotype). There were no significant effects of Inverdale genotype on the mean diameter
of the oogonia (Table 1).
At Day 90 of gestation. There were no obvious differences between the + +, known I+, and pII/I+ genotypes
with respect to crown-rump length (Table 2), adrenal
weight, or pituitary weight (data not shown). One set of the
II/I+ ovaries had ovarian volumes similar to those in the
known I+ group; these were classified as pI+. The other
set of II/I+ ovaries differed from those of the known I+
group (p < 0.05) but were similar to those of the + +
genotype; these were classified as pII (Table 2). Ovaries at
Day 90 of development had a well-defined cortex and medulla, and the germ cells in the cortex were present in the
form of oogonia, nests of oocytes in meiotic prophase, and
primordial follicles that contained diplotene stage oocytes
enclosed in a squamous layer of follicular cells. When the
total populations of germ cells in the pI+ ovaries were
compared with those in the pII ovaries, they were separable.
In the pI+ fetuses, the germ cell populations were similar
to those of the known I+, whereas in the pII fetuses, the
germ cell populations were similar to those in the + + ge-
1186
SMITH ET AL.
TABLE 2. Effect of the Inverdale gene on crown-rump length, mesonephros weight, ovarian volume, and number and diameter of germ cells per ovary
in fetal sheep at Day 90 of gestation.
Inverdale genotype*
Fetal
characteristic
++
Crown-rump length (cm)
23.0"
(22.4, 23.6)
13.0"
(11.3, 15.1)
150'
(107, 213)
2"
(1, 14)
29
(18, 48)
74',
(46, 118)
33"
(26, 42)
77.6'
(71.6, 83.8)
14
3
Ovarian volume (mm )
Total number of germ cells x10'
Atretic germ cells x10
Number of oogonia x10 1
'Number of oocytes x10
2Number of primordial
x10 follicles
' 2Meiotic germ cells
(% of total germ cells)
Number of female fetuses
23.9"
(23.3, 24.6)
17.6',
(14.4, 19.2)
3061'
(211, 442)
24'
(13, 40)
42,'
(21, 82)
168 '
(100, 280)
36"
(21, 61)
73.8'
(67.2, 80.6)
12
pl+
p+
pll
24.3.'
(23.0, 25.2)
18.5b
(15.7, 21.8)
268')
(190, 378)
21'
(14, 30)
22'
(14, 35)
1761'
(106, 293)
21'
(8, 53)
80.9
(76.7, 85.3)
9
23.7'
(20.2, 27.7)
11.8'
(9.5, 14.6)
102,'
(71, 148)
8.'
(4, 16)
2"'
(0, 67)
75.'
(39, 141)
25"'
(20, 30)
85.3
(56.6, 95.0)
6
*Values are geometric means (and 95% confidence limits). pi+, pll refer to putative I+ and putative II, respectively.
,b For each row, values with different alphabetic superscripts are significantly different. a vs. b = p < 0.05 (Duncan's Multiple Range test).
notype (Table 2). Moreover, the pII group had significantly
lower numbers of degenerating (i.e., atretic) germ cells, oogonia, and oocytes compared to the pI+ group. At Day 90,
the mean diameters of the isolated oocytes and oocytes enclosed by follicle cells, but not oogonia, were significantly
larger in the pII and + + groups as compared to the I+ and
pI+ groups (Table 3). When controlled for genotype, significant partial correlations were noted between the number
of germ cells and the diameter of isolated oocytes (R =
-0.47, p < 0.05) as well as between the number of germ
cells and the diameter of oocytes enclosed by follicle cells
(R = -0.48, p < 0.05), but not between the number of
germ cells and the diameter of oogonia.
At Day 105 of gestation. There were no obvious differences between the ++, known I+, and pII/I+ genotypes
with respect to crown-rump length, ovarian volume, adrenal
weight, or pituitary weight (data not shown). When the
morphology of the ovaries of the pI+/II genotypes was examined in detail (Fig. 1, a-c), one subset of ovaries contained a high frequency of abnormally large oocytes (Fig.
Ic) and some follicular-like structures that were devoid of
oocytes as previously observed in ovaries in adult II sheep
TABLE 3. Effect of the Inverdale gene on the diameter (eim) of the germ
cells and primordial follicles in the ovaries of fetal sheep at Day 90 of
gestation.
Germ cell or
follicle type
Oogonia
Isolated oocytes
++
Inverdale genotype*
I+
pl+
pl1
14.8'
14.2"
14.1"'
14.4"
(13.5, 15.4) (13.6, 14.6) (13.4, 15.1) (13.8, 15.8)
20.5'
17.91,
18.01
19.8
27.2'
25.81
24.81
27.3.'
