0021-972X/98/$03.00/0
Journal of Clinical Endocrinology and Metabolism
Copyright © 1998 by The Endocrine Society
Vol. 83, No. 11
Printed in U.S.A.
Premature Response to Luteinizing Hormone of
Granulosa Cells from Anovulatory Women with
Polycystic Ovary Syndrome: Relevance to Mechanism
of Anovulation*
DEBBIE S. WILLIS, HAZEL WATSON, HELEN D. MASON, RAY GALEA,
MARK BRINCAT, AND STEPHEN FRANKS
Department of Reproductive Science and Medicine, Division of Pediatrics, Obstetrics and Gynecology,
Imperial College of Science, Technology, and Medicine, St. Mary’s Hospital (D.S.W., H.W., H.D.M.,
S.F.), London, United Kingdom W2 1PG; and the Department of Obstetrics and Gynecology, University
of Malta, St. Luke’s Hospital (R.G., M.B.), Gwardamangia, Malta
ABSTRACT
Polycystic ovary syndrome is the most common cause of anovulatory infertility. Anovulation in polycystic ovary syndrome is characterized by the failure of selection of a dominant follicle with arrest of
follicle development at the 5–10 mm stage. In an attempt to elucidate
the mechanism of anovulation associated with this disorder we have
investigated at what follicle size human granulosa cells from normal
and polycystic ovaries respond to LH.
Granulosa cells were isolated from individual follicles from unstimulated human ovaries and cultured in vitro in serum-free medium 199 in the presence of LH or FSH. At the end of a 48-h incubation
period, estradiol (E2) and progesterone (P) were determined in the
granulosa cell-conditioned medium by RIA.
In ovulatory subjects (with either normal ovaries or polycystic
ovaries), granulosa cells responded to LH once follicles reached 9.5/10
P
OLYCYSTIC ovary syndrome (PCOS) is a common endocrine disorder associated with hyperandrogenemia,
infertility, and insulin resistance (1). Although this syndrome
is the most common cause of anovulatory infertility, the
mechanism of anovulation remains an enigma. It has become
apparent, however, that hyperinsulinemia and insulin resistance are features of anovulatory women with polycystic
ovaries and not equally of hyperandrogenemic, weightmatched ovulatory women with normal ovaries or polycystic
ovaries (2–5). The possibility, therefore, exists that insulin
may be involved in the mechanism of anovulation associated
with PCOS. This hypothesis is supported by several clinical
observations, in particular the detrimental effects of obesity
(itself associated with insulin resistance) on fertility in
women with polycystic ovaries.
Obese women with polycystic ovaries are more likely to be
anovulatory and experience menstrual disturbance, and are
less likely to ovulate in response to clomiphene citrate or
Received March 3, 1998. Revision received July 14, 1998. Accepted
July 24, 1998.
Address all correspondence and requests for reprints to: Dr. Debbie
Willis, Department of Obstetrics and Gynecology, Imperial College
School of Medicine, St. Mary’s Hospital, London, United Kingdom W2
1PG. E-mail: [email protected].
* This work was supported by the Medical Research Council (to
D.S.W. and H.W.).
mm. In contrast, granulosa cells from anovulatory women with polycystic ovaries responded to LH in smaller follicles of 4 mm. Granulosa
cells from anovulatory women with polycystic ovaries were significantly more responsive to LH than granulosa cells from ovulatory
women with normal ovaries or polycystic ovaries (E2, P , 0.0003; P,
P , 0.03). The median (and range) fold increase in estradiol and
progesterone production in response to LH in granulosa cell cultures
from size-matched follicles 8 mm or smaller were E2, 1.0 (0.5–3.9) and
P, 1.0 (0.3–2.5) in ovulatory women and E2, 1.4 (0.7–25.4) and P, 1.3
(0.3–7.0) in anovulatory women.
Granulosa cells from anovulatory (but not ovulatory) women with
polycystic ovaries prematurely respond to LH; this may be important
in the mechanism of anovulation in this common endocrinopathy.
(J Clin Endocrinol Metab 83: 3984 –3991, 1998)
gonadotropin administration despite similar increases in
FSH than their lean counterparts (6 – 8). Conversely, improvement in the insulin status of anovulatory women with
PCOS by weight loss, accompanied by a reduction in insulin
levels, or insulin-lowering drugs can lead to a resumption of
ovulatory cycles and fertility in some patients (9 –12).
We have previously shown that despite peripheral insulin
resistance in women with PCOS, the polycystic ovary remains responsive to insulin (in terms of steroidogenesis) and
that insulin acts via its own receptor rather than the type I
insulin-like growth factor receptor (13, 14). Preincubation
with insulin sensitizes human granulosa cells to LH, greatly
enhancing granulosa cell progesterone (P) production in response to LH stimulation (14). This suggests enhanced granulosa cell differentiation by insulin in vitro (14). We propose,
therefore, that hyperinsulinemia in anovulatory women with
PCOS may lead to premature maturation of granulosa cells
in vivo.
If this hypothesis is correct, granulosa cells from anovulatory women with polycystic ovaries (anovPCO) may respond to LH at an earlier stage of development than granulosa cells from ovulatory women with polycystic ovaries
(ovPCO) or normal ovaries (since hyperinsulinemia is associated with anovulation in women with polycystic ovaries).
