Neural connections

There are numerous and extensive neural pathways connecting the hypothalamus with the rest of the brain. The majority of afferent hypothalamic nerve fibres run in the lateral hypothalamic areas, whilst efferent pathways are more medially placed. One important efferent connection is the supraopticohypophyseal nerve tract carrying fibres from the supraoptic and paraventricular nuclei to the infundibular process of the pituitary, whilst other fibres carry hypothalamic-releasing or -inhibiting factors from the medial and basal parts of the hypothalamus to the anterior pituitary (Figure 15.2).

Figure 15.2 Representation of neurochemical interactions which are important in the control of gonadotrophin-releasing hormone (GnRH) secretion.

Gonadotrophin-releasing hormone (GnRH)-secreting neurones appear in the medial olfactory placode and enter the brain with the nervus terminalis, a cranial nerve that projects from the nose to the septal preoptic nuclei in the brain. By migration during embryogenesis, the cells, between 1000 and 3000 GnRH-producing neurones, predominantly settle in the arcuate nucleus of the hypothalamus (Schwanzel-Fukunda et al 1989). Failure of this migration has been shown to result in Kallmann’s syndrome; a disorder associated with an absence of GnRH secretion and a defect of smell-anosmia (a failure of both olfactory axonal and GnRH neuronal migration from the olfactory placode). A 5–7-fold increased frequency of this condition is found in males, indicating that an X-linked transmission is the most common, although autosomal-dominant and autosomal-recessive modes of transmission have also been established (Waldstreicher et al 1996).

The mutations responsible for this syndrome result in the failure to produce a protein, homologous to members of the fibronectin family, responsible for cell adhesion and protease inhibition that is necessary for migration of these neurones (Bick et al 1992). GnRH neurones have cilia — as do olfactory epithelial cells in the nose — and the olfactory origin and structural similarity of these cells suggest an evolution from reproduction controlled by pheromones.

GnRH

In 1971, Schally et al isolated pure preparations of porcine LH-releasing hormone (LHRH) from hypothalamic extracts. Subsequently, its structure was discovered and synthesis was achieved (Matsuo et al 1971a, b). The finding of FSH-releasing activity of this LHRH led to the hypothesis of a single hypothalamic-releasing hormone, GnRH, controlling secretion of both LH and FSH from the pituitary gland, with the suggestion that sex steroids might play a role in modulating the proportions of LH and FSH released. The amino acid sequence of GnRH is shown in Figure 15.3; it is a decapeptide secreted with a large pre- and postprecursor molecule.

Figure 15.3 The structure of pre–pro-gonadotrophin-releasing hormone (GnRH). Highlighted is the decapeptide of the active molecule GnRH (molecular weight 1181). Sites of cleavage from the gonadotrophin-associated peptide are shown, as well as the main sites of enzymatic degradation of GnRH by endo- and carboxyamide peptidases in the pituitary.

GnRH neurone system

The GnRH neurone system has been mapped in detail using primarily immunocytochemical methods. The GnRH neurones are not grouped into specific nuclei but form a loose network in several anatomical divisions. However, GnRH neurone bodies are found principally in two areas: the preoptic anterior hypothalamic area and the tuberal hypothalamus, particularly the arcuate nucleus and periventricular nucleus. Axons from GnRH neurones project to many sites in the brain; the most distinct tract is from the medial basal hypothalamus to the median eminence, where extensive plexuses of boutons are found on the primary portal vessels. GnRH, then, has ready and direct access to the anterior pituitary gonadotroph cells via the portal capillary plexus (see Figure 15.2). There are also numerous projections of GnRH-secreting neurones to the limbic system and the circumventricular organ, other than the median eminence. The role of these connections is currently unknown, but they may connect with other cells or GnRH may bind to different receptors (type II) which may modulate sexual behaviour or sexual arousal.