Follicle enclosed
(19.6, 21.3) (17.2, 18.5)
(17.3, 18.7) (18.3, 21.5)
'
oocytes
Primordial
follicles
(24.0, 25.6) (26.7, 27.9)
(26.2, 28.3) (25.4, 26.3)
34.9"
30.9 '
31.91
34.3"
(34.5, 35.4) (29.9, 31.9) (31.3, 32.2) (32.4, 36.3)
Number of female
fetuses
14
12
9
6
* Values are geometric means (and 95% confidence limits). pl+, pll refer
to putative I+ and putative II respectively.
,bFor each row, values with different alphabetic superscripts are significantly different from one another. a vs. b = p < 0.05 (Duncan's Multiple
Range test).
[3]. The oocyte-free follicles at this and later gestational
ages were noted to be discrete entities and to be separated
from the surrounding interstitial tissue and rete ovarii by a
basement membrane. In the present study, oocyte-free follicles were rarely observed in the known I+ or ++ genotype. Accordingly, those animals having ovaries with a high
frequency of abnormally large oocytes in primordial and
primary follicles with abnormal granulosa cells were tentatively classified as pII, whereas those containing a low
frequency of abnormal oocytes in both primordial and primary follicles were classified as pI+. In the pII genotype,
38% of the primordial follicle population and 47% of the
primary follicle population contained enlarged oocytes with
diameters up to 80 Gum. In addition, 75% of the primary
follicles in the pII genotype contained an abnormal arrangement of granulosa cells (Fig. 1). By contrast, in the putative
I+ genotype, 15% of the primordial follicle population and
5% of the primary follicle population contained enlarged
oocytes with diameters up to 55 pLm. Importantly, there was
no evidence in the pI+, I+, and control genotypes for abnormal arrangements of granulosa cells in the primary follicles. When the p+ and pII genotypes were examined
with respect to crown-rump length, ovarian volume, total
number of germ cells, and number of oocytes, primordial
follicles, or primary follicles, no significant differences
were noted. The overall geometric means (and 95% confidence limits) irrespective of genotype for the aforementioned fetal or ovary characteristics are summarized in Table 4.
When the effects of the Inverdale gene on oocyte or
follicular diameter were examined with respect to genotype,
no differences were noted. The overall geometric mean
(and 95% confidence limits) diameters were: oocytes, 27.1
(25.0, 29.3) Lm; oocytes in primordial follicles, 34.3 (33.4,
35.2) Aim; primordial follicles, 43.3 (42.1, 44.5) jIm; oocytes in primary follicles, 49.1 (47.8, 51.4) u.m; and primary follicles, 72.2 (70.3, 73.3) Ipm (n = 34 animals).
At Day 120 of gestation. As occurred with the Day 105
fetuses, the pII and I+ fetuses were distinguishable by
ovarian morphology (Fig. 1, d-f). In the pII ovaries, 10%
of primordial follicles had enlarged oocytes (i.e., up to 78
pxm in diameter); 16% of primary follicles had enlarged
OVARY AND ENDOCRINOLOGY OF INVERDALE FETUSES
1187
FIG. 1. Sections through Inverdale ovaries. At Day 105 of gestation, normal primary follicle growth was evident in the ++ genotype (a) and I+
genotype (b), but enlarged oocytes with abnormal somatic cell arrangements around the oocyte were evident in the II genotype (c). At Day 120 of
gestation, normal primary to secondary follicle growth was evident in the ++ genotype (d) and I+ genotype (e), but abnormal development was evident
in the II genotype (f). At Day 135 of gestation, normal secondary follicle growth was evident in the + + genotype (g) and I+ genotype (h), but enlarged
oocytes with abnormal somatic cell arrangements found in the oocyte were evident in the II genotype (i). In adults, normal antral follicle development
was evident in the ++ genotype (j) and I+ genotype (k), but enlarged oocytes with abnormal somatic cell arrangements were still evident in the II
genotype (I). For a-f, i, I, bar = 50 ipm; for g and h, bar = 75 Lm; for j and k, bar = 150 Lm.
oocytes (i.e., up to 80 jim), and over 90% of these contained abnormal arrangements of granulosa cells (Fig. If).