Hence, the aims of this study were 1) to determine at what
3984
LH RESPONSE IN HUMAN GRANULOSA CELLS
follicle size human granulosa cells from normal ovaries and
polycystic ovaries respond to LH in terms of steroid production in vitro, and 2) to compare the response to LH in
granulosa cells from anovulatory women with polycystic
ovaries with that in ovulatory women with either normal or
polycystic ovaries.
3985
mm were cultured separately, and follicles 4 mm or smaller from
individual patients were pooled, except in the case of two 3.5-mm and
five 4-mm follicles from which sufficient cells were obtained to be
cultured separately. Where granulosa cell number permitted, cells
were plated at 5 3 104 viable cells/well in 200 mL serum-free medium
199 supplemented with antibiotics (penicillin and streptomycin) and
200 mmol/L glutamine (Life Technologies, Paisley, Scotland), and
experiments were performed in triplicate. In the majority of experiments, when insufficient cells were obtained to achieve this, we used
a reduced plating density in duplicate or, if necessary, singleton
wells.
The granulosa cells were incubated with LH (5 ng/mL) or, as a
positive control, FSH (5 ng/mL) in the presence of 5 3 1027 mol/L
testosterone, as aromatase substrate, for 48 h. When the number of
granulosa cells in a single follicle was sufficient (as in dominant or
preovulatory follicles) additional doses of LH and FSH (0.625–10
ng/mL) were tested. FSH concentrations of 2.5 and 5 ng/mL were
commonly used because in previous studies these were found to give
maximal estradiol (E2) responses (16). Gonadotropin preparations
were highly purified human pituitary FSH (,16 IU LH/mg) and LH
(,7 IU FSH/mg) supplied by Dr. S. Lynch of Endocrine Services
(Bidford on Avon, UK). At the end of the 48-h incubation, the medium
was removed and stored at 220 C until measurement of E2 and P by
RIA as previously described (14). Viable granulosa cell number was
assessed using the trypan blue exclusion test after isolation and in
some cases at the end of the incubation period using a commercially
available kit (CellTiter 96 nonradioactive cell proliferation assay,
Promega Corp., Southampton, UK).
Where experiments were conducted in triplicate wells, the results are
shown as the mean and se of these triplicates. In these experiments, if
error bars are not visible, the errors are contained within the mean bar.
When comparing the response of granulosa cells to gonadotropins in
multiple patients, statistics were performed using Student’s t test, Wilcoxon signed rank test, and Mann-Whitney U test as appropriate. Significance was assumed when P , 0.05.
Subjects and Methods
Ovarian tissue was obtained from 23 women [10 with normal ovaries
(N), 10 ovPCO, and 3 anovPCO)] undergoing bilateral oophorectomy for
nonovarian disease. Details of each patient’s age, body mass index,
ovarian morphology, menstrual cycle history, indication for surgery,
follicle sizes used in the study, and serum hormone concentrations are
given in Table 1. No patient had been receiving treatment for suppression or stimulation of the ovaries for at least 3 months before surgery.
Approval for the use of ovarian tissue in this study was obtained from
Kensington, Chelsea, and Westminister Health Authority ethical committees and the research ethics committee of the Faculty of Medicine and
Surgery of the University of Malta.
Polycystic ovaries were defined on the presence of 3 or more of the
following criteria: enlarged ovarian volume (.9 mL), 10 or more follicles
of 2– 8 mm in diameter, and increased density and volume of stroma or
a thickened tunica (15). Patients with polycystic ovaries and a history of
oligomenorrhea or amenorrhea and whose ovaries contained no recent
corpora lutea were designated anovulatory, and those patients with
regular cycles and recent corpora lutea were designated ovulatory.
The number and size range of follicles used in the study were: 25
follicles 13 mm or smaller and four preovulatory follicles from normal
ovaries, 62 follicles 12 mm or smaller and two preovulatory follicles from
ovPCO, and 35 follicles 8 mm or smaller from anovPCO.
Granulosa cell cultures
Granulosa cells were isolated and cultured as previously described
(14). Briefly, granulosa cells from individual follicles greater than 4
TABLE 1. Clinical details of the patients from whom ovarian tissue was obtained
Patient
no.