The GnRH terminals in the median eminence remain outside the blood–brain barrier and can therefore be exposed to chemical agents within the general circulation. The GnRH neurones themselves have receptors for a number of neurotransmitters, in common with other neurones, and release them in their contacts with other neurones. In addition, their activity and release of GnRH can be influenced by GnRH agonists and antagonists. Exposure to GnRH agonists causes a greater release of GnRH but at less frequent pulse intervals. Exposure to GnRH antagonists induces a slower and non-pulsatile release of GnRH. Hence, GnRH neurones have an internal feedback loop for control of GnRH release through receptors for their own secretory product.

It has been estimated that there may be as few as 2000–3000 GnRH neurones dispersed within the hypothalamus, located predominantly in the preoptic and mediobasal areas. Considerable reduction in number can occur without affecting the pulsatile release of gonadotrophins. The pulsatile nature of gonadotrophin release is dependent upon small numbers of GnRH neurones working in synergy; the so-called ‘pulse generators’.

Regulation of gonadotrophin secretion by GnRH

The GnRH gene sequence was first isolated by Seeburg and Adelman in 1984. The GnRH decapeptide is derived from the post-translational processing of a larger precursor molecule that has been termed ‘pre–pro-GnRH’. This appears to be a tripartite structure with a preceding 23-amino acid sequence joined to the decapeptide GnRH, which is then attached via a 3-amino acid sequence, glycine–lycine–arginine (GLY–LYS–ARG), to a 56-amino acid terminal peptide, which is termed the ‘gonadotrophin-associated peptide’ (GAP). The post-GnRH-decapeptide GLY–LYS–ARG 3-amino acid section is an important site for proteolytic processing. GAP itself is thought to have some prolactin-inhibiting properties (Figure 15.3). GnRH genes are encoded from a single gene located on the shorter arm of chromosome 8.

Using radioimmunoassays, GnRH has been demonstrated in the hypophyseal portal blood from a number of animal species (Carmel et al 1976). Electrical stimulation of the preoptic area of the brain in female rats on the day of pro-oestrus increases the GnRH concentration in portal blood, and the stimulus induces a marked release of LH from the anterior pituitary. In contrast, administration of antibodies against GnRH prevents this electrically stimulated LH release. These data provide evidence favouring a cause-and-effect relationship between GnRH release by the hypothalamus and LH release by the anterior pituitary.

It is now firmly established that GnRH can stimulate the secretion of both LH and FSH in animals and humans. Following intravenous administration of synthetic GnRH, a significant rise in serum LH, and sometimes FSH, will be seen within 5 min, reaching a peak within approximately 30 min, but FSH peaks are often delayed further. LH release has a linear log-dose relationship up to doses of 250 µg but no such relationship can be found for FSH.

In the female, the magnitude of gonadotrophin release, particularly LH, in individuals varies with the stage of the menstrual cycle, being greatest in the preovulatory phase, less marked in the luteal phase and least in the follicular phase of any individual cycle (Yen et al 1972, Shaw et al 1974) (Figure 15.4).

Figure 15.4 Luteinizing hormone (LH) and follicle-stimulating hormone (FSH) release following 100 mg gonadotrophin-releasing hormone at different phases of the same menstrual cycle in six normal women (mean ± SD).

From Shaw RW, Butt WR, London DR, Marshall JC 1974 Variation in response to synthetic luteinizing hormone-releasing hormone (LHRH) at different phases of the same menstrual cycle in normal women. Journal of Obstetrics and Gynaecology of the British Commonwealth 81:632–639.

GnRH is thus the humoral link between the neural and endocrine components controlling LH and FSH release.

Mechanism of action of GnRH on pituitary cells

The first step in the action of GnRH on the pituitary gonadotroph is recognition of a specific receptor. GnRH receptor complexes often form clusters and become internalized, then undergo degradation in the lysosomes. The receptor fragments then pass back rapidly to the surface of the cell. This recycling process is causally related to upregulation of the receptor by GnRH. GnRH receptors tend to be of 60 kDa, to be glycoproteins and to have a transmembrane character of a complex nature with seven transmembrane domains. A negatively charged domain interacts predominantly with ARG in position 8 of the GnRH molecule.

