Hypothalamo-Pituitary Axis

Understanding the iner-relationship between the different endocrine glands is important in practicing as a Reproductive Endocrinologist. No endocrine gland acts in isolation without affecting, or being affected by the other glands. In this field the pituitary, ovaries, thyroid and adrenal glands have special integrated roles in controlling our wellbeing and reproduction. Other glands are also important, but are outside the remit of this book. Each gland acts a subordinate to a central hypothalamic-pituitary co-ordinator. It is now evident that this central control which was once thought to be the maestro, is in fact continuously affected by its subordinate target gland. Nevertheless, the pituitary gland still remains an important part of the neuroendocrine system through which the brain controls growth, metabolism, general health and reproduction, among many other vital human functions. This is affected through different neurotramitters which are conveyed first to the hypothalamic nuclei which in turn control the pituitary gland. The hypothalamo-pituitary axis is made of two developmentally different parts called the neurohypophysis and adenohypophysis.

The neurohypophysis is developed from the ectoderm of the diencephalon (mid brain). It is made of 3 different parts:

The neural lobe is connected to the median eminence by the infundibular stem. It is made of nerve axons arising in the hypothalamus with their final terminals ending in the vicinity of small blood vessels. This neurovascular association lacks a blood brain barrier, and is drained by the posterior hypophyseal veins. The cell bodies of these nerves are located in the supraoptic and paraventricular nuclei. Accordingly, the neurohypophysis is not a true endocrine gland, and forms the terminal stop for the release of oxytocin and antidiuretic hormone. These two hormones pass down as granules in the nerve axons along the infundibular stem. The target organs for oxytocin are the breasts and uterus, where as antidiuretic hormone exerts its main effect though the collecting tubules of the kidneys.

A detailed description of the neurohypophysis structure and function is beyond the remit of this chapter, which is mainly concerned with the development, maturation, inter-relationships and aging of the hypothalamo-pituitary-ovarian axis. All respectable physiology books give detailed and illustrated manuscripts, and can be consulted when necessary.

The adenohypophysis originates during the 4^th week of fetal life from the ectoderm of Rathke’s pouch which is a diverticulum of the primitive foregut. It elongates cranially and contacts the neural diverticulum destined to form the neurohypophysis by the 5^th week. It loses its connection to the foregut by the 6^th week. The adenohypophysis is fully developed by the 16^th week of intrauterine fetal life. By then, it is tightly close to the neurohypophysis, but remains functionally different. It is also made of 3 different parts:

The pars distalis forms most the gland. It is responsible for the secretion of follicle stimulating hormone (FSH), luteinising hormone (LH), thyroid stimulating hormone (TSH), adrenocorticotrophic hormone (ACTH), growth hormone (GH), prolactin, melanocyte stimulating hormone (MSH) and endorphins. Each of these hormones secretion is controlled through both neural and humoral mechanisms. A negative feedback mechanism, affected at the level of the pituitary gland and hypothalamus, coordinate the secretion of the target endocrine glands. It also prevents overproduction of the related tropic hormones by the pituitary gland. Neural control of the pars distalis hormones is affected through neurotransmitters, which facilitate the production of these tropic hormones. Prolactin stands out as the only pituitary hormone inhibited through dopamine secretion.

The pars intermedia lies between the pars distalis and infundibular process, hence its name. It is non-functional in adults and is only seen during fetal life and in pregnant women. It is normally separated from the pars distalis by the hypophyseal cleft. The pars tuberalis covers the pituitary stalk as a collar and carries the portal vessels connecting the pars distalis with the hypothalamus. Its exact endocrine function has not been conclusively established, but was thought to have a role in the short loop feedback control mechanism of gonadotrophins secretion.

In the context of this chapter, the main script will concentrate on the hypothalamo-pituitary-ovarian axis which is the main neuroendocrine system responsible for the development of secondary female sexual characteristics and reproduction. The adrenal and thyroid glands will be discussed in more detail separately in different chapters.