By contrast, in the pI+ ovaries, 5% of the primordial and
5% of the primary follicles contained abnormally large oocytes (i.e., up to 60 jm). However, there was no evidence
of abnormal arrangements of granulosa cells in the primary
follicle populations of p+, control, or I+ ovaries. Also in
pII ovaries, appreciable numbers of isolated oocytes (i.e.,
having no surrounding granulosa cells) were located in the
medullary region of the ovary (the numbers of these were
SMITH ET AL.
1188
TABLE 4. Effect of day of gestation on crown-rump length, ovarian volume, the number of germ cells, and the number of isolated oocytes, primordial follicles, and primary and secondary follicles in female Inverdale
fetuses irrespective of genotype.
Fetal
characteristic
Crown-rump
length (cm)
3
Ovarian volume (mm )
Total number of
germ cells (xl0 3)
Number of oocytes
Number of primordial
follicles (x10 )
Number of primary
follicles
Number of secondary
follicles
Number of female
fetuses
105
Day of gestation*
120
29.4a
(28.5, 30.3)
17.9a
(16.6, 19.2)
202;'
(169, 241)
10,964'
(4265, 27,925)
168a
(144, 196)
9'
(2, 38)
0'
34
b
3 5 .7
(35.0, 36.4)
b
20.9
(19.2, 22.9)
1561'
(131, 185)
796
(249, 2511)
143'
(118, 172)
110 b
(33, 354)
2'
(0, 3)
39
135
39.0'
(37.9, 39.7)
25.3'
(22.6, 28.4)
b
142
(114, 177)
39'
(8, 167)
135'
(108,169)
1401
(37, 524)
15 '
(5, 46)
37
* Values are geometric means (and 95% confidence limits). The ovarian
data are those for the left ovary. At each age, n is 11-16 for the ++
genotype, 8-11 for the I+ genotype, 7-9 for the putative I+ genotype,
and 6-8 for the putative II genotype.
~,bFor each row, values with different alphabetical superscripts are significantly different. a vs. b = p < 0.05 (Duncan's Multiple Range test).
t Excludes the pll genotype.
not determined). These "lost" oocytes were rarely seen in
control or I+ ovaries and were not obvious in the pII genotype at Day 105 of gestation. Using these criteria, no
differences were noted between ++, known I+, pI+, and
pII with respect to fetal weight, crown-rump length, adrenal
weight or pituitary weight, ovarian volume, total number
of germ cells, or numbers of isolated oocytes or primordial
or primary follicles. The overall geometric means (and 95%
confidence limits) for the aforementioned fetal characteristics are summarized in Table 4. Over all genotypes at Day
120 of gestation, there was a significant increase in crownrump length (p < 0.05), ovarian volume (p < 0.05), and
the numbers of primary follicles (p < 0.05) together with
a significant fall in the numbers of germ cells and oocytes
(both p < 0.05) in comparison to those characteristics at
Day 105 of gestation (Table 4). In some but not all of the
+ +, known I+, and pI+ animals, a few secondary follicles
were noted (Table 4), but these were not observed in any
of the II animals.
No effect of the Inverdale gene was noted for the mean
diameters of oocytes enclosed in primordial or primary follicles or for the diameters of primordial or primary follicles
themselves; these mean diameters were not different from
those described at 105 days of gestation. In the pII genotype, follicles devoid of oocytes, or follicles containing abnormally large oocytes, were excluded from the aforementioned analyses. The mean diameters of the oocytes in secondary follicles or of the secondary follicles in the + +,
known I+, or pI+ were not different from one another;
overall the geometric mean (and 95% confidence limit) diameter for the oocyte in secondary follicles was 66.4 (63.2,
68.3) tpm and for secondary follicles was 108 (102, 114)
pxm (n = 31 ewes).