The transmembrane domains have many formats which are common to humans and other species, and ligand-binding sites are fairly superficial within the receptor. In the human, the C-terminal tail is lost on the type 1 receptor (as found in the pituitary). Thus, once GnRH binds to the receptor, decoupling does not occur rapidly. This may be beneficial in allowing protracted LH release to occur.

Various GnRH agonists and antagonists have different binding ligands. An understanding of the site and mechanisms of binding to the GnRH receptor have allowed specific structural changes to be made to GnRH analogues to produce more potent pharmacological agents and to attempt to develop non-peptide antagonists.

Prolonged exposure to GnRH leads to suppression of LH release, called ‘downregulation’, which is associated with reduced numbers of GnRH receptors. This phenomenon is vitally important in understanding the mechanism of action of the GnRH analogues.

There are three principal positive actions of GnRH on gonadotrophin secretion:

The binding of GnRH to its receptor induces a complex series of intracellular responses, which result in hormone secretion and biosynthesis of the α and β subunits of LH and FSH. In addition, dimerization of α and β subunits and the glycosylation processes are induced. The mechanism of action of GnRH is depicted in Figure 15.5. Within seconds of GnRH binding to and activating GnRH receptors on the pituitary gonadotrophs, intracellular free Ca²⁺ concentrations increase. This Ca²⁺ is initially mobilized from intracellular stores (e.g. endoplasmic reticulum) but also, to maintain sustained LH release, extracellular Ca²⁺ enters the gonadotroph through receptor-regulated voltage-dependent Ca²⁺ channels.

Figure 15.5 Schematic representation of gonadotrophin-releasing hormone (GnRH)-receptor signal activation. GTP, guanosine triphosphate; PLC, phospholipase-C enzyme, AMP, adenosine monophosphate; cAMP, cyclic AMP; P, phosphate group.

The initial mobilization of intracellular Ca²⁺ is induced by inositol triphosphate, released as a consequence of receptor activation of the membrane-bound phospholipase-C enzyme. Diacylglycerol is also released by the action of phospholipase-C and in turn activates the phosphorylating enzyme protein kinase C. The adenyl cyclase complex is also stimulated and cyclic adenosine monophosphate (cAMP) is generated. Ca²⁺, protein kinase C and cAMP then interact to stimulate release of stored LH and FSH and subsequent biosynthesis (for review, see Clayton 1989).

GnRH agonist analogues

The current generation of GnRH analogues in clinical use are predominantly agonists. GnRH is degraded by peptidases in the pituitary and hypothalamus which cleave the native decapeptide at the Gly⁶-Leu⁷ bond and at position 10. Substitution at position 6 with another amino acid renders the analogue less susceptible to enzymatic breakdown, resulting in a half-life 2.5–6 times longer than the native GnRH, whilst still allowing the analogue to bind to the pituitary gonadotroph GnRH receptor site. Several agonistic analogues are available for clinical use (see Table 15.2). Following an initial few days of increased secretion of LH and FSH after the first administration of a GnRH agonist, downregulation of the GnRH receptor is achieved with subsequently suppressed release of LH and FSH, and hence a reduced stimulus to the ovary resulting in failure of follicle growth and reduced ovarian steroidogenesis.

GnRH antagonist analogues

GnRH antagonists are characterized by multiple modification of the amino acid sequence of native GnRH; predominantly the pyroglutamic and glycine termini (position 1 and 10) and deletion and substitution of hydrophobic D amino acids at positions 2 and 3 (see Table 15.2).

GnRH antagonists bind to the GnRH receptor without affecting its internalization and initiating gonadotrophin synthesis and release. The GnRH antagonists have the advantage over agonists of immediately reducing circulating gonadotrophin levels with rapid reversal on withdrawal. However, to date, the compounds available are of relatively low bioactivity and hydrophobic, requiring repeated daily administration. Their clinical roles are mainly restricted to their use to prevent premature luteinization in assisted conception cycles.

The wide-ranging group of clinical conditions in which GnRH, GnRH agonists or GnRH antagonists have found application are listed in Box 15.1. These are discussed in greater depth in Chapters 22 and 32–34.