Any malfunctioning at any point along this trail would lead to physical and psychological abnormalities which could affect the sexual identity, reproductive capacity and quality of life of the individual concerned. However chromosomal sex is the most important and driving force behind this sequence of developmental stages, depending whether a sperm with a Y or X chromosome fertilises the egg. This would dictate the development of the primitive gonad into a testicle or ovary respectively which in turn controls the development of the genital organs. Testicular production of testosterone and dihydrotestosterone would promote differentiation of internal as well as external male genital organs. Failure of testicular development, rather than ovarian development, would allow female external genitals differentiation from the urogenital sinus. According to genital sex assignment at birth, an individual would be brought up as a boy or a girl by the parents and society, which moulds his or her identity through sex of rearing. However, the issue of gender identity is a complex one, especially in women exposed to excessive amount of androgens during the intrauterine period of life as in cases of congenital adrenal hyperplasia.

The initial step in primitive gonadal development entails genital ridge formation by thickening of the coelomic epithelium on the medial aspect of the mesonephros. This is followed by migration of primordial germ cells from the wall of the yolk sac to the genital ridge before and during differentiation of the gonad into a testicle or an ovary. This is an autosomally controlled process at this early stage which is similar in both sexes. Further development of the primordial germ cells into the sex cords stage depends on the presence of one X chromosome. However, further development of the primitive gonads depends on the presence or absence of testicular determining factor encoded in the sex determining region gene (SRY). This is located in the short arm of the Y chromosome (Sinclair AH 1990 (1) and converts the primitive gonads into testicles. Absence of this factor would allow growth of the primitive gonads into ovaries instead. Further development beyond the primary oocytes stage depends on the presence of two X chromosomes. In the absence of a second X chromosome, the primordial follicles would undergo rapid atresia leading to degeneration of the ovaries into streak gonads. The primitive gonad would normally be destined to develop as an ovary by the 7^th or 8^th week of intrauterine fetal life.

Four different stages have been recognised leading to full maturation of the HPO axis to its adult state.

The first or developmental stage begins with the differentiation of the different parts of the axis during fetal life as described before. The system becomes functional during the second trimester with maximum secretion of gonadotrophins by the pituitary gland under the control of the fetal hypothalamus. This coincides with a maximum number of 7 million primordial follicles in the ovaries by the 24^th weeks of intrauterine life. The negative feedback mechanism controlling gonadotrophins releasing hormone (GNRH) secretion starts late during pregnancy and is marked by progressive decline in gonadotrophins secretion by the fetal pituitary gland. This phase is coupled by increased oocytes atresia and loss of ovarian primordial follicles leaving behind only one million in both ovaries at birth. Following birth and loss of the inhibitory placental steroids, the neonatal hypothalamo-pituitary unit is reactivated resulting in more gonadotrophins secretion which could exceed adult levels for about 3-4 months. This again is followed by a slow decline in gonadotrophins secretion to almost undetectable levels by the end of the second year of life.

This is followed by an inhibitory stage which extends up to the age of 8-9 years. This juvenile hypothalamic pause is characterised by minimal, if any, production of GnRH. The exact molecular changes which lead to this arrest of the hypothalamic GnRH pulse generator are not fully explored. Central neural suppression is frequently quoted as the main cause. However, there is substantial evidence now relating such arrest to an inhibitory gamma aminobutyric acid (GABA) effect on the hypothalamus. There is increased level of the hypothalamic mRNAs encoding for the enzyme glutamic acid decarboxylase, which is responsible for the production of GABA, at the time of GnRH generator arrest (El Majdoubi et al 2000 (2) and El Majdoubi et al 2000 (3)). Reduction of this inhibitory tone is associated with the onset of puberty (Mitsushima et al 1994 (4); Terasawa and Fernandez 2001 (5). A similar effect has also been related to a high level of melatonin during the inhibitory period of development (Garcia et al 2002 (6). A reduction in melatonin level to a critical value by the age of 10 years is associated with a release of this inhibitory effect, and increased production of GnRH and LH (Aleandri et al 1996 (7). Removing the ovaries during this stage would not lead to any increase in GnRH or gonadotrophins secretion, which is a reflection of the extent of such hypothalamic pause. The pituitary gland itself is less responsive to intravenous infusion or subcutaneous administration of GnRH during this inhibitory stage of life. This is a reflection of the fact that some aspects of the pituitary gland maturation might be independent of the hypothalamic GnRH pulse generation.