Age
(yr)
BMI
(kg/m2)
Ovarian
morphology
Day of
cycle
Cycle
history
1
2a
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21a
22a
48
50
37
41
43
38
38
39
26
48
35
32
20
6
26
14
8
5
Regular
Regular
Regular
Regular
Regular
Regular
Regular
Regular
Regular
Regular
Regular
Irregular
Irregular
Regular
Regular
Regular
Regular
Regular
Regular
30
45
29
N
N
N
N
N
N
N
N
N
N
ovPCO
ovPCO
ovPCO
ovPCO
ovPCO
ovPCO
ovPCO
ovPCO
ovPCO
ovPCO
anovPCO
anovPCO
23
35
32
anovPCO
44
40
38
39
38
47
39
25
25
17
25
24
30
24
20
19
30
36
21
12
16 just ovulated
10
13
17 just ovulated
5
13
15
16
luteal
21
No. of follicles (size range,
mm) used in the study
Indication for
surgery
Meno, fibroids
PP
Meno
Meno
PP
PP
PID
PP
Meno, fibroids
Meno, fibroids
Meno
PP
Meno
PP
PP
Meno
Meno, fibroids
#10 mm
5
1
3
5
5
2
7
6
3
1
16
10
(2–7)
(7.5)
(4 – 8)
(5–9)
(3–10)
(5–7)
(3– 6.5)
(3–7.5)
(5–5.5)
(7)
(3.5–9.5)
(2.5–9.5)
16
24
20
3
4
5
(3–7)
(3– 8.5)
(2– 8.5)
(3– 4.5)
(3–10)
(3– 6)
Luteal
1.5 months ago
Irregular
Ca cervix stage III
Irregular
Endo
18 (3–7.5)
1 (6.5), remainder
sent to histology
13 (3.5–7.5)
.10 mm
Serum hormone levels
LH
(IU/L)
FSH
(IU/L)
Fasting
insulin
(mU/L)
8.1
7.0
7.6
3.6
3.8
4.9
4.3
1.7
1.2
1 (18)
1 (18)
1 (12)
4.9
2.9
1 (12)
1 (20)
1 (17)
5.3
9.3
6.8
2.8
11.4
3.0
1.2
5.6
5.3
3.0
1.8
9.7
8.4
5.1
8.0
3.8
3.1
6.5
7.0
15.1
8.3
6.4
12.8
1 (19)
1 (13)
1 (18)
24
2.7
2.7
10.4
Meno, Menorrhagia; PP, pelvic pain; PID, pelvic inflammatary disease; Endo, endometrious. A fasting serum insulin concentration less than
10 mU/L was considered normal (5).
a
One ovary only.
3986
WILLIS ET AL.
Follicular fluid steroid levels
Follicular fluid E2 and androstenedione concentrations were determined in some follicles to assess the health of the individual follicles.
These were determined using the same RIAs as those employed for the
determination of steroid levels in the culture medium and were previously described in detail (14, 17).
Serum hormone measurements
Serum LH and FSH levels were measured using a commercially
available kit from Chelsea Kits, Division of Molecular Endocrinology,
Hammersmith Hospital (London, UK) (17a). Fasting serum insulin levels were determined by RIA as previously described (5). The assay used
a polyclonal antibody to insulin that cross-reacts with proinsulin (100%)
and C peptide (3.8%).
Results
Two major findings were apparent from this study. The
first was the marked difference in basal steroid production
and the magnitude of the response to gonadotropins between granulosa cells obtained from different follicles of the
same patient. The second finding was that granulosa cells
from follicles 8 mm or smaller from anovulatory women with
polycystic ovaries were significantly more responsive to LH
than granulosa cells from size-matched follicles from ovulatory women with either normal ovaries or polycystic ovaries. In anovPCO women, follicles 4 mm or larger contained
granulosa cells responsive to LH, whereas in ovulatory
women with normal ovaries or polycystic ovaries, follicles
were responsive to LH only when they had reached a size of
9.5 mm or larger.
JCE & M • 1998
Vol 83 • No 11
either E2 or P production in these granulosa cell cultures (E2,
P . 0.9; P, P . 0.4). Figures 1 and 2 show the effect of LH
and FSH on E2 and the corresponding P accumulation, respectively, in granulosa cell cultures from a range of follicle
sizes from women with normal ovaries. Figures 1i and 2i
illustrate the data obtained from granulosa cells pooled from
1.5-, 2-, and 3-mm follicles (a); from a 5-mm follicle (b); and
from a 7-mm follicle (c) all isolated from the same patient
with normal ovaries who was in the midfollicular phase of
her menstrual cycle. There was no effect of gonadotropins on
steroid levels in granulosa cell cultures from 1.5-, 2-, and
3-mm follicles. As expected, FSH augmented E2 (Fig. 1i) and
P (Fig. 2i) accumulation in granulosa cells from the 5- and
7-mm follicles. As the 7-mm follicle was the dominant follicle, more cells were obtained in which to study gonadotropin action, and FSH and LH were tested at two concentrations (2.5 and 5 ng/mL). FSH dose-dependently
stimulated E2 and P production, whereas there was no effect
of LH on steroidogenesis at either concentration tested (only
data for the 5 ng/mL dose are shown in Figs. 1i and 2i).
There was no effect of LH on steroid production in granulosa cells from 7.5-, 8-, or 9-mm follicles (Figs. 1ii and 2ii, 1iii
and 2iii, and 1iv and 2iv, respectively) from normal ovaries,
but as expected, FSH enhanced steroid levels in these cell
cultures.
Antral follicles (.9 mm)
The effect of gonadotropin treatment on E2 production by
granulosa cells isolated from a range of follicle sizes from
6 of the 10 patients with normal ovaries is shown in Figs. 1
and 2.
Six follicles larger than 9 mm were obtained from normal
ovaries (one 10 mm, one 13 mm, and four preovulatory
follicles). In granulosa cell cultures from follicles 10 and 13
mm in diameter, both FSH and LH augmented E2 (Fig. 1, v
and vi, respectively) and P (Fig. 2, v and vi, respectively)
accumulation. Results obtained for the preovulatory follicles
are shown below, together with those obtained from preovulatory follicles from ovPCO.
