Reproductive Endocrine Unit and National Center for Infertility Research, Massachusetts General Hospital, Boston, Massachusetts 02114
Address all correspondence and requests for reprints to: Frances J. Hayes, M.B., M.R.C.P.I., Reproductive Endocrine Unit and the National Center for Infertility Research, Massachusetts General Hospital, BHX 5, Fruit Street, Boston, Massachusetts 02114.
![]() |
Introduction |
---|
![]() ![]() ![]() ![]() ![]() ![]() |
---|
![]() |
The inhibin hypothesis |
---|
![]() ![]() ![]() ![]() ![]() ![]() |
---|
Inhibin is a glycoprotein hormone secreted by the Sertoli cells of the
testis and the granulosa and theca cells of the ovary. It is composed
of an -subunit disulfide-linked to one of two ß-subunits, the
ßA-subunit to form inhibin A or the
ßB-subunit to form inhibin B (1). A variety of molecular
forms are present in the circulation, including mature (32-kDa) and
partially processed
ß-dimers, as well as free
-subunit forms.
Although the free
-subunits lack biological activity, the larger
molecular mass dimers as well as inhibin A and inhibin B have been
shown to have the ability to suppress FSH secretion (4).
![]() |
Difficulties in charting inhibin physiology |
---|
![]() ![]() ![]() ![]() ![]() ![]() |
---|
An additional problem encountered in the inhibin field has been the lack of inhibin preparations in sufficient quantity and purity to administer to humans. To date, therefore, most of the evidence for the participation of inhibins in differential gonadotropin control in the human has been based upon correlative studies rather than experimental manipulations. Finally, there appears to be important species specificity in the relative role of inhibin in gonadotropin feedback; e.g. rats have a considerably lower FSH response to inhibin antibody administration than do nonhuman primates (15, 16). For all of these reasons, inferences about inhibin physiology in the human are limited and indirect.
![]() |
Inhibin physiology in the adult male |
---|
![]() ![]() ![]() ![]() ![]() ![]() |
---|
Sites of inhibin subunit gene expression: Data on the
localization of inhibin subunits in the human male are derived from
fetal and early postnatal, rather than adult, testes.
Immunocytochemical studies employing purified monoclonal antibodies
have demonstrated the presence of - and ßB-inhibin
subunits predominantly in the Sertoli cells and, to a lesser extent, in
the Leydig cells of the testis (17). Consistent with the absence of
circulating inhibin A in male serum, immunostaining against the
ßA-subunit has been negative.
Inhibin levels in normal and infertile men: Testing one of
the basic tenants of the inhibin hypothesis, namely that there should
be an inverse relationship between serum inhibin and FSH levels, has
proven difficult in intact models because gonadotropin stimulation of a
functioning testis is known to increase inhibin secretion. Thus, the
alterations in gonadal stimulation associated with normal puberty,
endogenous gonadotropin suppression, and exogenous gonadotropin
administration are all marked by parallel changes in inhibin and FSH
(18). When the inhibin hypothesis was initially tested in various
clinical models of primary and secondary hypogonadism, a reciprocal
relationship between serum inhibin and FSH was not observed; indeed,
inhibin levels were shown to be normal or even raised in men with
severe spermatogenic failure (19). This raised doubts about
the physiological relevance of inhibin in the male until it was
recognized that the Monash RIA used in these studies had
cross-reactivity of up to 288% with the biologically inactive
-subunit (7, 8).
Studies using an assay initially thought to measure total dimeric inhibin, but which was subsequently found to have only minimal cross-reactivity with recombinant inhibin B (14), showed that inhibin levels were virtually undetectable in normal and GnRH-deficient men (20). With the development of assays specific for inhibin A, Illingworth et al. demonstrated that in both normal men and men with reproductive disorders, inhibin A concentrations were too low to be of physiological relevance as an endocrine signal (21). Subsequently, a two-site ELISA for inhibin B demonstrated an inverse relationship between inhibin B and FSH across a wide range of FSH concentrations in normal and infertile men (21, 22, 23), consistent with a potential role of inhibin B as an endocrine feedback regulator of FSH secretion in the male.