At Day 135 of gestation. At Day 135, the ovaries of the
pII and pI+ genotypes were clearly distinguishable on the
basis that one subset contained follicles undergoing abnor-
mal development: this subset was assigned the pII genotype
(Fig. 1, g-i). In the pII genotype, enlarged oocytes up to 85
[pm were noted in 18% of the primordial follicles. Likewise,
enlarged oocytes (i.e., up to 85 pIm) were noted in 22% of
the primary follicles, with 95% of these containing an abnormal arrangement of granulosa cells (Fig. li). By contrast
in the pI+ genotype, enlarged oocytes up to 55 pxm were
noted in 9% of primordial follicles and in 4% of primordial
follicles. As reported for ovaries at Day 120 of gestation,
there was no evidence of abnormal arrangements of granulosa cells in the primary follicle populations of pI+, I+,
and control animals. The pI+ but not pII ovaries also contained some normal-looking secondary follicles, as was the
case with the known I+ and + + genotype (Fig. h). Some
follicles with two or more complete layers of granulosa
cells were observed in the pII, but the arrangement of the
granulosa cells was always disorganized (Fig. 2). Indeed,
follicles of the type observed in Figure 2b were not noted
in any other genotype. These follicles were regarded as
abnormal and were not included in the data reported in
Table 2. In addition to abnormal secondary follicle development in the pII ovaries, there was evidence for appreciable numbers of oocyte-free follicles, oocytes, and unusual
clusters of interstitial-like cells in the medullary region of
the ovary (numbers were not determined). The interstitiallike cells in the medulla were often intermingled with a
conspicuous extracellular matrix and collagen network. The
oocytes and cell masses in the medullary region were not
part of the rete system and were not observed in either
control or I+ ovaries. No differences with respect to genotype were noted for any of the parameters measured.
These values averaged over all genotypes with respect to
crown-rump length, ovarian volume, total number of germ
cells, number of oocytes, and number of primordial, primary, and secondary follicles are summarized in Table 4.
No genotype differences were noted for adrenal and pituitary weight or for the diameter of the oocytes in these follicles (data not shown). When all the genotype data were
pooled, there was a significant increase in mean crownrump length (p < 0.05), ovarian volume (p < 0.05), and
mean numbers of secondary follicles (p < 0.05) but a significant decline in the mean number of isolated oocytes (p
< 0.05) as compared to those characteristics at Days 105
and 120 of gestation (Table 4).
Morphology and Morphometry of Ovaries from Adult
Sheep
The results from the morphometric studies of the adult
ovaries from + +/I+ and II ewes are summarized in Table
5. The mean ovarian volume of the ovaries in the II animals
was significantly smaller than that in either the I+ or + +
genotype (p < 0.01). The mean total number of germ cells,
and mean numbers of primordial and primary follicles per
ovary in the II animals, were not different from those observed in the I+ and ++ ewes. However, the II ovaries
showed no evidence of normal follicular growth beyond the
primary stage of development (e.g., compare Fig. 1, j and
k with Fig. 11); virtually all the germ cells were enclosed
in primordial or primary follicles within the ovarian cortex.
In addition, some follicles in the II ovaries were present as
abnormal primary structures, with follicular cells not concentrically aligned with the oocyte and with a poorly developed zona pellucida (Fig. 11). Numerous nodules (oo-
OVARY AND ENDOCRINOLOGY OF INVERDALE FETUSES
1189
FIG. 2. Sections through Inverdale ovaries: a normal secondary follicle in the
++ genotype (a) and an abnormal follicle
in the II genotype showing an enlarged
oocyte and irregular organization of granulosa cells (b). For both figures, bar = 50
Jim.
cyte-free follicles) were present, and some follicular structures contained enlarged oocytes but only one layer of granulosa cells. No corpora lutea or corpora albicans could be
found in any of the sections from II ovaries.
significantly different (p < 0.05 in all instances) from those
for the + +, I+, and pI+ genotypes; the constants for the
+ +, I+, and pI+ genotypes were not different from one
another.
Relationship Between Granulosa Cells and Oocyte
Diameter
The relationships for each genotype that could be described by the equation y = Aecx are summarized in Figure
3, where y = the number of granulosa cells, X = the diameter of the oocyte, A = the intercept, and C = the rate
of increase in the number of granulosa cells relative to the
diameter of the oocyte. The relationships could be expressed by the following equations: y = 1.344 e .0 554X (R2
= 0.8393), y = 1.273e0 .05 22X (R2 = 0.7404), y =
0.975e ° ° 56 9 X (R2 = 0.8686), and y = 2.092e °.O385X (R2 =
0.7404) for the ++, I+, pI+, and pII genotypes, respectively. From multiple t-tests of the constants A and C, it
was evident that the values for the pII genotype were both
L.
Qa
E
1a
T
TABLE 5. Effect of the Inverdale gene on ovarian volume, cortical volume, and number of germ cells per ovary in adult ewes.
Genotype
Total number of
germ cells x10 3
Number of primordial
follicles x10
Number of primary
follicles
Number of secondary
follicles
Number of animals
++
I+
II
707"
(561, 912)
56a
(40, 84)
44"
(26, 74)
2691"
(1513, 4786)
660"
(169, 2511)
5
.
723
(545, 970)
44.