The pubertal stage would follow, and it signals the start of the HPO axis maturation. It is an age dependent process which follows genetically controlled CNS maturation necessary for the pulsatile release of GnRH (Ojeda 2006 (8). For some time, this process was thought to be related to adrenarche which indicates the start of adrenal androgens secretion, by the zona reticularis. There is gradual increase in adrenal androgen secretion over a period of 2 years before the onset of puberty. This is coupled with a progressive increase in the size of the zona reticularis of the adrenal cortex. However, this concept is no longer valid and other factors are becoming more evident as initiators or facilitators of puberty onset. Attaining a minimum BMI with minimum and critical body fat mass of about 22% is one such factor. This results in an increase in the level of blood leptin which is a peptide hormone produced by adipocytes. At a certain blood level threshold, leptin would facilitate the development of puberty, so long as other critical control mechanisms are operational (Mann and Plant 2002 (9). Developmentally, leptin reflects the amount of body fat and energy reserve. It signals to the brain that enough energy reserves are available for initiating reproductive function. Its level starts rising by the age of 7 - 8 years and peaks by the age of 13 – 15 years. Thereafter, the levels of serum leptin would parallel those of LH and oestradiol. Leptin has been found to induce gonadotrophins production by stimulating GnRH pulse generation. It also increases LH more than FSH by acting directly at the level of the pituitary gland. Another facilitatory factor for initiation of puberty is the kissprotein, which is encoded by the KiSS-1 gene, and its receptor GPR54. This KiSS-1/GPR54 system is an important regulator of puberty in all mammals. KiSS-1 mRNA and GPR54 mRNA are both increased in the hypothalamus at the onset of puberty, with robust expression in the region of the arcuate nucleus (Shahab et al 2005 (10).

Other important endocrine changes during early puberty include increase in the levels of growth hormone releasing factor and growth hormone (GH) itself, mainly at night time. GH has been shown to stimulate FSH induced granulosa cell differentiation. It also increases intraovarian levels of IGF-1, and enhance ovarian response to gonadotrophins. Together with IGF-1, they exert a paracrine intraovarian control on steroidogensis.

Changes in the level of leptin and KiSS-1/GPR54 mRNA transduction are coupled by a reduction in the hypothalamic GABA tone as already mentioned to coincide with the onset of puberty. All these factors, plus a genetically controlled CNS maturation lead to a release of the block on the GnRH pulse generation, allowing the appearance of large nocturnal GnRH pulses and nocturnal LH secretion, during sleep.

With final maturation of the system into the fourth or adult stage, these large nocturnal LH pulses become more frequent with smaller amplitude, probably secondary to increased nocturnal dopamine activity. A 24-hour pulsatile GnRH and LH secretion pattern forms the next step in this adult maturation stage, which is later on capped by maturation of the oestrogen positive feedback and LH surge mechanisms.

GnRH is a decapeptide (10 amino acids) produced in pulses by the median eminence of the hypothalamus, which do not necessarily correspond to follicle stimulating hormone (FSH) or luteinising hormone (LH) pulses. Development of the positive feedback mechanism indicates the final step in the maturation of the HPO axis. It might take a couple of years to develop after menarche and is one of the reasons why the early menstrual cycles are not ovulatory. It is also the first mechanism to be lost following any hypothalamic GnRH dysfunction. With this maturation step, very small GnRH pulses upregulate their own receptors in the pituitary gland without causing any LH secretion. This pattern is seen just before an LH surge which is affected by a dose and time controlled exposure of the pituitary gland to oestradiol just before ovulation. For organised follicular development, FSH and LH should be produced in tonic and cyclic patterns at different parts of the cycle. Tonic release indicates continuous production of both hormones and the cyclic production is responsible for the positive feedback mechanism. Both tonic and cyclic production are pulsatile in nature. The short first half life of LH (30 minutes) allows better perception of the pulsatile nature of its production than FSH which has a longer first half life of 60 minutes. This difference is secondary to the amount of carbohydrate moiety in the two hormones, being higher in FSH. Such differences in half life could explain the lack of synchrony between the pulse patterns of both hormones in relation to the short half life of GnRH pulses, which is 2.7 minutes. Furthermore, some GnRH pulses might not stimulate LH production due to temporary refractoriness of the pituitary gland. One important extra observation was that GnRH in the peripheral blood could be produced by the pancreas, or could leak from the organum vasculosum, which is outside the blood brain barrier.