Dopamine

The hypothalamic tuberoinfundibular dopaminergic pathway is formed by neurones, with cell bodies located in the arcuate nucleus and axons which project to the external layer of the median eminence in close juxtaposition to portal vessels. The coexistence of dopamine- and GnRH-containing axons in the same region of the median eminence suggests the possibility of dopaminergic involvement in the control of gonadotrophin secretion.

The addition of dopamine to pituitaries coincubated with hypothalamic fragments increases the release of LH, while the addition of phentolamine, an α-receptor blocker, prevents dopamine-induced LH release. These early in-vitro experiments suggested that the hormonal background was capable of modifying the response to dopamine, since it seemed ineffective in ovariectomized animals or during oestrus or dioestrus day 1 of the oestrous cycle. Dopamine was more effective at pro-oestrus or in oestrogen- and progesterone-primed rats (McCann et al 1974, Ojeda and McCann 1978).

In humans, the inhibitory role of dopamine and its agonists on LH, as well as that of prolactin release, has been demonstrated (LeBlanc et al 1976, Lachelin et al 1977). Elevated levels of prolactin can also stimulate dopamine turnover in the hypothalamus, and it is postulated that the stimulated dopamine secretion alters GnRH secretion and hence reduces FSH and LH release in situations of hyperprolactinaemia. Hence, dopamine in the human may principally have an inhibitory effect on GnRH secretion.

The contradictory roles played by dopamine in GnRH release are, in all likelihood, the consequence of more than one action of dopamine on the GnRH-secreting neurone. The steroid environment appears to modify the components involved in the dopaminergic control of GnRH secretion, with oestrogen appearing to affect the population of excitatory or inhibitory dopamine receptors, and suggesting that the feedback control of GnRH output by oestrogen is partly exerted at a hypothalamic level by reducing dopamine neuronal activity.

Noradrenaline

Most experimental evidence supports a stimulatory role for noradrenaline in the control of gonadotrophin release. Turnover of hypothalamic noradrenaline is increased during the preovulatory surge of gonadotrophins at pro-oestrus, and noradrenaline synthesis in the anterior hypothalamus is enhanced in ovariectomized rats. These effects on GnRH secretion appear to be mediated by α receptors, since phenoxybenzamine, an α blocker, suppresses the postcastration rise in gonadotrophins in male rats, and phentolamine blocks the pulsatile release of LH in ovariectomized monkeys (for review, see McCann and Ojeda 1976).

Selective blockade of noradrenaline synthesis prevents the preovulatory LH surge and that induced by gonadal steroids; the above data suggest that noradrenergic terminals in the preoptic or anterior hypothalamic area synapse with GnRH neurones involved in the control of the preovulatory surge of gonadotrophins.

Most of the cell bodies that synthesize noradrenaline are located in the mesencephalon and lower brainstem, and these cells also synthesize serotonin. The axons from these cells ascend into the medial forebrain bundle to terminate in various brain structures including the hypothalamus. The probable mode of action of catecholamines is to influence the frequency of GnRH release (see Figure 15.7).

Figure 15.7 Opposing roles of dopamine and noradrenaline on the release of gonadotrophin-releasing hormone (GnRH).

Serotonin

High concentrations of serotonin are found in the median eminence, with most of the serotonin-containing neurones originating from the raphe nucleus in the midbrain–pons area. Present evidence shows that serotonin plays a predominantly inhibitory role in gonadotrophin release (see McCann and Ojeda 1976).

Neuropeptide Y

Gonadal steroids regulate the secretion and gene expression of neuropeptide Y in hypothalamic neurones (Sahu et al 1992). Neuropeptide Y stimulates pulsatile release of GnRH and potentiates gonadotrophin response to GnRH (Pau et al 1995). In the absence of oestrogens, neuropeptide Y inhibits gonadotrophin secretion, and increased amounts of neuropeptide Y are found in cases of under-nutrition and the cerebrospinal fluid of women with anexoria and bulimia nervosa. These findings explain a link between under-nutrition and suppressed LH and FSH secretion in those conditions.

