Dominance: a term assigned to the status of the follicle destined to ovulate given its presumed key role in regulating the size of the ovulatory quota.
Established ovarian regulators: a group of mostly endocrine but at times autocrine principles, the
indispensability of which to ovarian function has been repeatedly and convincingly demonstrated.
Pituitary gonadotropins represent the most striking example of an established ovarian regulator.
Putative intraovarian regulator: a group of generally peptidergic principles the highly regionalized
and exquisitely timed expression of which is presumed to account for the physiologic phenomena that
cannot be fully attributed to conventional endocrine concepts. In this connection, it is hoped that
improved understanding of putative intraovarian regulators may provide the necessary clues to elucidate
the precise cellular mechanism(s) responsible for the apparent differential fate of distinct follicular
Recruitment: the process wherein the follicle departs from the resting pool to begin a well-
characterized pattern of growth and development. Recruitment, although obligatory, does not guarantee
ovulation. Stated differently, recruitment is necessary but not sufficient for ovulation to occur.
Selection: the final winnowing of the maturing follicular cohort by atresia down to a size equal to the
species-characteristic ovulatory quota.
The ovary, an ever-changing tissue, is a multicompartmental organ with a broad range of distinct
biologic properties. Responding to cyclic pituitary gonadotropin secretion, the various follicular
compartments interact in a highly integrated and seemingly programmed manner. All of this, of course,
is designed to subserve a single central objective; i.e., the generation of a mature fertilizable ovum for the
consequent preservation of the species. At
the heart of the ovarian life cycle is the follicle recognized as the fundamental functional unit of the ovary
since the middle of the sixteenth century.
The ovarian life cycle begins at the most unlikely location; i.e., the wall of the yolk sac and the ventral
wall of the hindgut near the origin of the allantoic evagination. It is here that the primordial germ cells
originate, either from or among the primitive endodermal cells some time toward the end of the third
week of gestation. This, in turn, is followed by migration of the germ cell elements to the primitive
gonadal folds during weeks 3 to 5 of life.
This remarkable translocation is accompanied by a steady increase in cell number through mitotic
divisions. Locomotion is thought to be accomplished by ameboid movements, the use of pseudopodia,
and most certainly, some form of chemotactic guidance.
Upon arrival at the genital ridge by the fifth week of gestation, the premeiotic germ cells, now referred to
as oogonia, continue to multiply as they settle down. During the subsequent two weeks of intrauterine
life, weeks 5 to 7 of gestation (often referred to as the "indifferent stage"), the primordial gonadal
structure constitutes no more than a bulge on the medial aspect of the urogenital ridge.
From this point on, the oogonial endowment is subject to three simultaneous ongoing processes:
mitosis, meiosis, and atresia (degeneration). Meiosis not only converts oogonia into primary oocytes
well before actual follicle formation, but unlike mitosis, also provides temporary protection from
oogonial atresia, thereby allowing the germ cells to invest themselves with granulosa cells and to form
primordial follicles. As a result of the combined impact of mitosis counterbalanced by atresia, the
number of germ cells peaks at 6 to 7 x 106 by 20 weeks of gestation, at which time two-thirds of the
total germ cells are intrameiotic dictyate primary oocytes while the remaining third can still be viewed as
oogonial. Some of the former have in the interim invested themselves with a single layer of spindle-
shaped (non-cuboidal) primordial (pre)granulosa cells, thereby giving rise to primordial follicles, the
formation of which begins around 16 weeks of gestation.
From midgestion onward, however, relentless and irreversible attrition progressively diminishes the
ovarian germ cell endowment by way of follicular (rather than oogonial) atresia, which begins around
month six of gestation and continues throughout life. Ultimately, some 50 years later, what has been
referred to as the oocytic "gene bank" is finally exhausted. Consequently, newborn female infants enter
life, still far from realizing reproductive potential, having lost as much as 80% of their germ cell
endowment. By the onset of puberty, virtually 95% of all follicles have been lost, only 400 to 500 of
which (i.e., < 1% of the total) will in fact ovulate in the course of a reproductive life span.
