Human Reproduction Vol.17, No.1 pp.

8891, 2002
CASE REPORT
The role of LH and FSH in ovarian androgen secretion and
ovarian follicular development: Clinical studies in a patient
with isolated FSH deciency and multicystic ovaries
Randall B.Barnes
1,5
, Anne B.Namnoum
2
, Robert L.Roseneld
3
and Lawrence C.Layman
4
Departments of Obstetrics and Gynecology,
1
University of Chicago, Chicago, IL 60637,
2
Emory University, Atlanta, GA 30322,
3
Department of Pediatrics and Medicine, University of Chicago, Chicago, IL 60637,
4
Department of Obstetrics and Gynecology,
Medical College of Georgia, Augusta, GA 30912, USA
5
To whom correspondence should be addressed at: Dept of Obstetrics and Gynecology, University of Chicago, 5841 Maryland
Avenue, Chicago, IL 60637, USA. E-mail: rbarnes@babies.bsd.uchicago.edu
Inactivating mutations have proven to be instructive in elucidating the role of FSH in human ovarian function. We
performed a detailed reproductive endocrine evaluation of a patient with inactivating mutations in the FSH
-subunit gene who was hypo-estrogenic and had LH excess. The patient underwent a pelvic ultrasound and
overnight frequent blood sampling followed by a human chorionic gonadotrophin (HCG) stimulation test. One
month later she received human recombinant FSH, followed 24 h later by a second HCG stimulation test. Despite
a mean LH serum concentration and LH pulse characteristics typical for polycystic ovaries (PCOS), baseline and
dexamethasone-suppressed free testosterone were lownormal. The administration of HCGled to minimal stimulation
of 17-hydroxyprogesterone and androgens. The patient had multicystic ovaries containing follicles 35 mm in
diameter and responded to FSH with prompt increases in estradiol and inhibin B. There were no clinical or
laboratory consequences of LH excess in this FSH-decient woman. These ndings support the hypothesis that
excessive LH stimulation alone does not cause ovarian hyperandrogenism. We also found that follicular development
was present in the absence of FSH. These antral follicles had apparently developed normally, since estradiol and
inhibin B increased promptly after FSH administration.
Key words: androgens/FSH deciency/LH excess/ovary
Introduction
Inactivating mutations have proven to be instructive in elucidat-
ing the role of FSH in human ovarian function (Layman and
McDonough, 2000; Themmen and Huhtaniemi, 2000). Our
studies of a patient with compound heterozygous mutations of
the FSH -subunit gene (Layman et al., 1997), undetectable
serum FSH and elevated serum LH have elucidated two aspects
of FSH action which challenge common wisdom. We report
here evidence that FSH is not necessary for the development
of small, healthy antral follicles readily responsive to FSH,
which contrasts with studies in FSH-knockout mice (Kumar
et al., 1997; Dierich et al., 1998). However, FSH is necessary
for ovarian theca cell function, contrary to expectations from
the 2-gonadotrophin, 2-cell model of ovarian steroidogenesis
(Barnes et al., 2000). The latter nding has important
implications for theories about the role of LH excess in the
pathogenesis of polycystic ovary syndrome (PCOS).
88 European Society of Human Reproduction and Embryology
Case report
Subject
The patient presented at 16 years of age with primary
amenorrhoea and absent breast development. Her baseline
and gonadotrophin-releasing hormone (GnRH)-stimulated
FSH concentrations were undetectable by immunoassay, while
LH concentrations at baseline (30 mIU/ml) and post-GnRH
(150 mIU/ml) were elevated. The patient was subsequently
treated with estrogen and underwent normal breast develop-
ment. Analysis of the FSH -subunit gene demonstrated that
she was a compound heterozygote for two different mutations,
a two base-pair deletion in exon three at codon 61 (Val61X)
inherited from her mother, and a missense mutation changing
a cysteine to a glycine at codon 51 (Cys51Gly) inherited from
her father. When these mutations were stably transfected into
Chinese hamster ovary cells, they demonstrated no measurable
FSH immuno- or bioactivity (Layman et al., 1997).
Clinical studies in a patient with FSH deciency
Figure 1. Outline of clinical research center protocols.
Dexamethasone, 0.5 mg four times daily, was started on day 1 and
continued until the blood sample was drawn 24 h after HCG.