Small antral follicles (#9 mm)
Ovulatory polycystic ovaries
In granulosa cells isolated from small antral follicles (#9
mm) from normal ovaries and then cultured in vitro, 16 of 20
cultures responded to FSH. FSH, as expected, significantly
stimulated E2 [median (range) fold increase, 2.0 (0.7–3.4); P ,
0.002] and P accumulation [median (range) fold increase, 2.5
(0.3–5); P , 0.003]. There was no effect of LH treatment on
As expected, more follicles were obtained from ovPCO
than from normal ovaries; hence, a wider range of follicle
sizes was available for study. Granulosa cells from ovPCO
had a similar response to gonadotropins as granulosa cells
from normal ovaries. Figure 3 illustrates the effects of LH and
FSH on E2 production in granulosa cells isolated from a range
Normal ovary
FIG. 1. Effects of LH (L) and FSH (F) at
5 ng/mL on E2 production in granulosa
cell cultures from six different ovulatory patients with normal ovaries. The
data in each graph, i.e. i, ii, iii, iv, v, and
vi, were obtained from different patients. The follicle size from which the
granulosa cells were isolated is shown
at the top of each graph. In i, the data
were obtained from a patient with normal ovaries in the midfollicular phase of
her menstrual cycle. The effects of FSH
and LH are shown in granulosa cell cultures pooled from 1.5-, 2-, and 3-mm
follicles (a); from a 5-mm follicle (b); and
from a 7-mm follicle (c). C, Control wells
(cells treated with testosterone alone).
LH RESPONSE IN HUMAN GRANULOSA CELLS
3987
FIG. 2. Effects of LH (L) and FSH (F) at
5 ng/mL on P accumulation in granulosa cell cultures from six different ovulatory patients with normal ovaries.
The data are the P data for the granulosa cell cultures shown in Fig. 1.
FIG. 3. Effects of LH (L) and FSH (F) at
5 ng/mL on E2 production in granulosa
cell cultures from ovulatory patients
with polycystic ovaries. The data in i,
a–l, were obtained from the same patient. In ii, the data were pooled from
three experiments in three different patients to show the significant effect of
LH at this stage in development (9.5
mm), and the data in iii were obtained
from a 12-mm follicle from a single patient. The follicle size from which the
granulosa cells were isolated is shown
at the top of each graph. C, Control wells
(cells treated with testosterone alone) *,
P , 0.05.
of follicle sizes obtained from three ovulatory patients with
polycystic ovaries. The corresponding data for P accumulation in these granulosa cell cultures are shown in Fig. 4.
Figures 3i and 4i illustrate the response to gonadotropins in
granulosa cells from a range of follicle sizes obtained from
the same patient. It highlights the range of basal and gonadotropin-stimulated steroid levels in granulosa cell cultures
from follicles isolated from the same patient.
Small antral follicles (#9 mm)
In granulosa cells from follicles 9 mm or smaller from
ovPCO, 42 of 47 responded to FSH with a significant increase
in E2 [median (range) fold increase, 2.7 (0.8 –25.0)] and P
accumulation [median fold (range) increase, 3.7 (0.6 –31.5)]
above basal levels in the cell cultures (E2, P , 0.0001; P, P ,
0.001). Although a granulosa cell response to LH was observed in two of these follicles, there was no significant effect
of LH on steroidogenesis in granulosa cell cultures obtained
from follicles 9 mm or smaller (E2, P . 0.8; P, P . 0.9).
Antral follicles (.9 mm)
Nine follicles larger than 9 mm were obtained from
ovulatory polycystic ovaries (three 9.5-mm, one 10-mm,
three 12-mm, and two preovulatory follicles). The granulosa cells from all three of the 9.5-mm follicles (obtained
from different patients) significantly responded to FSH
and LH (Figs. 3ii and 4ii). In Figs. 3ii and 4ii, the data were
pooled from three individual experiments in which granulosa cells were isolated from three 9.5-mm follicles obtained from different patients to highlight the significance
of granulosa cells acquiring the ability to respond to LH
once follicles reach 9.5 mm in diameter in ovulatory
women with polycystic ovaries. The only 10-mm follicle
obtained from an ovulatory woman with polycystic ovaries was presumed to be atretic; it had few granulosa cells,
associated with low basal steroid production, and did not
respond to either gonadotropin (data not shown). This
patient had been in the midluteal phase of her menstrual
cycle; therefore, her ovary would not have been expected
to contain a healthy 10-mm follicle.
The responses of granulosa cells from three 12-mm follicles
obtained from different patients were varied. One, from a
patient in the late follicular phase of her menstrual cycle,
appeared to be a healthy dominant 12-mm follicle. This follicle was considered healthy because it was well vascularized
and contained 1,000,000 granulosa cells, and the follicular
3988
WILLIS ET AL.
JCE & M • 1998
Vol 83 • No 11
FIG. 4. Effects of LH (L) and FSH (F) at
5 ng/mL on P accumulation in granulosa cell cultures from ovPCO. The data
in i, a–l, were obtained from the same
patient. The data are the P data for the
granulosa cell cultures shown in Fig. 3.
*, P , 0.05.
fluid was estrogen dominant (18, 19) (E2, 3.11 mmol/L; androstenedione, 0.8 mmol/L). In granulosa cell cultures from
this healthy dominant follicle, LH augmented E2 and P accumulation (Figs. 3iii and 4iii, respectively).