While little information exists at this time regarding the morphogenic principles responsible for follicular
organization, it is quite certain that formation of primordial follicles (to end no later than six months
postpartum), the first step in follicular development, is entirely gonadotropin-independent. Although
other factors are undoubtedly at play, it is virtually certain that even the earliest phases of follicular
development beyond the primordial follicle stage ar gonadotropin-dependent. Once recruited,
primordial follicles migrate towards the medullary region of the ovary for further development.
The next phase in follicular development, the so-called slow growth phase, is concerned with the
conversion of primordial follicles 60 um in diameter to primary follicles and eventually to mature
secondary but still preantral follicles 120 um in diameter. The process begins when the spindle-shaped
granulosa cell precursors of some primordial follicles differentiate into a single layer of cuboidal cells
surrounding a primary oocyte, thereby yielding primary follicles. Thereafter, proliferation of primary
follicular granulosa cells gives rise to multiple cellular layers, thereby yielding a preantral and ultimately
antral "secondary follicle," the maximal granulosa cell endowment of which is estimated at 600. It is at
this point that the granulosa cells become physiologically coupled by gap junctions. The resultant
electrical coupling yields an expanded, yet integrated and functional, syncytium concerned with
metabolic examination change and the transport of diffusible low molecular weight substances, thereby
compensating for the otherwise avascular intrafollicular environment. Moreover, the granulosa cells
extend cytoplasmic processes to form gestational gap-junction-like unions with the plasma membrane of
the oocyte. Undoubtedly, it is this latter communication system which is responsible in large measure for
the tight control exerted by the cumulus granulosa cells on the resumption of meiosis by the enclosed
Although the early "theca interna" has in fact been acquired at the end of the primary follicle stage, the
"theca externa" is a characteristic of the secondary follicle forming only as the follicle expands and
compresses surrounding stroma. Whereas the theca interna cells assume an epithelioid appearance and
characteristics of steroidogenic cells, the theca externa, in turn, retains its spindle-shaped configuration,
thereby merging with adjacent stromal cells.
It is at this juncture that the secondary, still pre-antral follicle embarks on an 85-day journey, spanning
three ovulatory cycles, during which a secondary follicle, 120 um in diameter, will be converted into a
Graafian preovulatory follicle, 20 mm in diameter. The first leg of the journey, the so-called accelerated
growth phase, constitutes the folliculogenic segment wherein preantral secondary follicles 120 um in
diameter are converted into antral follicles, 2 mm in diameter. This growth phase is characterized by a
600-fold increase in granulosa cell endowment concurrent with a greater than 15-fold increase in overall
follicular diameter. This overall increase in follicular size is accomplished not only through granulosa cell
proliferation but also through progressive enlargement of the antrum (central follicular fluid-filled cavity),
thereby establishing secondary antral follicles.
The term recruitment is employed here to indicate that a follicle has entered the final growth trajectory, a
well-characterized pattern of growth and development. It is the luteal pool of
2 mm secondary antral follicles which constitutes the launching pad from which the follicles destined to
ovulate in the next cycle will be recruited. Indeed, follicles must go through the terminal exponential
growth phase during which time 2 mm follicles achieve preovulatory (Graafian) status (and a 20 mm
diameter) at which point the oocyte occupies an eccentric position, surrounded by several layers of
cumulus granulosa cells.
Importantly, it is during this last phase of folliculogenesis that follicular selection is completed. This term
implies the final winnowing of the maturing but not quite yet dominant follicular cohort, by atresia, down
to a size equal to the species-characteristic ovulatory quota. In the human, follicular selection is
presumed to occur during the first five days of the cycle at a time when the leading follicular diameter is
5 to 10 mm.
The term dominance refers to the status of the follicle destined to ovulate, given its presumed key role in
regulating the size of the ovulatory quota. It is generally agreed in the human that a selected follicle
becomes dominant about a week before ovulation; i.e., as early as days 5 to 7 of the cycle at a time
when follicular diameter is around 10 mm. Only the dominant follicle can at this point in time boast
detectable levels of FSH in its follicular fluid. Expectedly, this same follicle also displays significant
follicular levels of estradiol.