Reproductive endocrine evaluation
The patient underwent a detailed reproductive endocrine
evaluation at 21 years of age in the University of Chicago
General Clinical Research Center (GCRC), after dis-
continuing the oral contraceptive pill for one month and giving
informed consent. The study protocols were approved by the
University of Chicago Institutional Review Board. Her height
was 155 cm and her weight was 53.2 kg. There was no
hirsutism (Ferriman and Gallway score of 4; normal 8),
breasts and pubic hair were Tanner stage 5, and the pelvic
examination was normal. An initial blood sample was drawn for
immunoactive FSH, LH and free testosterone. The patient then
received dexamethasone 0.5 mg four times daily for 4 days
prior to and continuing throughout the endocrine evaluation
to suppress adrenal function. Following admission to the
GCRC, she underwent a transvaginal pelvic ultrasound fol-
lowed by blood sampling every 10 min from 7.00 p.m. to 6.00
a.m. Serum immunoactive FSH and LH were determined for
each sample; inhibin A and B, bioactive FSH and free -
subunit were determined in a pooled sample. The following
morning, a human chorionic gonadotrophin (HCG) test was
performed. A blood sample was drawn for baseline testoster-
one, free testosterone and estradiol. The following steroid
intermediates were also determined: 17-hydroxyprogesterone
(17-Prog), androstenedione, 17-hydroxypregnenolone and
dehydroepiandrosterone. HCG 5000 IU was then administered
i.m. and blood was drawn 24 h later for repeat steroid
measurements.
The patient remained off any sex steroids for another month,
then returned to the GCRC for a second HCG study which
was like the rst except she received 300 IU of recombinant
human FSH s.c. in the morning, 24 h prior to the HCG test.
A blood sample was drawn 24 h later for steroids and inhibin
A and B, and HCG 5000 IU was administered. She then
returned 24 h after HCG (48 h after FSH) for a nal blood
sample for steroid measurements (Figure 1).
All samples for steroids were frozen at 20C and
measured in the same assay using previously published methods
(Barnes et al., 1989; Roseneld et al., 1994). Serum immuno-
active LH and FSH were measured using a -subunit-specic
assay (Dela, Wallach, Finland; lower limit of sensitivity
0.15 mIU/ml for LH and 0.2 mIU/ml for FSH). Bioactive
FSH, inhibin A, inhibin B and free -subunit were measured
using previously published methods (Christin-Maitre et al.,
89
Figure 2. Transvaginal ultrasound of the left ovary before FSH
treatment. The ovary is fusiform in shape, measuring ~61.5 cm.
The stroma is scant but there are multiple follicles throughout the
ovary, the largest measuring 5 mm. The right ovary (not shown)
was similar to the left.
1996; Lavoie et al., 1998; Welt et al., 1999). LH pulse analysis
was performed using the ULTRA program (Van Cauter and
Copinschi, 1981).
Results
The ovarian ultrasound prior to FSH treatment is shown in
Figure 2. The ovaries were multicystic containing follicles
35 mm in diameter, but unlike typical polycystic ovaries
(Adams et al., 1986), there was no dense stroma and the
follicles were scattered throughout the ovary rather than being
predominately subcapsular.
Immunoactive LHwas 46 mIU/ml and FSHwas undetectable
(0.2 mIU/ml) in the single sample drawn before dexametha-
sone. There was no immunoactive FSH detected in the samples
collected overnight (Table I), and no bioactive FSH was
detected in the pooled sample. The mean immunoactive
LH in the overnight samples was 46 mIU/ml; mean pulse
amplitude was 15.8 mIU/ml and nine pulses were detected in
11 h (Figure 3). This LH prole is similar to that reported for
PCOS (Waldstreicher et al., 1988). Free -subunit measured
in the pooled 11 h sample was in the postmenopausal range
(3501740 pg/ml; Table I).
Although LH concentrations were elevated, the patient had
no laboratory evidence of ovarian hyperandrogenism. The
patients baseline free testosterone was 6 pg/ml (normal
range 310 pg/ml; ovarian hyperandrogenism 10 pg/ml)
(Roseneld et al., 1994). After 4 days of dexamethasone
suppression the free testosterone fell to 2 pg/ml (normal range
8 pg/ml; ovarian hyperandrogenism 8 pg/ml; Roseneld
et al., 1994). The 17-Prog concentration 24 h after HCG was
40 ng/dl which was 2.7 standard deviations (SD) below the
mean in women with ovarian hyperandrogenism (288 ng/dl)
(Levrant et al., 1997) and 1.5 SD below the mean of normal
follicular phase women (139 ng/dl: Levrant et al., 1997).
One month later, during the second GCRC study,
estradiol increased substantially 24 h after FSH administration
(2880 pg/mL). Inhibin B was near the lower limit of the
normal range at baseline and rose to above the range for
R.B.Barnes et al.