Experimental studies.
Inhibin administration and immunoneutralization in the
subhuman primate: Studies in the monkey have demonstrated that
treatment of gonadectomized males with recombinant inhibin A prevents
the castrate rise in pituitary FSHß messenger ribonucleic acid (mRNA)
levels and circulating FSH concentrations (24). Although these data
initially suggested a critical role for inhibin in the regulation of
FSH secretion, it is important to note that the serum levels of inhibin
A achieved were found to be pharmacological rather than physiological
when remeasured with the inhibin A ELISA (25). In addition, although
there are presently no data to suggest that inhibin A and inhibin B
differ in potency, it is inhibin B, rather than inhibin A, that appears
to be the physiologically relevant inhibin in male monkeys and adult
men (20, 21, 25). However, despite these reservations,
immunoneutralization experiments support an endocrine role for inhibin
in the male with the demonstration that a continuous infusion of an
antiserum to the N-terminal portion of human -inhibin subunit in
male macaques results in a 2- to 3-fold increase in circulating FSH
concentrations (16).
Inhibin B levels: A number of studies have explored the
inhibin B response to a variety of hormonal manipulations in normal and
infertile men (22, 23, 26). Our group has previously chronicled the
serial changes in inhibin B during the experimental induction of sexual
maturation with exogenous GnRH in men with isolated GnRH deficiency
(22, 23). At baseline, inhibin B levels in GnRH-deficient men were
significantly lower than those in normal controls, but showed a
surprising degree of variation. Although more than one third of
patients had levels at or below the working range of the assay, several
had concentrations well within the normal adult range. Correlational
analysis revealed that two interrelated factors showed a strong
positive relationship with inhibin B: pretreatment testicular volume
and a prior history of some degree of spontaneous puberty (Fig. 1). The higher inhibin B levels observed
in patients with a history of puberty and, therefore, higher testicular
volume, suggest that prior stimulation of the testis sufficient to
induce seminiferous tubular maturation has an abiding impact on inhibin
B secretion that is sustained after sexual maturation. This hypothesis
of a developmental threshold for testicular inhibin B secretion is
supported by the pattern of inhibin B secretion observed in boys with
central precocious puberty (CPP). In this model, inhibin B
concentrations are in the normal adult range at the time of active
precocity, but during treatment with a GnRH analog at a dose sufficient
to render gonadotropins undetectable, they fall to levels that remain
easily detectable, similar to the concentrations found in
GnRH-deficient men with a history of spontaneous puberty (27).
|
|
|
As inhibin B secretion is dramatically lowered in primary testicular disorders, a recent study explored whether differences in the degree of suppression of spermatogenesis after exogenous testosterone administration could be assessed by changes in circulating concentrations of inhibin B (29). During testosterone treatment, FSH levels fell rapidly, followed by a parallel decline in inhibin B and sperm concentrations. However, there was no difference in the inhibin B levels, either before treatment or during therapy, between men rendered azoospermic and those in whom sperm continued to be present in the ejaculate. Thus, measurement of plasma inhibin B was of no value in predicting the spermatogenic response to sex steroid administration for contraceptive purposes. The observation that inhibin B levels only fell to 30% of baseline despite undetectable gonadotropin levels is consistent with our data in men with GnRH deficiency and boys treated for precocious puberty (23, 27). Similar evidence is provided by studies in the monkey, in which inhibin B concentrations in juveniles were approximately 40% higher than those in infants, despite the fact that the juvenile period is characterized by marked hypogonadotropism (25). These data provide consistent support for a gonadotropin-independent component to inhibin B secretion in the male that is abolished only in the presence of primary testicular disease, suggesting that inhibin secretion from Sertoli cells may be regulated by interaction with germ cells (30).