(22, 88)
34"
(10, 118)
2818"
(457, 18,197)
630"
(79, 5011)
6
164"1
(86, 310)
57,
(28, 114)
51"
(15, 170)
3388A
(1659, 6918)
0b
5
* Values are geometric means (and 95% confidence intervals).
For each row, values with different alphabetical superscripts are significantly different from one another (p < 0.01; Duncan's Multiple Range
test).
,,b
0
O
p I+
-
P 11
0 3 85
y = 2.0919eO.
i
x
2
R= 0.7404
7-
Inverdale genotype*
Ovarian volume (mm3 )
I+
++
L.
0
0
_Mist
0
50
100
50
100
Oocyte diameter (m)
FIG. 3. Relationship between the number of granulosa cells surrounding
the oocyte in the optical section containing the nucleolus and the diameter of the oocyte in that same section. ++, I1+, p+, and pll refer to
noncarrier, heterozygous carrier, putative heterozygous carrier, and putative homozygous carrier of the Inverdale fecX' gene.
1190
SMITH ET AL.
Plasma FSH
Plasma LH
A
4.5 -
1
b
[E] ++
1
4-
a
b
[m++
p 1+
ll+
3.5-
rp II
1
3E
E
2.5-
I
a
2-
I
,C,
j
a
LL
2-
1.51
0.5
0
Day 90
Day 105
Day 120
Day 135
Adult
Gestational Age
1211 5 6
Day 90
uay 1U0
uay 12u
Day 135
Adult
Gestational Age
FIG. 4. Plasma concentrations of FSH in Inverdale ++, I+, pl+, and
pll female fetuses at Days 90, 105, 120, and 135 of gestation and during
adult life. Histograms represent geometric means, and the vertical bars
represent 95% confidence limits. a vs. b, p < 0.01 (Duncan's Multiple
Range test). Numbers under the histograms represent the number of fetuses.
FIG. 5. Plasma concentrations of LH in Inverdale ++, I-, putative putative
pl,
and pll female fetuses at Days 90, 105, 120, and 135 of gestation and
during adult life. Histograms represent geometric means, and the vertical
bars represent 95% confidence limits. a vs. b, p < 0.01 (Duncan's Multiple Range test). Numbers under the histograms represent the number of
fetuses.
Endocrinology of the 11/1+ Fetuses and II Adults Carrying
the Inverdale Gene
available to distinguish between the pI+ and II genotypes.
However, as I+ female fetuses could be produced from
mating I rams with + + females, it was possible to use this
known genotype as a reference to distinguish between the
pI+ and II genotypes. Using this criterion it appears that
the pattern of germ cell development at Days 40 and 90 of
gestation, but not thereafter, was retarded in the I+ and pI+
relative to the control ++ fetuses. For example, in the I+
and pI+ fetuses, there were fewer germ cells at Day 40
and more at Day 90 relative to the + + genotype. At Day
40, the germ cells were present in all genotypes as mitotically active oogonia as in other sheep breeds [9]. Likewise
at Day 90, the germ cells were present as a mixture of
degenerating cells, oogonia, and oocytes with no follicular
development beyond the primordial stage of growth [9].
From previous studies [9, 11], we noted that most germ
cells (i.e., - 80%) are lost through atresia between Days
75 and 90. The finding of lower numbers of atretic germ
cells and oocytes in the + + and plI genotypes, as compared with the known I+ and pI+ genotypes, suggests that
the wave of atresia had largely been completed in the + +
and II genotypes whereas in the I+ genotype, atresia was
still widespread. It appears that the extent of atresia is as
severe in the I+ genotypes as in the + + and II genotypes,
because by Day 105, the total numbers of germ cells including the number of oocytes and primordial follicles were
essentially the same in all genotypes.
At Day 90 many of the germ cells are dividing meiotically [8, 9]. The development of oogonia into oocytes is
also the time of X-chromosome reactivation in female germ
cells [15]. Entry into meiosis is not an autonomous property
of germ cells per se but is thought to be induced by the
mesonephric-derived rete cells [16]. The finding of similar
proportions of meiotic germ cells in the various genotypes
During fetal development, FSH and LH were not detectable in either the pituitary gland or peripheral plasma
at Day 40 of gestation. However, both hormones were detectable at Days 90, 105, 120, and 135 of gestation, and
these results together with those for the adult animals are
summarized in Figures 4 and 5. For plasma FSH, no differences were noted between the genotypes during fetal life.