Endogenous opioids

Endogenous opioids play a central role in the neural control of gonadotrophin secretion by way of an inhibitory effect on hypothalamic GnRH secretion. These are a fascinating group of peptides, with ‘endorphins’ being the name coined to denote a substance with morphine-like action, of endogenous origin, in the brain. Endorphin production is regulated by gene transcription, and since these are precursor peptides, all opioids are derived from three precursor peptides:

A single injection of morphine, administered to oophorectomized monkeys, brings about immediate cessation of GnRH pulse generation (Yen et al 1985). Hypothalamic opioidergic neurones are found in the arcuate nucleus of the medial basal hypothalamus, in close contact with GnRH neurones. The administration of an opioid antagonist, naloxone, produces an increased frequency and amplitude of GnRH and LH secretion. These changes are most marked in the luteal phase of the cycle, and it is thought that the negative feedback of oestrogen may partly be effected through opioid-induced inhibition of GnRH secretion (Ropert et al 1981), as is the negative feedback of progesterone on GnRH secretion.

Changes in opioid tone seem to mediate the hypogonadotrophic state seen in hyperprolactinaemia, exercise and other causes of hypothalamic amenorrhoea. Treatment of patients with hypothalamic amenorrhoea (who have suppressed GnRH pulsatile secretion) with naltrexone — an opioid receptor blocker — allows return of pulsatile LH and FSH secretion and return of ovulation (Wildt et al 1993).

Corticotrophin-releasing hormone

Experimental studies indicate that corticotrophin-releasing hormone inhibits hypothalamic GnRH secretion both directly and by augmenting endogenous opioid secretion. Corticotrophin-releasing hormone infusions inhibit gonadotrophin release in primates (Xiao et al 1989), and suppress the electrophysiological activity of the GnRH pulse generator (Knobil 1989). Women with hypothalamic amenorrhoea have increased levels of cortisol in their circulation, suggesting that this is the mechanism by which stress influences the hypothalamic–pituitary–ovarian axis.

Leptin

Leptin, identified in 1994, is a 167-amino acid protein transcribed by the ob gene. The human leptin gene is on chromosome 7q31, and leptin is mainly produced in white adipose tissue with a small amount from brown adipose tissue. In humans, leptin concentrations increase with the onset of puberty, and low levels may contribute to the amenorrhoea seen in anorexia nervosa. Leptin only acts in the hypothalamus where it can produce rapid synaptic modulatory effects and increase GnRH release from the medial basal hypothalamus. Nocturnal changes in LH pulse parameters are associated temporarily with the rise in plasma leptin at night. These fluctuations in leptin are synchronized with serum LH changes. Leptin’s predominant role is to act as an adipostat, having the capacity to improve glucose homeostasis and to stimulate energy utilization by increasing thermogenic activity and capacity.

Prolactin

Prolactin gene expression occurs on the lactotrophes of the anterior pituitary gland, in the myometrium and decidualized endometrium. Differences in mRNA in these sites indicate that there are differences in prolactin gene regulation, and only that secreted by the pituitary gland is likely to be relevant in feedback on the hypothalamic–pituitary–ovarian axis. Transcription of the prolactin gene is regulated by a transcription factor which binds to the 5’ promoter region, and this is also involved in growth hormone and TSH secretion. The main function of prolactin is in lactogenesis, and the secretion of prolactin is under the inhibitory control of hypothalamic dopamine released into the portal circulation. In addition to direct inhibition of prolactin gene expression, dopamine binding to the D₂ receptor inhibits lactotrophe development and growth.

Several factors exert a stimulatory effect on prolactin secretion, particularly thyrotrophin-releasing hormine, vasoactive intestinal peptide and oestrogen, whilst endogenous opioids inhibit release, probably by modulating the dopaminergic neurones.

Elevated circulating levels of prolactin result in reduced pulse amplitude and frequency of gonadotrophin release with resultant anovulation during lactation and in conditions which cause hyperprolactinaemia.