Table I. Gonadotrophin and inhibin concentrations in an FSH-decient patient
FAS LH FSH Bioactive FSH Inhibin A Inhibin B
(pg/ml) (mIU/ml) (mIU/ml) (mIU/ml) (pg/ml) (pg/ml)
Baseline
a
1247 46 0.2 4 0.6 25.3
24 h after FSH 8.4 0.8 283
Normal women 103257 015 320 440 0.680 15220
a
Baseline is mean of every 10 min samples over 11 h for LH and FSH. For all others, baseline is a value of
the pooled 11 h sample. Normal values are from Christein-Maitre et al., 1996; Lavoie et al., 1998; Welt
et al., 1999.
FAS free alpha subunit.
Figure 3. LH pulse analysis. LH was measured at 10 min intervals
over 11 h. Signicant pulses are indicated by an asterisk.
menstruating women 24 h after FSH injection (Table I). In
contrast, inhibin A was present at low concentrations.
The response of testosterone and the steroid intermediates
to the two HCG tests have been previously reported (Barnes
et al., 2000). In brief, all steroid concentrations 24 h after the
second HCG injection (48 h after FSH) were greater than at
24 h after the rst HCG injection (HCG alone). The difference
between the two HCG tests was most dramatic for testosterone,
which was unaffected by the rst HCG test (12 ng/dl before
and after HCG), but increased from 16 to 41 ng/dl after the
second, FSH-primed HCG test.
Discussion
This patient, with well characterized inactivating mutations in
the FSH -subunit gene, was hypo-estrogenic with secondary
LH excess. Two features were noteworthy: antral follicular
development was present in spite of the lack of FSH, and
there were no clinical or laboratory consequences of the
LH excess.
The point at which FSH is necessary for follicular develop-
ment in the human ovary is debated (Gougeon, 1996; McGee
and Hsueh, 2000). Despite having no detectable FSH, our
patient had multicystic ovaries with follicles up to 5 mm in
diameter. Similar sized follicles have been reported in some
women with FSH receptor mutations (Aittomaki et al., 1996;
Beau et al., 1998); however, those FSH receptor mutations
90
may not be completely inactivating. These ndings are in
contrast to FSH - and FSH receptor-knockout mice, in which
antral follicles are not maintained (Kumar et al., 1997; Dierich
et al., 1998). In studies of human ovarian xenografts trans-
planted into the kidney capsule of immunodecient and hypo-
gonadotrophic mice, FSH was required for the growth of
follicles beyond the two-layer granulosa cell stage (Oktay
et al., 1998), which is about the point at which the FSH
receptor gene is rst expressed in human follicles (Oktay et al.,
1997). Although the xenograft data are in contrast to our
ndings, it is likely that the endocrine and growth factor milieu
of the in-situ human ovary is very different than that in the
immunodecient, hypogonadotrophic mouse.
The prompt estradiol and inhibin B responses 24 h after
exogenous FSH suggest that our patients antral follicles
contained granulosa cells that had developed normally in the
absence of FSH. Our ndings are similar to those in two other
patients with isolated FSH deciency who ovulated and had a
successful pregnancy after ~14 days of menotrophin therapy
(Rabinowitz et al., 1979; Matthews et al., 1993). Ovulation
after a short exposure to menotrophins implies that some
healthy follicles had reached the point of recruitability without
FSH exposure, since ~14 days are required for an ovulatory
follicle to develop from follicles a few millimeters in diameter,
in contrast to the 3 months required to develop from the pre-
antral stage (Gougeon, 1996). These earlier reports and our
ndings suggest that in the complete absence of FSH stimula-
tion, human ovarian antral follicles can develop up to 5 mm
in diameter.
Our patient had no clinical or laboratory evidence of ovarian
hyperandrogenism despite a mean LH concentration, LH pulse
characteristics and ovarian follicular sizes typical for PCOS.
On ultrasound her multicystic ovaries lacked the excess stroma
of classic polycystic ovaries. Indeed, her ovaries produced
little, if any, androgen. Her baseline and dexamethasone-
suppressed free testosterone were lownormal. The administra-
tion of HCG led to minimal stimulation of 17-Prog or other
thecal cell steroids. However, we have previously shown that
exogenously administered FSH augmented her LH- and HCG-
stimulated production of testosterone and all steroids of thecal
cell origin (Barnes et al., 2000). This suggests that FSH action,
probably via granulosa cell-produced paracrine intermediates,
is necessary for thecal cells to respond to LH. One of many
such paracrine factors is inhibin B, which increased markedly
with FSH administration in this patient (Barnes 1998).