However, in the adult animals, the plasma concentrations
of FSH in the II genotype were significantly higher (p <
0.001) than in either the I+ or ++ genotype (Fig. 5). For
plasma LH, the only genotype difference noted during fetal
life was at Day 90 of gestation (Fig. 5): at this time the
concentrations in the + + genotype were significantly higher than in the known I+ or pI+ and II genotypes (p <
0.01). In adult animals the plasma concentrations in the II
genotype were significantly higher than in either the + +
or I+ genotype (p < 0.01).
DISCUSSION
The results of these studies on female fetuses carrying
the Inverdale gene have revealed differences in ovarian development that are dependent on whether the fetuses carry
one or two copies of the fecX l gene. For animals with one
copy of the X-linked mutation, differences were noted in
the numbers of germ cells present at Days 40 and 90 but
not at Days 105 or thereafter. In contrast, for animals
thought to carry two copies of the X-linked mutation, differences were not noted in the numbers of germ cells at
any stage of fetal life but were observed in ovarian follicular development at Day 105 and thereafter.
At the time of these studies, DNA markers were not
OVARY AND ENDOCRINOLOGY OF INVERDALE FETUSES
at Day 90 (Table 2) suggests that the fecX gene may not
exert its influence on the factor(s) that induce meiosis.
At both Days 40 and 90 of gestation, the mean diameters
of the oogonia were not different between the genotypes.
However, an unexpected result was the finding of differences between the genotypes in the mean diameters of isolated oocytes and follicle-enclosed oocytes at Day 90 but
not at Day 105 or thereafter. At Day 90 the mean diameters
of the follicles in the known I+ and pI+ genotypes were
smaller than those in the + + and II genotypes; these differences in the mean diameter of the primordial follicles
were due, at least in part, to the differences in the mean
oocyte diameters. It is possible that the variations in oocyte
diameter were due to the differences in the numbers of
oocytes at this time, reflecting the pattern of germ cell development between the genotypes. In both the I+ and pI+
genotypes at Day 90, the mean numbers of oocytes were
approximately twice those in the + + and II genotypes.
Moreover, significant linear relationships were found for the
number of isolated or follicle-enclosed oocytes and their
respective diameters.
The pattern of germ cell development, including that of
follicle formation in the pII fetuses, paralleled that in the
+ + genotype. Indeed it was not until Day 105 of gestation
and thereafter that differences between the pII animals and
the other genotypes were noted. At Days 105, 120, and 135
of gestation, pII ovaries contained a higher frequency of
abnormally large oocytes and oocyte-free nodules. The intraovarian rete is a prominent feature of fetal and neonatal
ovaries and is often associated with, or is adjacent to, primordial germ cells, oocytes, and developing follicles [3,
16]; this was also the case in the present study. In the pII
genotype, the oocyte-free follicles were confirmed to be
surrounded by a basement membrane and not open-ended
structures associated with the ovarian rete. In the pII genotype, follicles developing normally beyond the primary
stage of growth were not observed in either the fetal or
adult ovaries. From comparison of oocyte diameters with
the number of granulosa cells in the largest cross section
of follicles, it seems clear that in the II but not the I+ or
+ + genotype, oocyte enlargement is not accompanied by
a parallel increase in the number of granulosa cells and/or
an organized arrangement of granulosa cells. Although no
evidence was found for effects of the fecX gene in homozygotes before the initiation of follicular growth (i.e.,
before Day 105), it seems clear that two copies of the Xlinked mutation has a detrimental effect on follicular
growth from the primary stage of development during both
fetal and adult life. The finding of lost oocytes and masses
of interstitial-like cells in the medullary region at Days 120
and 135 may well be a consequence of abnormal germ cellsomatic cell interactions from Day 105 onward. These unusual features in the medullary region were similar to those
previously reported for adult ovaries [3].
In adult ewes, no differences were noted among the three
genotypes (i.e., ++, I+, and II) with respect to the total
numbers of germ cells. However, the volume of the ovaries
of the II genotype was significantly (p < 0.01) smaller than
that of the I+ and + + genotype. This finding, together with
the observation of no normal follicular growth beyond the
primary stage of development in the II genotype, confirms
our previous observations [3]. In the pII genotype, the volume of the ovary did not differ from that in the + + genotype during fetal life. It seems that the difference in volume evident in adult life occurs at some time during the
first 6 mo of postnatal life (unpublished results).
1191
There have been reports for females of other species
showing that mutations of the X chromosome affect fertility. For example, in women, X-chromosome translocations
are often associated with primary or secondary amenorrhea,
suggestive of a deficiency of oocytes [17]. An inherited
deletion in the long arm of the X chromosome also causes
premature menopause with bilateral streak ovaries [18, 19].