The intrapituitary paracrine–autocrine system

Growth factors and cytokines provide a paracrine–autocrine system which regulates cell development and replication, and pituitary hormone synthesis and secretion. As with other organs in the body, the pituitary contains epidermal growth factors, fibroblast growth factors, insulin-like growth factors, interleukins, activin, inhibin, endothelin and many others. Interactions between all of these substances constitute a complex interplay, but one group of compounds are better understood — those of the activin-inhibin and follistatin substances (see Figure 15.8).

Figure 15.8 Paracrine and autocrine control of follicle-stimulating hormone (FSH) and release and function. LH, luteinizing hormone.

Inhibin

Inhibin is secreted by granulosa cells in the ovary and selectively inhibits FSH but not LH secretion. Inhibin consists of two peptides — the α and β subunits — linked by a disulphide bond. Two forms of inhibin have been purified — inhibin A and inhibin B with identical α subunits but separate, although similar, β subunits. Messenger RNA for the α and β subunits is found in granulosa cells but also in pituitary gonadotrophes. Whilst inhibin predominantly suppresses FSH secretion, limiting the numbers of follicles recruited at the start of a cycle, cells actively synthesizing LH respond to inhibin by increasing GnRH receptor numbers. Inhibin A reaches peak levels in the luteal phase and falls in concentration as the corpus luteum undergoes luteolysis.

Activin

Activin is produced by granulosa cells but is also present in pituitary gonadotropes. It contains two subunits identical to the β subunits of inhibin A and B, but some variants of the β subunits have also been identified, although their role and function are not thought to be important. Activin augments the secretion of FSH and inhibits prolactin, ACTH and growth hormone response (Corrigan et al 1991, Blumenfeld 2001). Activin increases pituitary responsiveness to GnRH by enhancing GnRH receptor formation, and its effects are blocked by inhibin and follistatin.

Follistatin

This is a peptide secreted by a variety of pituitary cells including gonodatrophes. It inhibits FSH synthesis and secretion, and the FSH response to GnRH by binding to activin and thus decreasing the activity of activin (Besecke et al 1997).

Overview of GnRH secretion

GnRH thus stimulates gonadotrophin synthesis and secretion as well as activin-enhancing and follistatin-suppressing GnRH activity. Numerous other neurotransmitters and ovarian steroids also interplay in altering and modifying pituitary responsiveness to GnRH, which results in the various changes on pulse amplitude and frequency observed throughout the menstrual cycle.

Negative feedback

Negative feedback control or inhibition of pituitary LH and FSH release has been postulated since 1932, when Moore and Price considered that the ovary and adenohypophysis were linked in a rigid system of hormonal interactions. The quantitative relationship between ovarian steroids and gonadotrophin release can be demonstrated by disturbing the negative feedback loop by oophorectomy which produces, over a period of days, an increase in circulating LH and FSH; this reaches a plateau at approximately 3 weeks with levels which are some 10 times higher than preoperative values. Alternatively, the administration of exogenous oestrogens to oophorectomized or postmenopausal women will result in rapid suppression of elevated circulating gonadotrophin levels (Figure 15.9).

Figure 15.9 The negative feedback effect of oestrogen on serum luteinizing hormone (LH) and follicle-stimulating hormone (FSH) in a postmenopausal woman.

Source: Shaw 1975, unpublished data.

The negative feedback changes result from both a direct pituitary site of action of oestradiol, with a decrease in sensitivity of the gonadotroph to GnRH (McCann et al 1974), and an action within the hypothalamus and a decrease in GnRH secretion, possibly via increased inhibitory dopaminergic and opiate activity.

The threshold for the negative feedback action of oestrogen is set to bring about suppression of gonadotrophin release with relatively small increases in oestradiol-17β in the normal female. This negative feedback loop is the main factor which maintains the relatively low basal concentrations of plasma LH and FSH in the normal female. Circulating levels of oestradiol-17β within the range of 100–200 pmol/l will suppress the early follicular phase gonadotrophin rise which initiates follicular development.