In the other two 12-mm follicles, FSH, but not LH, augmented E2 production. However, LH-stimulated P accumulation in one of these cultures. Although these two 12-mm
follicles were the largest follicles obtained in each patient,
they did not appear to be healthy dominant follicles; they
contained relatively few granulosa cells (300,000 and
600,000), and the follicular fluid was androgen dominant in
both follicles (E2, 0.03 and 0.02 mmol/L; androstenedione, 11
and 12 mmol/L, respectively).
Preovulatory follicles from N ovaries and ovPCO
In total, six preovulatory follicles were obtained for investigation. In four of six follicles (three N and one ovPCO),
FSH and LH stimulated both E2 and P accumulation in the
granulosa cell cultures (Fig. 5, i and ii). In the other two
preovulatory follicles (one N and one ovPCO), FSH and LH
augmented granulosa cell P accumulation only and had no
effect on E2 accumulation above the already high basal E2
levels [E2 (Fig. 5iii) and P (Fig. 5iv)].
Anovulatory polycystic ovaries
Ovarian tissue was obtained from three anovulatory
women with PCOS. Two of the patients were hyperinsulinemic, as judged by the fasting serum insulin concentration
(Table 1) (5). Figures 6 and 7 illustrate the effects of LH and
FSH on steroid production in granulosa cells obtained from
a range of individual follicle sizes from two of the three
anovulatory patients with polycystic ovaries. Figures 6i and
7i show data from follicles 4 mm or smaller from one patient,
whereas Figs. 6ii and 7ii show results from follicles larger
than 4 mm from a second patient. As seen previously in N
and ovPCO groups, there was a wide range of basal and
gonadotropin-stimulated steroid levels among follicles from
the same patient. Although some of the follicles were presumed to be atretic (with low granulosa cell number and low
basal steroid levels that were not augmented by FSH or LH;
Fig. 6, ib, iic, and iid), the proportion of atretic follicles in the
anovPCO was similar to that in N ovaries, in that 22 of the
28 granulosa cell cultures responded to FSH. The median
(range) fold increases in E2 and P accumulation above the
control values in response to FSH were 1.4 (0.3–22.9) and 1.5
(0.3–25.2), respectively.
As observed in ovulatory subjects, granulosa cells from
follicles smaller than 4 mm from anovulatory women responded to FSH but not LH (Fig. 6i). However, in marked
contrast to the results obtained in ovaries from ovulatory
subjects, granulosa cells isolated from 4- to 8-mm follicles
from anovulatory women with polycystic ovaries that responded to FSH also responded to LH (Figs. 6 and 7). Analysis of responses in all follicles #8 mm from anovPCO
showed that FSH and LH significantly stimulated E2 (FSH,
P , 0.01; LH, P , 0.03) and P (FSH, P , 0.0001; LH, P , 0.02)
accumulation.
Interestingly, granulosa cells isolated from the four largest
follicles from one anovulatory patient (two 7.5-mm follicles
and both of her 8-mm follicles) had a response to gonadotropin stimulation similar to that seen in two of the preovulatory follicles described above. There was no effect of LH or
FSH on E2 production (Fig. 6ii, h– k), but both gonadotropins
augmented P accumulation in these same cultures (Fig. 7ii,
h– k, respectively), further suggesting that granulosa cells
in anovPCO may be prematurely advanced in their
development.
Comparison of the response to LH in granulosa cells from
ovulatory and anovulatory women
The effect of LH on granulosa cell steroidogenesis in ovulatory women (normal and polycystic ovaries) was compared to that in anovulatory women (with polycystic ovaries)
in all FSH-responsive granulosa cell cultures from follicles 8
mm or smaller (as the maximum follicle size obtained from
anovPCO was 8 mm in this study). The groups were matched
for follicle size.
The responses to LH (expressed in terms of fold increase
LH RESPONSE IN HUMAN GRANULOSA CELLS
FIG. 5. Effects of LH (L) and FSH (F) at 5 ng/mL on E2 (i and iii) and
P (ii and iv) production in granulosa cell cultures from preovulatory
follicles. i and ii are the pooled data from four preovulatory follicles
in which LH and FSH stimulated E2 production. iii and iv are the
pooled data from two preovulatory follicles in which LH and FSH had
no effect on E2 production. C, Control wells (cells treated with testosterone alone). *, P , 0.05.
in steroid production) were pooled from the results in all
ovulatory women (whether normal ovaries or ovPCO) and
compared to the responses to LH in granulosa cell cultures
from anovPCO. The median and range fold increases in E2
and P production in response to LH (5 ng/mL) are shown in
Table 2. Granulosa cells from follicles 8 mm or smaller from
anovPCO were significantly more responsive to LH than
granulosa cells from size-matched follicles from ovulatory
women with normal ovaries or polycystic ovaries (E2, P ,
0.0003; P, P , 0.03).
In summary, in ovulatory subjects, granulosa cells begin to
respond to LH when healthy follicles reach 9.5/10 mm in
diameter. In contrast, in anovPCO, granulosa cells from follicles as small as 4 mm respond to LH. Granulosa cells from
follicles 8 mm or smaller from anovPCO were significantly
more responsive to LH than granulosa cells from sizematched follicles from normal ovaries or ovPCO.
Discussion
This is the first report that provides evidence to support
the suggestion that granulosa cells from anovulatory women
3989
with PCOS are at a prematurely advanced stage of development. Granulosa cells from anovPCO were significantly
more responsive to LH than granulosa cells from sizematched follicles from normal and ovPCO. Hence, the premature response to LH is associated with anovulation in
women with PCOS and not the polycystic ovarian morphology per se.