Ovaries with deficient oocytes have also been reported in
infants with trisomy 18 [20] or trisomy 21 (Down's syndrome) [21]. However, it has not been possible to study the
etiology of the above-mentioned conditions through quantitative studies of ovarian development in humans. More
often, animal models have been used to study the development of mutant ovaries. In XO mice, a massive loss of
oocytes occurs during the meiotic prophase [22, 23], and
this is thought to be due to anomalies in pairing of the X
chromosomes [24]. As discussed earlier, it seems unlikely
that the X-linked mutation in Inverdale ewes causes major
differences in germ cell atresia during meiotic prophase as
is the case with XO mice.
In I+ fetuses, the finding of significantly lower numbers
of germ cells at Day 40 of gestation and significantly greater numbers of germ cells at Day 90 relative to the control
values is similar to that for homozygous (BB) or heterozygous (B+) carriers of the Booroola gene relative to the
controls [9]. In the case of the Booroola sheep fetus, in
which detailed studies of germ cell development were done
at Days 40, 55, 75, 90, 95, 120, and 135 days of gestation,
it was hypothesized that the patterns of germ cell development, including follicular formation and growth, were
retarded in BB/B+ relative to those in + + animals. One
important difference between Inverdale and Booroola sheep
highlighted by this study is that the developmental fates of
germ cells in the II and I+ are different whereas in the BB
or B+ genotypes they are the same.
No genotype differences were found for the concentrations of FSH or LH in peripheral plasma at any stage of
gestation. The only exception to this was noted for the plasma LH concentrations at Day 90. At this time the concentrations in the + + genotype were significantly higher than
in the I+, p+, and pII genotypes. Although follow-up
studies on this observation are needed, the values found for
the I+ and II genotypes are somewhat lower than those
obtained for other breeds of sheep and for Booroola ewes
at this time [11]. Since plasma FSH concentrations are
within the ranges found for other breeds, the possibility
exists that at Day 90 and earlier, something may be different about the regulation of LH synthesis or secretion that
is related to the X-linked mutation. The values for FSH and
LH in adult ewes match those reported in previous studies
[3] and those found in ovariectomized ewes. These results
are consistent with the notion that the ovaries of the II
genotype secrete little or no steroid or inhibin. The only
exception is in cases in which the adult II ewes develop
abnormal ovarian structures thought to be tumors; these putative tumors, which occur in approximately one third of
the animals, are a major source of biologically active inhibin [3]. Despite significant differences in ovulation rate
between I+ and + + ewes, no differences were noted in
the FSH and LH concentrations in the plasma of these ewes
in the present study. These observations are similar to results reported elsewhere [4].
In summary, the evidence suggests that the fecX gene
influences the timing of germ cell development in I+ but
not II ovaries from as early as Day 40 of gestation. Moreover, when two copies of the fecX gene are present, normal
1192
SMITH ET AL.
follicular growth beyond the primary stage of development
during both fetal and postnatal life is impaired. No differences in plasma gonadotropin concentrations were noted
between the genotypes in fetal life except for the possibility
that the X-linked mutation may affect LH concentrations at
Day 90 of gestation but not thereafter.
ACKNOWLEDGMENTS
We wish to thank Leanne Still for preparation of tissues for histological
studies; Grant Shackell and the staff of the Invermay Agriculture Centre
for assistance in recovering the Inverdale fetuses; and Stan Lun, Louise
Shaw, and Lisa Condell for preparation of pituitary extracts and assay of
FSH and LH. The RIA reagents were obtained from the National Hormone
and Pituitary Program and the NIADDK.
10.
11.
12.
13.
14.
REFERENCES
I. Davis GH, McEwan JC, Fennessy PF, Dodds KG, Farquhar PA. Evidence for the presence of major gene influencing ovulation rate on
the X chromosome of sheep. Biol Reprod 1991; 44:620-624.
2. Davis GH, McEwan JC, Fennessy PF, Dodds KG, McNatty KP, O
W-S. Infertility due to bilateral ovarian hypoplasia in sheep homozygous (FecX'FecXl) for the Inverdale prolificacy gene located on the
X chromosome. Biol Reprod 1992; 46:636-640.
3. Braw-Tal R, McNatty KP, Smith P, Heath DA, Hudson NL, Phillips
DJ, McLeod BJ, Davis GH. The ovaries of ewes homozygous for the
X-linked Inverdale gene (FecX') are devoid of secondary and tertiary
follicles but contain many abnormal structures. Biol Reprod 1993; 49:
895-907.