A negative feedback effect of progesterone on gonadotrophin secretion is now well established. Whilst progesterone, even in large doses, has little effect on baseline LH release, it can suppress the ovulatory surge of LH, as demonstrated in human females administered synthetic progestogens (Larsson-Cohn et al 1972), and oestrogen-induced positive feedback surges cannot be produced during the luteal phase in women (Shaw 1975). The principal negative feedback action of progesterone is thus upon the midcycle gonadotrophin surge, and it may be responsible for its short 24-h duration. It also seems likely that progesterone is an important factor in the reduced frequency of gonadotrophin pulses observed during the luteal phase of the cycle compared with their frequency in the follicular phase (see below).

Positive feedback

The fact that oestrogen may stimulate (positive feedback) rather than inhibit gonadotrophin release under certain circumstances was first proposed by Hohlweg and Junkmann (1932). Their proposal has since been substantiated by numerous experimental reports in animals and humans. Under physiological conditions, the positive feedback only operates in females; it is brought about by oestrogen and appears to be an essential component in producing the midcycle ovulatory surge of gonadotrophins. Administration of oestradiol-17β to females during the early or midfollicular phase of the cycle will induce a surge of gonadotrophins (Yen et al 1974, Shaw 1975) (Figure 15.10), but treatment with the same doses of oestrogen during the midfollicular phase induces a far greater release of LH than during the early follicular phase (Yen et al 1974). Studies on the dynamics of this positive feedback response to oestrogen, observed in greatest detail in the rhesus monkey, demonstrate an activation delay of some 32–48 h from the commencement of oestrogen administration until the onset of the positive-feedback-induced gonadotrophin surge, a minimum threshold level to be exceeded and a strength-duration aspect of the stimulus (Karsch et al 1973).

Figure 15.10 The positive feedback effect of exogenous oestrogen (E₂ = 200 µg ethinyl oestradiol/day) on gonadotrophin release; qualitative differences in the early and midfollicular phase of the cycle. •, luteinizing hormone; , follicle-stimulating hormone.

Reproduced with permission from: Yen SSC, van den Berg G, Tsai CC, Siler T 1974 Causal relationship between the hormonal variables in the menstrual cycle. In: Ferin M et al (eds) Biorhythms and Human Reproduction. John Wiley, New York, pp 219–238.

Oestrogen elicits gonadotrophin release by increasing pituitary responsiveness to GnRH and possibly by stimulating increased GnRH secretion by the hypothalamus. In the normally menstruating female, oestrogen pretreatment produces an initial suppressive action on pituitary responsiveness (Shaw et al 1975a) followed by a later augmenting action which is both concentration and duration dependent (Shaw et al 1975a, Young and Jaffe 1976). The augmenting effect of oestrogen on GnRH pituitary responsiveness is demonstrated in Figure 15.11.

Figure 15.11 Oestrogen augmentation of pituitary response to 100 mg gonadotrophin-releasing hormone bolus in four women receiving 2.5 mg oestradiol benzoate IM, 2 h after initial control response on day 4 of cycle and retested 48 h later on day 6 of cycle.

Source: Shaw RW 1975 A Study of Hypothalamic–Pituitary–Gonadal Relationships in the Female. MD Thesis, Birmingham University, Birmingham.

These data and others suggest that the midcycle oestrogen-induced surge of gonadotrophins may occur without the need for any increased output of hypothalamic GnRH, and indeed, this occurs in patients with endogenous GnRH deficiency receiving pulsatile GnRH treatment at a constant rate.

Progesterone by itself does not appear to exert a positive feedback effect. However, when administered to females in whom the pituitary has undergone either endogenously induced or exogenously administered oestrogen priming, progesterone can induce increased pituitary responsiveness to GnRH (Shaw et al 1975b; Figure 15.12). Since circulating progesterone levels are increasing significantly during the periovulatory period, this action may be of importance in determining the magnitude and duration of the midcycle gonadotrophin surge.

Figure 15.12 Effect of progesterone pretreatment (12.5 mg IM) on luteinizing hormone (LH) response to 100 mg gonadotrophin-releasing hormone IV during the early and midfollicular phases of the cycle, showing the increased priming effect of oestrogen. C, control; T, after progesterone.