In the end, it is the dominant follicle which, under the influence of the midcycle LH surge, undergoes
dramatic transformations designed to effect further oocyte maturation as well as follicular rupture. All
told, it is the rupture of the follicle which concludes the life cycle of the follicle proper while initiating the
life cycle of its successor, the corpus luteum.
As midcycle approaches, a dramatic rise is noted in the circulating levels of estradiol followed in turn by
an LH (and to a lesser extent FSH) surge, the ability of which to trigger follicular rupture is well
established. It is the midcycle gonadotropin surge that marks the end of the follicular phase of the cycle
and precedes actual rupture by as much as 36 hours. For reasons not well understood but possibly
because of unique microenvironmental circumstances, one (rarely, more than one) follicle ovulates and
gives rise to a corpus luteum during each menstrual cycle.
Mechanically, ovulation consists of rapid follicular enlargement followed by protrusion of the follicle
from the surface of the ovarian cortex. Ultimately, rupture of the follicle results in the extrusion of an
oocyte-cumulus complex. Fortuitous endoscopic visualization of the ovary around the time of ovulation
reveals that elevation of a conical "stigma" on the surface of the protruding follicle precedes rupture.
Rupture of this stigma is accompanied by gentle, rather than explosive, expulsion of the oocyte and
antral fluid, suggesting that the latter is not under high pressure.
After ovulation, the dominant follicle reorganizes to become the corpus luteum. Thus, following rupture
of the follicle, capillaries and fibroblasts from the surrounding stroma proliferate and penetrate the basal
lamina. This rapid vascularization of the corpus luteum may be guided by angiogenic factor(s) readily
detected in the follicular fluid. Concurrently, the mural granulosa cells undergo morphologic changes
collectively referred to as "luteinization." These latter cells, the surrounding theca-interstitial cells, and
the invading vasculature intermingle to give rise to a corpus luteum. Clearly, it is this endocrine gland
which is the major source of sex steroid hormones secreted by the ovary during the postovulatory phase
of the cycle. An important aspect of this phenomenon is the penetration of the follicle basement
membrane by blood vessels, thereby providing the granulosa/luteal cells with circulating levels of LDL.
Normally, the functional span of the corpus luteum is 14 + 2 days. Thereafter, the corpus luteum
spontaneously regresses, to be replaced (unless pregnancy occurs) at least five cycles later by an
avascular scar referred to as the "corpus albicans." The mechanisms underlying luteolysis remain
unclear. However, there is little doubt as to the central role of LH in the maintenance of corpus luteum
function. Thus, withdrawal of oligohydramnios support under a variety of experimental circumstances
has virtually invariably resulted in luteal demise. However, in the event of an intervening pregnancy,
hCG secreted by the fetal trophoblast maintains the ability of the corpus luteum to elaborate
progesterone, thereby enabling the maintenance of early gestation until the luteoplacental shift.
Preantral granulosa cells are predominantly targeted by FSH. Indeed, a negligible number of LH
receptors is observed in preantral granulosa cells. At that point the binding of LH is confined to theca-
interstitial cells. Importantly, however, granulosa cells of antral follicles appear capable of binding both
LH and FSH. Thus, in contrast to the presence of FSH receptors in granulosa cells from follicles of all
sizes, LH receptors are found only in granulosa cells of large preovulatory follicles. These observations
are in keeping with the notion that the ontogenetic acquisition of LH receptors is under the influence of
Both LH and FSH hormonal action appears to require the intermediacy of the membrane-associated
enzyme adenylate cyclase. Indeed, it is generally accepted that gonadotropin-mediated stimulation of
adenylate cyclase results in the conversion of intracellular ATP to cAMP. The latter, in turn, is thought
to bind to the regulatory subunit of a protein kinase (commonly referred to as A-kinase) whereupon the
catalytic subunit of the enzyme is activated and dissociated. The latter, in turn, phosphorylates key
intracellular proteins central to the signal transduction sequence. However, the exact nature of the
proteins involved remains unknown at this time.