4. Shackell GH, Hudson NL, Heath DA, Lun S, Shaw L, Condell L,
Blay LR, McNatty KP. Plasma gonadotrophin concentrations and
ovarian characteristics in Inverdale ewes which are heterozygous for
a major gene (FecX 1) on the X-chromosome that influences ovulation
rate. Biol Reprod 1993; 48:1150-1156.
5. Singh RP, Carr DH. The anatomy and histology of XO human embryos and fetuses. Anat Rec 1966; 155:369-384.
6. Lyon ME Hawker SG. Reproductive lifespan in irradiated and unirradiated XO mice. Genet Res 1973; 21:185-194.
7. Fitch N, de Saint Victor J, Richer CL, Pinsky L, Sitahal S. Premature
menopause due to a small deletion in the long arm of the X-chromosome: a report of three cases and a review. Am J Obstet Gynecol
1982; 142:968-972.
8. Mauleon P. Oogenesis and folliculogenesis. In: Cole HH, Cupps PT
(eds.), Reproduction in Domestic Animals. New York, London: Academic Press; 1969: 175-202.
9. Smith P, O W-S, Hudson NL, Shaw L, Heath DA, Condell L, Phillips
DJ, McNatty KP Effects of the Booroola gene (FecB) on body weight,
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
ovarian development and hormone concentrations during fetal life. J
Reprod Fertil 1993; 98:41-54.
Zamboni L, Bezard J, Mauleon P The role of the mesonephros in the
development of the sheep fetal ovary. Ann Biol Anim Biochim Biophys 1979; 19:1153-1178.
Smith P, Braw-Tal R, Corrigan K, Hudson NL, Heath DA, McNatty
KP. Ontogeny of ovarian follicle development in Booroola sheep fetuses that are homozygous carriers or non-carriers of the FecBB gene.
J Reprod Fertil 1994; 100:485-490.
Tisdall DJ, Watanabe K, Hudson NL, Smith P, McNatty KP FSH
receptor gene expression during ovarian follicle development in sheep.
J Mol Endocrinol 1995; 15:273-281.
Gundersen HJG. Stereology of arbitrary particles: a review of unbiased number and size estimators and the presentation of some new
ones. J Microsc 1986; 143:3-45.
McNatty KP, Hudson NL, Collins F, Fisher M, Heath DA, Henderson
KM. Effects of oestradiol-17p, progesterone or bovine follicular fluid
on the plasma concentrations of FSH and LH in ovariectomized Booroola ewes which were homozygous carriers or non-carriers of a fecundity gene. J Reprod Fertil 1989; 87:573-585.
Kay GF, Penny GD, Patel D, Asworth A, Brockdorff N, Rastan S.
Expression of Xist during mouse development suggests a role in the
initiation of X chromosome inactivation. Cell 1993; 72:171-182.
Byskov AG. Does the rete ovarii act as a trigger for the onset of
meiosis? Nature 1974; 252:396-397.
Madan K, Hompes PGA, Schoemaker J, Ford CE. X-autosome translocation with a breakpoint in Xq22 in a fertile woman and her 47.
XXX infertile daughter. Hum Genet 1981: 59:290-296.
Krauss CM, Turksoy RN, Atkins L, McLaughlin C, Brown BS, Page
DC. Familial premature ovarian failure due to interstitial deletion of
the long arm of the X-chromosome. N Engl J Med 1987; 317:125127.
Veneman TF, Beverstock GC, Exalto N, Molevanger P Premature
menopause because of an inherited deletion in the long arm of the
X-chromosome. Fertil Steril 1991; 55:631-633.
Russell LB, Altschuler G. The ovarian dysgenesis of trisomy 18. Pathology 1975; 7:149-155.
Hojager B, Peters H, Byskov AG, Faber M. Follicular development
in ovaries of children with Down's syndrome. Acta Paediatr Scand
1978; 67:637-643.
Burgoyne PB, Baker TG. Oocyte depletion in XO mice and their XX
sibs from 12 to 200 days post partum. J Reprod Fertil 1981; 61:207212.
Burgoyne PG, Baker TG. Meiotic pairing and gametogenic failure.
In: Evans CW, Dickinson HG (eds.), Controlling Events in Meiosis.
Cambridge: Company of Biologists; 1984: 349-362.
Burgoyne PB, Baker TG. Perinatal oocyte loss in XO mice and its
implications for the aetiology of gonadal dysgenesis in XO women. J
Reprod Fertil 1985; 75:633-645.