Sustained large GnRH pulses could desensitise the pituitary gland by downregulating its own receptors on the surface of the gonadotrophs. This suppression is always preceded by a short flare period with increased blood levels of gonadotrophins and oestradiol. GnRH receptors are also found within the ovaries at the follicular level. Accordingly, sustained non-physiological doses of GnRH could interfere directly with ovarian function. They could reduce FSH induction of follicular aromatase enzymatic activity, leading to reduced oestradiol production. Furthermore, downregulation of ovarian LH receptors would also result in reduced progesterone production.

Both FSH and LH are glycoproteins with identical a chain but have different amino acid sequence in their β chains. FSH receptors are located in the granulosa cells where as LH has receptors in both granulosa and theca cells. As for other protein hormones, these receptors are located in the cell membrane and have short half life of 30 minutes, indicating rapid turnover. Only 2% of the receptors need to be occupied to initiate a local response. Follicles are responsive to FSH stimulation only when they reach or exceed 60-cell stage. The characteristic functions of FSH could be summarised as follows:

This is a very intricate process, and could easily be affected adversely leading to abnormal ovulation or even anovulation. Different events occur at the level of the hypothalamus, pituitary gland, and ovary in preparation for this event.

· The hypothalamus secrets small and rapid GnRH pulses to upregulate its own receptors at the level of the pituitary gland.

· This increase in pituitary GnRH receptors level is also associated with an increase in the pituitary storage of LH itself. The LH surge occurs when both attain a critical level, and the pituitary gland is exposed to a critical level of oestradiol for a critical period of time.

· Changes within the ovary include increased production of oestradiol before the LH surge, neovascularisation, increased levels of prostaglandins and plasmin. These are also associated with increased osmotic pressure and fluid influx into the follicle.

The actual LH surge starts at midnight. It peaks just before noon on the following day and lasts for about 48 hours, when the accumulated pituitary LH reserve is exhausted. Ovulation is expected to occur within 36-40 hours after the start of the LH surge.

For completion purpose, factors controlling the gonadotrophins negative feedback should be alluded to. Both oestradiol and progesterone could have a negative effect at the level of the hypothalamus. Increased level of oestrogen is associated with increased hypothalamic dopamines and reduced adrenergic activity. Progesterone also increases the level of hypothalamic endorphins; hence both steroids could affect GnRH pulse generation. At the level of the pituitary gland both oestrogen and progesterone could reduce gonadotrophins secretion by affecting GnRH postreceptor activity, but not the receptors themselves.

Many factors could affect the HPO axis function with variable consequences on the ovaries. The initial response would be loss of the LH surge, with further progression to inadequate ovulation, dysfunctional uterine bleeding, infrequent ovulation and finally anovulation. These factors could be environmental, neural or endocrine in nature. The vulnerability of the HPO axis is reflected by the fact that it is modulated by corticotrophins, cortisol, adrenaline, nor adrenaline dopamine, serotonin, acetylcholine and gamma amino butyric acid, just as examples.

The most important environmental factors which could affect the hypothalamo-pituitary axis are excessive weight gain or loss, bulimia even without weight change, excessive exercise, morbid stress, depression and recreational drugs. They could affect the hypothalamus in different ways and hence interfere with GnRH pulse generation. They would be discussed in more detail in different chapters in this book.

The interrelationship between the different endocrine glands makes it difficult, if not impossible, for a gynaecologist to practise reproductive medicine without thorough knowledge of these interactions. This is especially so for dysfunctions of the thyroid and adrenal glands as well as hyperprolactinaemia. More information about these inter-relations would be given in Chapters 3 and 4. Polycystic ovarian syndrome is the most common female endocrine dysfunction and would be discussed in chapter 6 in this book. Different LH and FSH pulse patterns have been described (Abdel-Gadir et al (11), but there is general agreement regarding increased LH pulse amplitude and probably pulse frequency as well. However, the main characteristics of the syndrome are hyperandrogenisation and anovulation.