To our knowledge this is also the first study documenting
the steroid responses to LH in individual follicles in the
human ovary. A clear, developmentally regulated response
to LH was observed in human granulosa cells, consistent
with previous studies (20 –24), but we also showed that granulosa cells were able to respond to LH in ovulatory subjects
(normal and polycystic ovaries) once follicles reached 9.5/10
mm in diameter.
Marked differences in basal steroid production and the
magnitude of response to gonadotropins among cultures of
granulosa cells obtained from different follicles of the same
ovary were observed. This finding was anticipated because
individual follicles in the same ovary are likely to be at
varying stages of development or atresia. This is an important concept in follicle selection. It is generally assumed that
at the time of selection, only one follicle is selected in humans
because that follicle is at a slightly more advanced stage of
development than other follicles in its cohort (25). Despite the
observed variations in basal steroid levels and in the responsiveness to gonadotropins within patients, we found that
granulosa cells from anovPCO women with polycystic ovaries were significantly more responsive to LH than granulosa
cells from ovulatory women with normal ovaries or polycystic ovaries. In anovulatory women with polycystic ovaries, granulosa cells from follicles as small as 4 mm responded to both FSH and LH. Granulosa cells from
anovulatory women with polycystic ovaries were significantly more responsive to LH than granulosa cells from
size-matched follicles 8 mm or smaller from ovulatory
women with normal or polycystic ovaries. This adds force to
the argument that granulosa cells in anovPCO are at a prematurely advanced stage of development.
We propose that raised insulin levels in anovulatory
women with polycystic ovaries may be a major factor causing
premature maturation of granulosa cells in vivo, as we have
previously shown in vitro that insulin increases the ability of
granulosa cells to respond to LH (14). Hillier has suggested
that the follicle requires LH throughout its development, but
if LH levels are raised above a certain threshold level, premature luteinization may be initiated (the so-called ceiling
hypothesis) (26). LH levels that are inappropriately high for
the stage of follicle development have been shown to be
detrimental to subsequent follicle maturation and ovulation
(26 –29). Premature exposure of granulosa cells to LH inhibits
their proliferation, such that development of the dominant
follicle is arrested (26, 30 –32). Furthermore, in normal follicle
development, once granulosa cells can respond to LH (i.e.
gain LH receptors) at 9.5–10 mm, they only undergo two
more cell divisions to reach the preovulatory, 20 –25 mm
stage (33). In the anovPCO, granulosa cells were responsive
to LH in follicles as small as 4 mm in diameter. Therefore, if
the cells can only undergo two more divisions, they would
3990
WILLIS ET AL.
JCE & M • 1998
Vol 83 • No 11
FIG. 6. Effects of LH (L) and FSH (F) at
5 ng/mL on E2 production in granulosa
cell cultures from two anovPCO. The
data in i, a– d, were obtained from one
patient, and the data in ii were obtained
from a second patient. The follicle size
from which the granulosa cells were isolated is shown at the top of each graph.
C, Control wells (cells treated with testosterone alone).
FIG. 7. Effects of LH (L) and FSH (F) at
5 ng/mL on P accumulation in granulosa cell cultures from two anovPCO.
The data are the P data for the granulosa cell cultures shown in Fig. 6.
TABLE 2. Fold increase in steroid accumulation in response to
LH in (FSH-responsive) granulosa cells from follicles 8 mm or
smaller from ovulatory women (with normal or polycystic ovaries)
compared to anovulatory women (with polycystic ovaries)
Patients
Ovulatory (normal and ovPCO)
Anovulatory (anovPCO)
Fold increase in steroid
accumulation in response to LH
above control
Estradiol
Progesterone
1.0 (0.5–3.9);
(n 5 46)a
1.4 (0.7–25.4);
(n 5 17)b
1.0 (0.3–2.5);
(n 5 42)c
1.3 (0.3–7.0);
(n 5 20)d
a
vs. b, P , 0.0003; c vs. d, P , 0.03. The range and number of
subjects are given in parentheses.
be expected to reach a maximum size of around 8 –10 mm
(33).
A balance between FSH and LH is required for follicle
development and ovulation (26). In anovulatory women
with polycystic ovaries, raised insulin levels may upset this
balance by sensitizing granulosa cells to the effects of LH.
Administration of clomiphene citrate and exogenous FSH in
ovulation induction programs may then redress the balance
in favor of FSH (either by increasing the FSH/LH ratio or
reaching a threshold level of FSH). This is supported by
studies in rat granulosa cells where a FSH/LH ratio of 2 or
less increased E2 production but had no effect on proliferation. However, increasing the FSH level to a FSH/LH ratio
of 3 or larger stimulated granulosa cell proliferation (34).
As with all biochemical markers of PCOS, there are exceptions to the general trend that anovulatory women with
PCOS are hyperinsulinemic; some anovulatory women with
PCOS have normal serum insulin levels (5). The cause of
anovulation in these patients is therefore not easy to explain
on the basis of insulin action on the follicle. Such patients
may, however, have a raised serum LH level (35), and therefore, it is possible that there is an interaction between insulin
and LH such that either a raised insulin (and normal LH) or
LH RESPONSE IN HUMAN GRANULOSA CELLS
a raised LH (and normal insulin) level can lead to premature
arrest of follicle growth and thus anovulation. Hyperandrogenemia in women with PCOS may also exacerbate the effects of insulin and LH. Androgens augment FSH-induced
granulosa cell cAMP production and differentiation (36, 37).