Granulosa cells are the cellular source of the two most important ovarian steroids, estradiol and
progesterone. Although the granulosa cells and their luteinized counterparts are capable of producing
progesterone independent of other ovarian cell types, the biosynthesis of estrogens requires cooperation
between the granulosa cells and their thecal neighbors. The participation of these two cell types and of
the two gonadotropins (FSH and LH) in ovarian estrogen biosynthesis underlies the concept of the two
cell/two gonadotropin hypothesis, an integrative process required for ovarian estrogen biosynthesis.
According to this view, theca-derived, LH-dependent, aromatizable androgens (androstenedione and
testosterone) are acted upon by FSH-inducible granulosa cell aromatase activity. A broader view of
this concept could and probably should allow its extension to include intercellular exchanges of other
steroidogenic substrates (e.g., C21 progestins).
The granulosa (like the theca-interstitial) cell is amply endowed to carry out progestin biosynthesis.
Central to this process is the availability of abundant supplies of cholesterol which serves as the starting
material for the steroidogenic cascade. Recent studies have shown that cholesterol used for steroid
hormone production is derived primarily from circulating serum low-density lipoprotein (LDL) rather
than from de novo cellular biosynthesis from acetate. LDL particles are known to bind to specific
membrane receptors, the LDL-receptor complexes entering the cell by receptor-mediated endocytosis.
The resultant free cholesterol is re-esterified and is stored in the cytoplasm in lipid droplets. Faced with
steroidogenic demands, the cholesterol ester is hydrolyzed and the free cholesterol transported to
mitochondria for standard steroidogenic processing. Accordingly, cholesterol is converted to
pregnenolone by way of the rate-limiting mitochondrial enzyme cholesterol side chain cleavage. The
subsequent conversion of pregnenolone to progesterone occurs relatively readily by virtue of the relative
abundance of the cytoplasmic enzymes 3 beta-hydroxysteroid dehydrogenase/D5,D4-isomerase.
There is little doubt that the granulosa and theca-interstitial cells are capable of elaborating a large
number of proteins as assessed by gel fractionation analysis. It is equally clear that the identity of most
of the proteins elaborated remains a mystery at this time. On the other hand, a measurable number of
readily identifiable proteins has been studied. Many, such as steroidogenic enzymes and cell surface
receptors, are quite self-evident. Yet others are briefly discussed below.
Inhibin, an FSH-inducible (32kDa) protein, constitutes a unique granulosa cell marker,
the functional role of which in reproductive physiology is under active investigation.
Structurally, inhibin is a hetero dimer comprised of a common alpha-subunit (18kDa) but
different beta-subunits (14kDa). Both forms (a/BA and a/BB) of inhibin (A and B
respectively) possess similar physiological properties. Although inhibin is likely to play an
endocrine role by inhibiting pituitary gonadotropin release, recent studies indicate that
inhibin may also play a local intraovarian role.
Activin, unlike inhibin, is comprised of dimers of the beta subunits of inhibin (BA/BB or
BA/BA). Although possibly active at the level of the hypothalamic-pituitary unit, granulosa
cell-derived activin has also been shown to enhance the FSH-supported induction of
granulosa cell LH receptors.
Follistatin, which is a recent addition, constitutes a single-chain polypeptide (315 amino
acids) originally isolated from porcine follicular fluid. Although structurally distinct from
both inhibin and activin, this FSH-inducible granulosa cell-derived polypeptide appears
to suppress the release of pituitary FSH, but not LH, in a manner reminiscent of inhibin.
The potential relevance of follistatin to ovarian physiology, if any, remains unknown at
Despite its exposure to high circulating levels of gonadotropins, the postmenopausal ovary is an
atrophic, yellowish, lusterless structure with a wrinkled surface weighing less than 10 gm.