Clinical studies in a patient with FSH deciency
These ndings are relevant to the role of LH excess in the
pathogenesis of the ovarian hyperandrogenism of PCOS. It
has been reported that 75% of women with clinical evidence
of PCOS have an elevated LH concentration and 94% have
an increased LH/FSH ratio (Taylor et al., 1997). These
gonadotrophin secretory abnormalities have been thought to
play an important role in the development of the ovarian
hyperandrogenism characteristic of PCOS (Hall, 1993). In a
related study, we found that a woman with a constitutively
activating mutation of the LH receptor, identied because she
was the mother of two sons with gonadotrophin-independent
precocious puberty, had no clinical or laboratory evidence of
ovarian hyperandrogenism (Rosenthal et al., 1996). Thus,
increased LH stimulation, even in the presence of FSH, appears
to be insufcient to induce the hyperandrogenism and stromal
hyperplasia of PCOS. The ndings in the previous, as well as
the current, case report support the hypothesis that thecal cell
androgen secretion in response to excessive LH stimulation is
strictly limited by intra-ovarian factors. Ovarian hyperandro-
genism is more likely a result of escape from down-regulation
by these intra-ovarian factors than a result of elevated LH
concentrations (Ehrmann et al., 1995). Taken together, our
studies support the hypotheses that normal ovarian androgen
production depends on both FSH and LH and that excessive
ovarian androgen production is a result of abnormal intra-
ovarian regulation, and not of excessive LH stimulation.
Acknowledgements
We are greatly indebted to Patrick Sluss, Reproductive Endocrine Unit,
The Massachusetts General Hospital, Boston, MA, who performed
the FSH bioassay, FAS, inhibin A, and inhibin B assays; to Zubie
Sheikh, Department of Obstetrics and Gynecology, University of
Chicago, for performing the transvaginal ultrasound; and to Jennifer
M.Cunningham, Department of Medicine, University of Chicago, for
performing the LH pulse analysis. We also appreciate the helpful
discussions of J.Larry Jameson from the Center for Endocrinology,
Metabolism, and Molecular Medicine, Northwestern University,
Chicago, IL and William F.Crowly Jr from the Reproductive
Endocrine Unit and the National Center for Infertility Research, The
Massachusetts General Hospital, Boston, MA.
L.C.L. was supported by NIH grant support PHS NICHD HD33004;
R.L.R. was supported by NIH grant support RR-00055 (CRC).
References
Adams, J., Polson, D.W. and Franks, S. (1986) Prevalence of polycystic
ovaries in women with anovulation and idiopathic hirsutism. Br. Med. J.,
293, 355359.
Aittomaki, K., Herva, R., Stenman, U-H. et al. (1996) Clinical features of
primary ovarian failure caused by a point mutation in the follicle-
stimulating hormone receptor gene. J. Clin. Endocrinol. Metab., 81,
37223726.
Barnes, R.B. (1998) The pathogenesis of polycystic ovary syndrome: lessons
from ovarian stimulation studies. J. Endocrinol. Invest., 21, 567579.
Barnes, R.B., Roseneld, R.L., Burstein, S. et al. (1989) Pituitaryovarian
responses to nafarelin testing in the polycystic ovary syndrome. N. Engl.
J. Med., 320, 559565.
Barnes, R.B., Namnoum, A., Roseneld, R.L. et al. (2000) Effect of follicle-
stimulating hormone on ovarian androgen production in a woman with
isolated follicle-stimulating hormone deciency. N. Engl. J. Med., 343,
11971198.
91
Beau, I., Touraine, P., Meduri, G. et al. (1998) Novel phenotype related to
partial loss of function mutations of the follicle stimulating hormone
receptor. J. Clin. Invest., 102, 13521359.
Christin-Maitre, S., Taylor, A.E., Khoury, R.H. et al. (1996) Homologous
in vitro bioassay for follicle-stimulating hormone (FSH) reveals increased
FSH biological signal during the mid- to late luteal phase of the human
menstrual cycle. J. Clin. Endocrinol. Metab., 81, 20802088.
Dierich, A., Sairam, M.R., Monaco, L. et al. (1998) Impairing follicle-
stimulating hormone (FSH) signaling in vivo: targeted disruption of the
FSH receptor leads to aberrant gametogenesis and hormonal imbalance.
Proc. Natl Acad. Sci. USA, 95, 1361213617.