Drugs both prescribed and recreational could affect the HPO axis. The most commonly prescribed ones include antipsychotic and certain anti-hypertensive drugs. On the other hand, alcohol forms the most commonly used recreational chemical or drug. It has been shown to be detrimental on pubertal development, disrupts normal menstrual function and affects postmenopausal hormone levels (Emanuele et al 2002 (12). The effect of alcohol on puberty involves both the HPO axis, growth hormone and insulin growth factor-1 activities, which are functionally interrelated. Alcohol consumption among adolescent girls between the ages of 12-18 years was shown to suppress oestrogen levels for as long as 2 weeks, even after moderate consumption (Block et al 1993 (13). Lower LH (Rettori et al 1987 (14), and growth hormone (Dees and Skelley 1990 (15) blood levels have been reported after alcohol consumption. Furthermore, the development of regular menstrual pattern after menarche was similarly affected by alcohol intake (Dees et al 2000 (16).

The effect of alcohol on oestradiol level after the menopause depended on the amount taken and whether that individual was on HRT or not. Acute alcohol exposure has been shown to cause temporary increase in oestradiol level in women taking HRT. This was thought to follow decreased conversion of oestradiol to oestrone (Purohit 2000 (17). There was no similar effect on women not using HRT. An opposite effect of chronic or high alcohol consumption was noticed on the level of oestradiol in women using HRT (Johannes et al 1997 (18). The effects of many other recreational drugs including marijuana and cocaine have also been examined both in the acute phase and after chronic use.

The reproductive episode in human beings is very short. Maximum fertility potential spans for only few years between the ages of 23-29 years. This episode is preceded and followed by times of reduced fertility potential due to immaturity, to start with, and aging thereafter of the HPO axis respectively. To be more accurate, aging of the HPO axis starts during fetal life after the 24^th week of pregnancy. Accelerated loss of the primary follicles starts at that time, and drops from 7 million to one million at birth as mentioned before. This is followed by further atresia and only 0.7 million would be found in both ovaries by the time of menarche. It is estimated that about 100 follicles would be found by the time of menopause. Such ovarian aging is also associated with increased oocytes chromosomal abnormalities after the age of 35 years. It is estimated that 1: 4 eggs would be aneuploid by the age of 35, 1: 2 by 42 years and >80% of the eggs would be similarly abnormal by the age of 44 years (R). Accordingly, ovarian aging is marked with the presence of fewer oocytes with higher chromosomal abnormalities. The major and most significant early endocrine change during the same period is a substantial fall in the circulating levels of inhibin-B, with no significant change in inhibin-A or oestradiol. This would later progress on to a decline in inhibin-A and oestradiol blood levels, and a rise in FSH level without any further change in inhibin-B (Burger et al 2002 (19)

At the same time an independent hypothalamic aging process has been documented, not related to the ovaries. This was thought to be the cause of the high FSH blood levels and changes in LH pulse pattern during the early follicular phase of regularly menstruating women in their late thirties, compared to younger women. Further arguments have also been put forward to prove this concept of hypothalamic aging including:

Unfortunately, FSH is not a very reliable marker of ovarian ageing, at least in its early stages. This is because of the following reasons:

1. It is produced in pulses and timing the blood sample within that pulse could affect the FSH level. The peak, the rough or any point in between could be represented.

2. High FSH could be seen during some but not all cycles during the early stages of the climacteric period. This pattern becomes more frequent with time, till it becomes established during all cycles.

3. High oestradiol in the early follicular phase could suppress FSH production and lead to a low or normal FSH blood level. Accordingly, oestradiol level should be tested in the same sample for endocrine coupling. A high level ≥200 pmol/l is equally important as a high FSH level in reflecting a reduced ovarian reserve. Transvaginal ultrasound examination on day 2 or 3 of the cycle might show:

a. Rapid follicular recruitment with a single advanced or large follicle. Such follicle could produce high oestrogen (>200 pmol/l) which would affect FSH level. In such cases rapid growth and maturation of the dominant follicle would lead to a short follicular phase. This explains why polymenorrhoea is the first sign of the climacteric or incipient ovarian failure.

b. Multiple follicles recruitment, with many of them growing to a medium size during the early part of the follicular phase, could lead to high oestradiol level and falsely normal FSH blood level.

Figure 1 shows 18.5 mm follicle in the left ovary on day 5 of the cycle which was diagnostic of rapid follicular recruitment. This patient had polymenorrhoea with short follicular phase. Note the lack of any activity in the right ovary.