Raised insulin, LH, and androgen concentrations may, therefore, all have a role in prematurely advancing granulosa cell
differentiation.
With a heterogeneous syndrome such as PCOS, there may
be several causes of anovulation. The mechanism of anovulation proposed here may be relevant to hyperinsulinemic
women with polycystic ovaries. Other mechanisms that have
been suggested are inhibitors of aromatase or FSH, i.e. 5aandrostane-3,17-dione or the insulin-like growth factor-binding proteins (38, 39).
In conclusion, the data presented here suggest that granulosa cell differentiation is prematurely advanced in anovulatory women with PCOS. This may contribute to the mechanism of anovulation in these women. It is proposed that
hyperinsulinemia, by increasing the response of granulosa
cells to LH, amplifies the effects of LH (which may already
be elevated in these women). Thus, the effect of LH on the
maturing follicle is similar to that which, in the normal cycle,
is exerted only at later stages of development. This may lead
to premature differentiation of granulosa cells, as high levels
of LH inhibit granulosa cell proliferation and promote granulosa cell steroidogenesis and differentiation (26, 31, 32).
References
1. Franks S. 1995 Polycystic ovary syndrome. N Engl J Med. 333:853– 861.
2. Dunaif A, Graf M, Mandell J, Laumas V, Dobransky A. 1987 Characterization
of groups of hyperandrogenic women with acanthosis nigricans, impaired
glucose tolerance and/or hyperinsulinemia. J Clin Endocrinol Metab.
65:499 –507.
3. Conway GS, Jacobs HS, Holly JMP, Wass JAH. 1990 Effects of luteinizing
hormone, insulin, insulin-like growth factor-I and insulin-like growth factor
small binding protein-1 in the polycystic ovary syndrome. Clin Endocrinol
(Oxf). 33:593– 603.
4. Sharp PS, Kiddy DS, Reed MJ, Anyaoku V, Johnston DG, Franks S. 1991
Correlation of plasma insulin and insulin-like growth factor-I with indices of
androgen transport and metabolism in women with polycystic ovary syndrome. Clin Endocrinol (Oxf). 35:253–257.
5. Robinson S, Kiddy D, Gelding SV, et al. 1993 The relationship of insulin
insensitivity to menstrual pattern in women with hyperandrogenism and
polycystic ovaries. Clin Endocrinol (Oxf). 39:351–355.
6. Polson DW, Kiddy DS, Mason HD, Franks S. 1989 Induction of ovulation with
clomiphene citrate in women: the difference between responders and nonresponders. Fertil Steril. 51:30 –34.
7. Kiddy DS, Sharp PS, White DM, et al. 1990 Differences in clinical and
endocrine features between obese and non-obese subjects with polycystic
ovary syndrome: an analysis of 263 consecutive cases. Clin Endocrinol (Oxf).
32:213–220.
8. Hamilton-Fairley D, Kiddy D, Watson H, Paterson C, Franks S. 1992 Association of moderate obesity with a poor pregnancy outcome in women with
polycystic ovary syndrome treated with low dose gonadotrophin. Br J Obstet
Gynaecol. 99:128 –131.
9. Kiddy DS, Hamilton-Fairley D, Bush A, et al. 1992 Improvement in endocrine
and ovarian function during dietary treatment of obese women with polycystic
ovary syndrome. Clin Endocrinol (Oxf). 36:105–111.
10. Pasquali R, Antenucci D, Casimirri F, et al. 1989 Clinical and hormonal
characteristics of obese amenorrheic hyperandrogenic women before and after
weight loss. J Clin Endocrinol Metab. 68:173–179.
11. Nestler JE, Jakubowicz DJ. 1996 Decreases in ovarian cytochrome P450c17a
activity and serum free testosterone after reduction of insulin secretion in
polycystic ovary syndrome. N Engl J Med. 335:617– 623.
12. Dunaif A, Scott D, Finegood D, Quintana B, Whitcomb R. 1996 The insulinsensitizing agent troglitazone improves metabolic and reproductive abnormalities in the polycystic ovary syndrome. J Clin Endocrinol Metab.
81:3299 –3306.
13. Willis D, Franks S. 1995 Insulin action in human granulosa cells from normal
3991
and polycystic ovaries is mediated by the insulin receptor and not the type-I
insulin-like growth factor receptor. J Clin Endocrinol Metab. 80:3788 –3790.
14. Willis DS, Mason HD, Gilling-Smith C, Franks S. 1996 Modulation by insulin
of follicle stimulating hormone and luteinising hormone actions in human
granulosa cells of normal and polycystic ovaries. J Clin Endocrinol Metab.
81:302–309.
15. Adams J, Franks S, Polson DW, et al. 1985 Multifollicular ovaries: clinical and
endocrine features and response to pulsatile gonadotrophin releasing hormone. Lancet. 2:1375–1379.
16. Mason HD, Willis DS, Beard RW, Winston RML, Margara R, Franks S. 1994
Estradiol production by granulosa cells of normal and polycystic ovaries
(PCO): relationship to menstrual history and concentrations of gonadotropins
and sex steroids in follicular fluid. J Clin Endocrinol Metab. 79:1355–1360.