Response to induction of Sertoli cell damage: In a
prospective study of men with hematological malignancies treated with
chemotherapy, the dynamic relationship between inhibin B and FSH was
examined during progression from normal to severely impaired
seminiferous tubular function (12). Within 4 months of commencing
chemotherapy, inhibin B levels had fallen to 20% of the baseline
values accompanied by a reciprocal 5-fold rise in FSH, yielding an
inverse correlation between the two hormones. In contrast to the levels
of dimeric inhibin B, total immunoreactive inhibin levels measured by
the Monash RIA remained unchanged, whereas pro-C-containing inhibins
increased significantly, correlating directly with the rising FSH
level. This novel observation that a testicular insult is associated
with an increase in pro-
C-containing inhibin forms is consistent
with the in vitro demonstration that FSH stimulates free
-subunit secretion in Sertoli cell cultures (31). The reciprocal
changes seen in inhibin B and free
-subunit secretion in response to
Sertoli cell damage provide for the first time a definitive explanation
for the normal total immunoreactive inhibin levels in men with
spermatogenic failure reported in previous studies using the Monash RIA
(19).
Summary of inhibin physiology in the male: These studies indicate that inhibin B is the physiologically important form of inhibin in the human male. Although inhibin B is stimulated by FSH, there is also evidence for a gonadotropin-independent component to its regulation in men. Inhibin B appears to play a significant role in the negative feedback loop regulating FSH secretion and may provide a useful marker of Sertoli cell function.
Inhibin physiology in the female
Descriptive studies. The role of inhibin in female reproductive physiology has been explored using a number of approaches, including the determination of specific inhibin subunit mRNA expression in the ovary as well as the concentration of circulating immunoreactive inhibins across the menstrual cycle.
Sites of inhibin subunit gene expression: In a variety of
species, including the human, a changing pattern of inhibin subunit
mRNA expression has been demonstrated in ovarian granulosa, theca, and
lutein cells across the reproductive cycle (32). Expression of the
ßA-subunit is highest in the corpus luteum and the
dominant follicle; that of ßB is greatest in the
granulosa cells of antral follicles during the luteal-follicular
transition, whereas expression of the -subunit appears relatively
constant throughout follicular development after the antral stage.
Immunoreactive inhibin levels across the menstrual cycle: The Monash polyclonal RIA initially showed inhibin to be detectable throughout the normal menstrual cycle. Inhibin levels were low in the early to midfollicular phase, rose in the late follicular phase to a peak on the day of the midcycle LH surge, then fell briefly after ovulation before reaching a maximum in the midluteal phase (33). When similar studies were conducted with the two-site ELISA specific for dimeric inhibin A (13, 20, 34), the overall pattern of inhibin secretion across the menstrual cycle was similar to that seen with the Monash assay, except that the magnitude of the changes observed was greater. The peak in inhibin A in the luteal phase and its fall with luteolysis are consistent with its being a secretory product of the corpus luteum, as would be expected from the expression of the ßA-subunit in the corpus luteum (32). The inverse correlation between inhibin A and FSH levels in the luteal phase suggests that release from the negative feedback effect of inhibin may play a role in the rise in FSH and, thereby, the initiation of follicular development during the luteal-follicular transition. However, it has been suggested that the control of FSH during this critical phase may be entirely due to estradiol (35).