Figure 2 demonstrates multiple follicular recruitment on the 3rd day of the cycle. The patient was 31 years old. She had one 12 mm follicle in the left ovary and 3 others in the right one measuring 14.5 mm, 12.5 mm and 10.5 mm respectively. She also had short menstrual cycles. Note the small size of the left ovary. Her FSH blood level was 6.8 IU/L and oestradiol 248 pmol/l on the same day.

More information about other conditions which may affect FSH blood level can be found in Chapter 14

Normally, a single dominant follicle would produce enough oestradiol to switch off FSH production causing demise of the other recruited ones. However, during the late 30s or early 40s one follicle would not produce enough oestradiol to switch off FSH production by the pituitary gland, during the mid follicular phase. More than one follicle would then be needed to produce that amount of oestradiol. Accordingly, two or more follicles would reach maturation and ovulate at the same time probably with a wide ovulation window, depending on their size. This explains why binovular twins are more common in women in their late 30s, than in younger ones. Accordingly, such twining is a sign of reduced HPO axis integrity with compromised negative feedback mechanism to oestradiol. It is definitely not a sign of increased or enhanced fertility potential, as commonly thought.

Figure 3 shows double ovulation in a spontaneous monitored cycle with a corpus luteum in each ovary, in a 37 year old woman. On the other hand, Figure 4 shows a dominant follicle in the right ovary and a corpus luteum in the left one. A colour Doppler copy of this picture on the back cover shows a good vascular rim around the dominant follicle in the right ovary indicating imminent ovulation in a different patient of almost similar age. This case demonstrated a wide ovulation window with two follicles ovulating at different times, which is not uncommon during natural cycles in women in their late 30s

It can be safely stated that a single normal FSH test has a very low diagnostic or even screening value for ovarian reserve according to the facts mentioned before. On the other hand a single high level >13.0 IU/L performed in a reputable laboratory would indicate reduced ovarian reserve, even if repeating the test in more than one occasion showed normal values (Scott et al (20) and Martin et al (21). Accordingly, such repetition in subsequent cycles in a woman who had a single high value is not indicated and adds unnecessary costs.

Different ways have been tried to improve the predictive value of FSH as a predictor of ovarian reserve including:

These tests have been used as an indirect measurement of the ovarian reserve or follicular pool, and to predict the clinical response to ovulation induction in women >35 years of age, who are seeking to get pregnant. They are not sensitive, and better tests are now available for that purpose. They are not necessary outside this context and could not be used to predict future fertility or the exact time of cessation of fertility (Maseelall and McGovern 2008 (22), or the age at menopause. Inhibin B blood levels and antral follicle count on day 3 of the cycle have been used for some time, but have been superseded by antimullerian hormone (Knauff EA et al, 2009 (23), which has the added value of having a non-variable level at different times of the cycle. Accordingly, the test could be conducted at any time, and not restricted to the early follicular phase. It is especially useful for the assessment of follicular pool in young women with relatively high FSH and regular periods (incipient ovarian failure) or high FSH and oligomenorrhoea (transitional ovarian failure).

Other endocrine changes which occur with aging, but are not related to the climacteric are progressive decline in the level of dehydroepiandrosterone and dehydroepiandrosterone sulphate. On the other hand, despite a 50% fall in the level of testosterone between the ages of 20-40 yeas, little if any changes occur during the transitional period before the menopause, and its level might even rise thereafter (Burger et al 2002 (24).

Younger women might need to have their ovarian reserve tested after a shorter period of infertility in the following circumstances:

It is evident that the function and integrity of the HPO axis as a single unit can be affected by many factors, from within or without the axis itself. Accordingly, knowledge of the pathophysiology of this system relies on thorough knowledge of the normal and abnormal function of the other endocrine glands, which may interact to disrupt its function. This knowledge will hopefully be provided in the different chapters included in this book. In essences, this chapter was meant to be a platform or a launching pad for the rest of the book.

1. Sinclair AH, Berta P, Palmer MS, et al. A gene from the human sex-determining region encodes a protein with homology to a conserved DNA-binding motif. Nature 1990; 346:240-244.

2. El Majdoubi M, Sahu A and Plant TM. Changes in hypothalamic gene expression associated with the arrest of pulsatile gonadotrophin-releasing hormone release during infancy in the agonadal male rhesus monkey (Macaca mulatto). Endocrinology 2000; 141: 3273-3277.