17. Gilling-Smith C, Willis DS, Beard RW, Franks S. 1994 Hypersecretion of
androstenedione by isolated thecal cells from polycystic ovaries. J Clin Endocrinol Metab. 79:1158 –1165.
17a.Ferguson KM, Hayes M, Jeffcote SL. 1992 A standardised multicentre procedure for plasma gonadotrophin radioimmunoassay. Ann Clin Biochem.
19:358 –361.
18. McNatty KP, Baird DT. 1978 Relationship between follicle-stimulating hormone, androstenedione and oestradiol in human follicular fluid. J Endocrinol.
76:527–531.
19. McNatty KP, Smith DM, Osathanondh R, Ryan KJ. 1979 The human antral
follicle: functional correlates of growth and atresia. Ann Biol Anim Biochem
Biophys. 19:1547–1558.
20. Kammerman S, Ross J. 1975 Increase in numbers of gonadotropin receptors
on granulosa cells during follicle maturation. J Clin Endocrinol Metab.
41:546 –551.
21. Nimrod A, Bedrak E, Lamprecht SA. 1977 Appearance of LH-receptors and
LH-stimulatable cyclic AMP accumulation in granulosa cells during follicular
maturation in the rat ovary. Biochem Biophys Res Commun. 78:977–984.
22. Wang C, Hsueh AJW, Erickson GF. 1981 LH stimulation of estrogen secretion
by cultured rat granulosa cells. Mol Cell Endocrinol. 24:17–28.
23. Shima K, Kitayama S, Nakano R. 1987 Gonadotropin binding sites in human
ovarian follicles and corpora lutea during the menstrual cycle. Obstet Gynecol.
69:800 – 806.
24. Shaw HJ, Hillier SG, Hodges JK. 1989 Developmental changes in luteinizing
hormone/human chorionic gonadotropin steroidogenic responsiveness in
marmoset granulosa cells: effects of follicle stimulating hormone and androgens. Endocrinology. 124:1669 –1677.
25. Baird DT. 1987 A model for follicular selection and ovulation: lessons from
superovulation. J Steroid Biochem. 27:15–23.
26. Hillier SG. 1994 Current concepts of the roles of follicle stimulating hormone
and luteinizing hormone in folliculogenesis. Hum Reprod. 9:188 –191.
27. Friedrich F, Kemeter P, Salser H, Breitenecker G. 1975 Ovulation inhibition
with human chorionic gonadotrophin. Acta Endocrinol (Copenh). 78:332–342.
28. Chappel SC, Howles C. 1991 Reevaluation of the roles of luteinizing hormone
and follicle-stimulating hormone in the ovulatory process. Hum Reprod.
6:1206 –1212.
29. Shoham Z, Jacobs HS, Insler V. 1993 Luteinizing hormone: its role, mechanism of action and detrimental effects when hypersecreted during the follicular
phase. Fertil Steril. 59:1153–1161.
30. McNatty KP, Sawers RS. 1975 Relationship between the endocrine environment within the Graafian follicle and the subsequent rate of progesterone
secretion by human granulosa cells in vitro. J Endocrinol. 66:391– 400.
31. Yong EL, Baird DT, Yates R, Reichert Jr LE, Hillier SG. 1992 Hormonal
regulation of the growth and steroidogenic function of human granulosa cells.
J Clin Endocrinol Metab. 74:842– 849.
32. Yong EL, Baird DT, Hillier SG. 1992 Mediation of gonadotrophin-stimulated
growth and differentiation of human granulosa cells by adenosine-39,59-monophosphate: one molecule, two messages. Clin Endocrinol (Oxf). 37:51–58.
33. McNatty KP, Smith DM, Makris A, Osathanondh R, Ryan KJ. 1979 The
microenvironment of the human antral follicle: interrelationships among the
steroid levels in antral fluid, the population of granulosa cells, and the status
of the oocyte in vivo and in vitro. J Clin Endocrinol Metab. 49:851– 860.
34. Frishman GN, Luciano AA, Peluso JJ. 1992 Effect of the ratio of folliclestimulating hormone to luteinizing hormone on rat granulosa cell proliferation
and oestradiol-17b secretion. Hum Reprod. 7:1073–1078.
35. Holte J, Bergh T, Gennarelli G, Wide L. 1994 The independent effects of
polycystic ovary syndrome and obesity on serum concentrations of gonadotrophins and sex steroids in premenopausal women. Clin Endocrinol (Oxf).
41:473– 81.
36. Hillier SG, de Zwart FA. 1982 Androgen/antiandrogen modulation of cyclic
AMP-induced steroidogenesis during granulosa cell differentiation in tissue
culture. Mol Cell Endocrinol. 28:347–361.
37. Hillier SG. 1987 Intrafollicular paracrine function of ovarian androgen. J
Steroid Biochem. 27:351–357.
38. Agarwal SK, Judd HL, Magoffin DA. 1996 A mechanism for the suppression
of estrogen production in polycystic ovary syndrome. J Clin Endocrinol Metab.
81:3686 –3691.
39. Giudice LC. 1992 Insulin-like growth factors, and ovarian follicular development. Endocr Rev. 13:641– 669.