With the lack of change in inhibin A levels in the midfollicular phase
of the cycle, it initially appeared unlikely that inhibin was playing
any endocrine role in the fall in FSH observed at this time. However,
using the inhibin B-specific ELISA, patterns of differential secretion
of inhibin B and A secretion were subsequently documented across the
menstrual cycle (14, 34) (Fig. 4). Plasma
concentrations of inhibin B are maximal in the early and midfollicular
phases, falling in the late follicular phase before ovulation and then,
after a brief postovulatory rise, reaching a nadir in the midluteal
phase. During the luteal-follicular transition, inhibin B levels rise
rapidly, reaching a maximum 4 days after the peak in FSH. The timing of
the early follicular phase rise in plasma inhibin B and the positive
correlation with FSH (34) suggest that inhibin B is secreted by small
developing follicles in response to FSH stimulation. The source of the
peak in inhibin B on the day after ovulation is thought to represent
release from the ruptured follicle, rather than secretion from the
corpus luteum, as the latter does not express mRNA for the
ßB-subunit. The inability of the Monash assay to detect
the follicular phase rise in inhibin B secretion was subsequently
explained by the demonstration that this assay detects dimeric inhibin
A to a far greater degree than inhibin B (36). It has been suggested
that inhibin B may have an endocrine role in the fall in FSH that
occurs in the late follicular phase (14). However, studies to date have
not distinguished a potential negative feedback effect of inhibin B on
FSH secretion from that of estradiol at this cycle stage.
|
Experimental studies.
Response to GnRH analog: Children with CPP, before treatment and during GnRH agonist (GnRHa) therapy, provide a useful model of pubertal physiology in which to examine the differential regulation of gonadotropin secretion. In a study of girls with CPP, we have previously shown inhibin B levels to approximate those of women in the follicular phase, whereas inhibin A levels were at the limit of assay detection, suggesting that inhibin B is the predominant nonsteroidal modulator of FSH secretion during human female puberty (41). Treatment with GnRHa was associated with a significant fall in inhibin B to near undetectable levels in most patients, with values returning to baseline by 6 months off therapy. The response to reversible pituitary-gonadal suppression induced by GnRHa administration provides convincing evidence that inhibin B secretion during female puberty is dynamically regulated by gonadotropin stimulation of the ovary.
Response to gonadotropin stimulation: Several studies have
reported an increase in total immunoreactive inhibin levels after the
administration of exogenous gonadotropins for ovulation induction (37).
These studies have noted a correlation between the inhibin response and
the number of follicles stimulated as well as a decline in the inhibin
response with increasing reproductive age. More detailed information
has come from studies employing pituitary down-regulation with a GnRHa
before stimulation with recombinant FSH for in vitro
fertilization (IVF), as this regimen creates a hormonal milieu that
allows the impact of high doses of FSH on ovarian inhibin secretion to
be selectively elucidated. In a study of women with normal ovarian
function participating in such an IVF treatment protocol, Lockwood
et al. (42) demonstrated that both inhibin A and inhibin B
were significantly suppressed in association with pituitary
desensitization, whereas levels of pro--C were largely unaltered.
Levels of both inhibins rose markedly during stimulation with FSH. The
positive correlation observed between inhibin A levels in the late
follicular phase and the number of follicles larger than 10 mm suggests
that inhibin A may be useful as a marker of dominant follicle
development in IVF. These data indicate that ovarian production of
dimeric inhibin A and inhibin B in women is gonadotropin-dependent,
whereas the regulation of pro-
-C may have a significant
gonadotropin-independent component, consistent with our report of
extragonadal production of pro-
-C in the male (11).
Control of inhibin B by FSH in GnRH-deficient women: In the
intact model it is difficult to distinguish the relationship between
inhibin and FSH, as gonadotropin stimulation is known to stimulate
inhibin secretion, which, in turn, exerts a negative feedback on FSH.