Microscopically, the cortex is thin and usually devoid of follicles (11.25; formerly 11.14). Although
devoid of follicles, the menopausal ovary is not a defunct endocrine organ. Indeed, analysis of
peripheral and ovarian vein blood from postmenopausal women indicates that the postmenopausal
ovary secretes predominantly androstenedione and testosterone. In fact, the concentrations of
testosterone and androstenedione in ovarian venous effluents of postmenopausal women are 15 and 4
times higher, respectively, than their peripheral venous levels.
Occasionally, the postmenopausal ovarian cortex shows evidence of stromal hyperplasia. When
stromal hyperplasia is florid, the ovary may be enlarged, consisting almost entirely of hyperplastic
stromal nodules. In such cases, the lipid-rich luteinized cells of the hyperplastic stroma resemble the
theca interna cells of the follicle. Thus, ovaries with stromal hyperthecosis may produce enough
androgens to result in circulating testosterone levels within the male range, hirsutism, and virilization.
The medulla of the postmenopausal ovary is large in relation to the cortex, comprising corpora albicantia
and candicantia traversed by sclerosed blood vessels. Functionally, the most important medullary
component may well be the hilar cell comprising groups of large epithelioid cells closely connected to
bundles of non-myelinated nerve fibers and small vessels. Histochemically similar to the interstitial cells
of the testes, hilar cells are generally assumed to display considerable steroidogenic potential.
Rarely do hilar cells give rise to functional neoplasms; i.e., hilus cell tumors. The latter usually produce
excess amounts of androgens leading to signs and symptoms of virilism. However, signs and symptoms
of estrogen excess may also be evident in circumstances characterized by significant peripheral
Given the inevitable hypoestrogenic state consequent to the cessation of ovarian function, several key
complications may ensue. These include urogenital atrophy, hot flashes, osteoporosis, and increased
cardiovascular morbidity and mortality. These complications, most of which are partly, if not fully,
traceable to estrogen deficiency, are most appropriately managed by the provision of estrogen
replacement therapy. This, for the most part, can be done by providing estrogen for some or all of the
calendar month. Estrogen, in turn, may have to be supplemented by a course of progestin for women
with an intact uterus in whom protection of the endometrial lining is essential. If unchecked, persistent
unopposed estrogenic stimulation may lead to endometrial hyperplasia and even endometrial cancer.
There is little doubt that in the absence of contraindications to estrogen replacement therapy, all
estrogen-insufficient menopausal individuals are in principle eligible for estrogen replacement therapy.
Although the latter is associated with small, albeit uncertain, risks, current consensus favors the notion
that the benefits far outweigh whatever risks may be associated with this therapeutic approach.
Optimal ovarian estrogen biosynthesis is contingent upon the cooperation of the two
gonadotropins (LH and FSH) and the two ovarian somatic cell types (granulosa and theca). In
contrast, progesterone biosynthesis is primarily LH-dependent and is carried out at the level of
the granulosa-lutein cell.
The pronounced progestational capabilities of the corpus luteum reflect its highly vascular
nature, a phenomenon due to the breaching of the follicular basement membrane at the time of
ovulation along with neovascularization of the former follicular apparatus.
FSH reception appears to constitute an early feature of the granulosa cell. Consequently, it is
the FSH receptor that provides the granulosa cell with a window to the outside world through
which other signaling systems can be acquired.
Contrary to conventional wisdom, the indispensability of estrogens to intraovarian physiology is
now being challenged. A strong body of evidence would suggest that primate/human follicle
may not depend on estrogen for growth and maturation.
The ovary itself may, in fact, play a zeitgeber(German - "timegiver") role during the menstrual
cycle. A time-keeping function subserved by the activities of the cyclic structures of the
dominant ovary. The 28-day menstrual cycle is thus the result of the intrinsic life span of the
cyclic ovarian dominant structure and not the result of time changes dictated by the brain or
pituitary. The dominant follicle thus determines the length of the follicular phase; the corpus
luteum determines the length of the luteal phase.