Ehrmann, D.A., Barnes, R.B. and Roseneld, R.L. (1995) Polycystic ovary
syndrome as a form of functional ovarian hyperandrogenism due to
dysregulation of androgen secretion. Endocr. Rev., 16, 322353.
Gougeon, A. (1996) Regulation of ovarian follicular development in primates:
facts and hypotheses. Endocr. Rev., 17, 121155.
Hall, J.E. (1993) Polycystic ovarian disease as a neuroendocrine disorder of
the female reproductive axis. Endocrinol. Metab. Clin. North Am., 22, 7592.
Kumar, T.R., Wang, Y., Lu., N. et al. (1997) Follicle-stimulating hormone is
required for ovarian follicle maturation but not male fertility. Nature Genet.,
15, 201204.
Lavoie, H.B., Martin, K.A., Taylor, E., et al. (1998) Exaggerated free alpha-
subunit levels during pulsatile gonadotropin-releasing hormone replacement
in women with idiopathic hypogonadotropic hypogonadism. J. Clin.
Endocrinol. Metab., 83, 241247.
Layman, L.C. and McDonough, P.G. (2000) Mutations of follicle stimulating
hormone-beta and its receptor in human and mouse: genotype/phenotype.
Mol. Cell. Endocrinol., 161, 917.
Layman, L.C., Lee, E.J., Peak, D.B. et al. (1997) Delayed puberty and
hypogonadism caused by mutations in the follicle-stimulating hormone -
subunit gene. N. Engl. J. Med., 337, 607611.
Levrant, S.G., Barnes, R.B. and Roseneld, R.L. (1997) A pilot study of the
human chorionic gonadotrophin test for ovarian hyperandrogenism. Hum.
Reprod., 12, 14161420.
Matthews, C.H., Borgato, S., Beck-Peccoz, P. et al. (1993) Primary
amenorrhoea and infertility due to a mutation in the -subunit of follicle-
stimulating hormone. Nature Genet., 5, 8386.
McGee, E.A. and Hsueh, A.J.W. (2000) Initial and cyclic recruitment of
ovarian follicles. Endocr. Rev., 21, 200214.
Oktay, O., Briggs D. and Gosden, R.G. (1997) Ontogeny of follicle-stimulating
hormone receptor gene expression in isolated human ovarian follicles. J.
Clin. Endocrinol. Metab., 82, 37483751.
Oktay, O., Newton H., Mullan J. et al. (1998) Development of human
primordial follicles to antral stages in SCID/hpg mice stimulated with
follicle-stimulating hormone. Hum. Reprod., 13, 11331138.
Rabinowitz, D., Benveniste, R., Lindner, J. et al. (1979) Isolated follicle-
stimulating hormone deciency revisted. N. Engl. J. Med., 300, 126128.
Roseneld, R.L., Barnes, R.B. and Ehrmann, D.A. (1994) Studies of the
nature of 17-hydroxyprogesterone hyperresponsiveness to gonadotropin-
releasing hormone agonist challenge in functional ovarian
hyperandrogenism. J. Clin. Endocrinol. Metab., 79, 16861692.
Rosenthal, I.M., Refetoff, S., Rich, B. et al. (1996) Response to acute challenge
with gonadotropin releasing hormone agonist in two half-brothers and their
mother with a constitutively activating mutation of the luteinizing hormone
receptor. J. Clin. Endocrinol. Metab., 81, 38023806.
Taylor, A.E., McCourt, B., Martin, K.A. et al. (1997) Determinants of
abnormal gonadotropin secretion in clinically dened women with polycystic
ovary syndrome. J. Clin. Endocrinol. Metab., 82, 22482256.
Themmen, A.P.N. and Huhtaniemi, I.T. (2000) Mutations of gonadotropins
and gonadotropin receptors: elucidating the physiology and pathophysiology
of pituitarygonadal function. Endocr. Rev., 21, 551583.
Van Cauter, E. and Copinschi, G. (eds) (1981) Human pituitary
hormones: circadian and episodic variations. Martinus Nijhoff, The Hague,
pp. 221235.
Waldstreicher, J., Santoro, N.F., Hall, J.E., et al. (1988) Hyperfunction of the
hypothalamicpituitary axis in women with polycystic ovarian disease:
indirect evidence for partial gonadotroph desensitization. J. Clin. Endocrinol.
Metab., 66, 165172.
Welt, C.K., Adams, J.M., Sluss, P.M. et al. (1999) Inhibin A and inhibin
B responses to gonadotropin withdrawal depends on stage of follicle
development. J. Clin. Endocrinol. Metab., 84, 21632169.
Received on February 28, 2001; accepted on August 21, 2001