From Shaw RW, Butt WR, London DR 1975b The effect of progesterone on FSH and LH response to LH-RH in normal women. Clinical Endocrinology 4:543–550.

Follicular phase

In the early follicular phase, both the immediately releasable and the reserve pool of gonadotrophins are at a minimum. The increased FSH release, responsible for initiating follicle development, must indicate an increased output of GnRH, with the lowering of negative feedback action in the presence of low levels of oestrogen and progestogen. As follicles develop, oestrogen levels rise and the negative feedback action of oestrogen on GnRH increases, suppressing FSH levels. Increasing levels of inhibin from the rapidly dividing granulosa cells of the dominant follicle further suppress FSH secretion at a pituitary level.

With these progressive increases in oestradiol throughout the midfollicular phase, the quantitative estimates of the primary, immediately releasable, pool of gonadotrophins increase slightly, whilst the reserve pool increases greatly, thus demonstrating the augmentation action of oestradiol primarily upon the reserve pool. Since there is no marked increase in circulating gonadotrophin levels during this phase, secretion of GnRH must be minimal or else there is evidence of impedance of GnRH sensitivity by oestrogen.

Mechanism for preovulatory gonadotrophin surge

During the late follicular phase, there is an increase in the amount of oestradiol secreted by the ovary. Under this influence, the sensitivity of the gonadotroph to GnRH eventually reaches a phase when GnRH can exert its full self-priming action. The consequence is the transference of gonadotrophins from the secondary reserve pool to the releasable pool. The increased pituitary responsiveness to GnRH may be further enhanced by the slight progesterone rise which can affect its action on a fully oestrogen-primed anterior pituitary. These changes culminate in the production of the ovulatory gonadotrophin surge. It is possible that these events could occur even if the gonadotrophs were exposed to a constant level of GnRH. However, an increased secretion of GnRH, as reported in rhesus monkeys and rats (Sarker et al 1976), at midcycle, which would act synergistically with the changes in pituitary sensitivity, also seems likely in the human.

The LH and FSH surges begin abruptly (LH levels doubling over 2 h) and are temporarily associated with attainment of peak oestradiol-17β levels. The mean duration of the LH surge is 48 h, with an ascending limb of 14 h which is accompanied by a decline in oestradiol-17β and 17-hydroxyprogesterone concentrations but a sustained rise in inhibin levels.

The ascending limb of LH is followed by a peak plateau of gonadotrophin levels lasting some 14 h and a transient levelling of progesterone concentrations. The descending limb is long, lasting approximately 20 h, and accompanying this is a second rise in progesterone, a further decline in oestradiol-17β and 17-hydroxyprogesterone, and a rise in inhibin levels.

The concentration of inhibin during the periovulatory interval is not correlated with oestradiol-17β or progesterone changes. It may merely reflect the release from follicular fluid of stored inhibin.

Ovulation occurs 1–2 h before the final phase of progesterone rise or 35–44 h from the onset of the LH surge.

Luteal phase

The significantly lower basal gonadotrophin secretion in the face of high pituitary capacity during the midluteal phase suggests that endogenous GnRH should be very low.

A progressive decrease in sensitivity and reserve characterizes pituitary function during the late luteal phase and into the early follicular phase of the next cycle. This is probably due to a progressive decline in oestrogen and progesterone on which sensitivity and reserve are dependent. The role played by the proposed ovarian inhibin on preferentially controlling FSH secretion has still to be completely determined.

It is therefore apparent that the functional state of the pituitary gonadotroph as a target cell is ultimately determined by the modulating effect of ovarian steroid hormones via their influence on the gonadotroph’s sensitivity and reserves, and upon the hypophysiotrophic effect of GnRH.

With such a complex inter-related control mechanism, it is perhaps not surprising that many drugs which affect neurotransmitters, ill health or associated endocrine disorders can disrupt normal hypothalamic–pituitary–ovarian function, resulting in disordered follicular growth and suppression of ovulation and in severe cases cause amenorrhoea.