We, therefore, chose the model of GnRH-deficient women to explore the
relationship between FSH and inhibin B secretion during the
luteal-follicular transition by manipulating FSH levels via changes in
the frequency of exogenous GnRH administration (34). The GnRH pulse
frequency was either increased from every 4 h in the late luteal
phase to every 90 min on the day of menses to mimic normal cycling
women, or it was kept constant at a late luteal phase frequency through
the first 6 days of the subsequent cycle. The slower rate of rise in
FSH observed in the group in whom the GnRH pulse frequency was
maintained at the slow luteal phase frequency was associated with
significantly lower inhibin B levels in the early follicular phase,
suggesting that there may be a critical rate of rise in FSH required
for inhibin B stimulation (Fig. 5). In
these studies the difference in the inhibin B responses to varying FSH
levels was evident before changes were apparent in estradiol,
indicating that inhibin B may be an important early prognostic
indicator of follicular response during ovulation induction
therapy.
|
Effects of inhibin administration: Administration of recombinant inhibin A to subhuman primates during the luteal phase prevents the rise in FSH generally associated with the demise of the corpus luteum, whereas administration in the early follicular phase causes a progressive decrease in FSH levels (see Ref. 34 for review). These data provide more direct evidence for an endocrine role of inhibin A in the negative feedback regulation of gonadotropin secretion, which is consistent with the inverse relationship observed between inhibin A and FSH across the luteal-follicular transition.
Summary of inhibin physiology in the female: Inhibin A and inhibin B are differentially secreted in normal women, changing as a function of ovarian follicular development and possibly acting in a negative feedback mode to modulate FSH secretion. Inhibin A is a marker of the mature follicle and the corpus luteum, and may have an endocrine role in the luteal-follicular transition where a fall in inhibin A levels permits a rise in FSH to occur. Inhibin B secretion is stimulated by FSH, and data from older cycling women suggest that it may play an endocrine role in FSH secretion, although there is as yet no clear evidence that it contributes to the dynamic changes in FSH within a cycle.
Comparison of inhibin physiology in the male and female. In
the human there appear to be important gender differences not only in
the forms of inhibin present in the circulation, but also in their
regulation by gonadotropins. Although both inhibin A and inhibin B are
present in the female, only inhibin B circulates in physiological
concentrations in the male. In addition, there is evidence for a
gonadotropin-independent component to inhibin B secretion in males, but
not in females. For example, in children with CPP treated with a GnRH
analog, inhibin B levels in boys remain easily detectable, whereas in
girls they are completely suppressed despite a similar degree of
gonadotropin inhibition in both groups. Therefore, it appears that in
the male, inhibin B secretion persists even after gonadotropin
stimulation has been withdrawn unless there is a primary testicular
defect. In the female, on the other hand, inhibin levels respond
rapidly to gonadotropin changes throughout the menstrual cycle, with
inhibin A and inhibin B falling to near-undetectable levels during the
follicular and luteal phases, respectively. The mechanism for this
differential regulation in the male and female gonad is not known. It
may relate to tissue-specific regulation of - and ß-subunit gene
expression, or it may reflect fundamental differences in the relevant
cell types in the testis and ovary. The presence of Sertoli cells is a
consistent feature of the seminiferous tubule after early testicular
development. In contrast, the highly coordinated maturation and atresia
of ovarian follicles involves dynamic changes in the cell populations
involved in inhibin synthesis and secretion. Further studies are
required to address this issue.
Summary
Although considerable strides have been made in charting the physiology and pathophysiology of inhibin in the human, further progress awaits the development of recombinant inhibin suitable for administration in human studies. Measurements of total serum inhibin, although inadequate to chart the dynamic changes associated with normal and disordered pituitary-gonadal function, have proven to be of value as indices of ovarian tumor activity. Although the roles of specific dimeric inhibin measurements in clinical practice have not been clearly established, advances in our understanding of inhibin physiology and pathophysiology in the human suggest that inhibin B may have value as a marker of Sertoli cell function in men with infertility and as a prognostic indicator in women undergoing ovulation induction therapy.
![]() |
Acknowledgments |
---|
![]() |
Footnotes |
---|
Received December 18, 1997.
Revised February 17, 1998.
Accepted February 27, 1998.
![]() |
References |
---|
![]() ![]() ![]() ![]() ![]() ![]() |
---|