3. El Majdoubi M, Sahu A, Ramaswamy S and Plant TM. Neuropeptide Y: a hypothalamic break restraining the onset of puberty in primates. Proc Natl Acad Sci USA 2000, 97: 6179-6184.

4. Mitsushima D, Hei D and Terasawa E. Alpha aminobutyric acid is an inhibitory neurotransmitter restricting the release of luteinising hormone-releasing hormone before onset of puberty. Proc Natl Acad Sci USA 1994; 91; 395-399.

5. Terasawa E, Fernandez DL. Neurobiological mechanisms of the onset of puberty in primates. Endo Rev 2001; 22: 111-151.

6. Murcia García J, Muňoz Hoyos A, Molina Carballo A, Fernández García JM, Narbona López E and Uberos Fernández J. Puberty and melatonin. An Esp Pediatr 2002; 57(2): 121 - 126.

7. Aleandri V, Spina V and Morini A. The pineal gland and reproduction. Hum Reprod Update 1996; 2(3): 225 - 235.

8. Ojda SR, Roth C, Mungenast A, Heger S, Mastronard C, Parent AS, Lomniczi A, Jung H. Neuroendocrine mechanisms controlling female puberty: new approaches, new concepts; Int J Androl. 2006; 29(1): 256 -263.

9. Mann DR and Plant TM. Leptin and pubertal development. Semin Reprod Med 2002; 20(2): 93-102.

10.Shahab M, Mastronardi C, Seminara SB, Crowley WF and Ojeda SR. Increased hypothalamic GPR54 signalling: a potential mechanism for initiation of puberty in primates. PNAS 2005; 1026): 2129 - 2134

11.Abdel Gadir A, Khatim MS, Mowafi RS, Alnaser MI, Muharib NS and Shaw RW. Implications of ultrasonically diagnosed polycystic ovaries. II. Studies of dynamic and pulsatile hormonal patterns. 1992; 7(4): 458 - 461.

12.Emanuele MA, Wezeman F and Emanuele N. Alcohol’s effects on female reproductive function. Alcohol Res Health 2002, 26(4): 274 - 281.

13.Block GD, Yamamoto ME, Mallick A and Styche AJ. Effects on pubertal hormone by ethanol abuse in adolescents. Alcoholism: Clinical and Experimental Research 1993; 17: 505

14.Rettori V, Skelley CW, McCann SM and Dees WL. Detrimental effect of short-term ethanol exposure on reproductive function in the female rat. Biology of Reproduction 1987; 37: 1089 - 1096.

15.Dees WL and Skelley CW. Effects of ethanol during the onset of female puberty. Neuroendocrinology 1990; 51: 64-9.

16.Dees WL, Dissen GA, Hiney JK et al. Alcohol ingestion inhibits the increased secretion of puberty-related hormones in the developing female Rhesus monkey. Endocrinology 2000; 141: 1325-31.

17.Purohit V. Can alcohol promote aromatization of androgens to oestrogens? A review. Alcohol 2000; 22: 123 - 127.

18.Johannes C. Crawford S and McKinley S. The effect of alcohol and oestrogen replacement therapy (HRT) on oestrogen levels in postmenopausal women. A J Epidemiol 1997; 145: S1.

19. Burger HG, Cahir N, Roberston DM, Groome NP, Dudley E, Green A, Dennerstein L. Serum inhibin A and B fall differently as FSH rises in perimenopausal women. Clin Endocrinol (Oxf) 1998; 48(6): 809 - 813.

20.Scott RT Jr, Hofmann GE, Oehninger S, Muasher SJ. Intercycle variability of day 3 follicle-stimulating hormone level and its effect on stimulation quality in vitro fertilization. Fertil Steril 1990; 54: 297 – 302.

21.Martin JS, Nisker JA, Tummon IS, Daniel SA, Auckland JL, Feyles V. Future in vitro fertilization pregnancy potential of women with variably elevated day 3 follicle-stimulating hormone levels. Fertil Steril 1996; 65: 1238 – 1240.

22.Maseelall PB, McGovern PG. Ovarian reserve screening: what the general gynaecologist should know. Women Health 2008; 4(3): 291 -3 00.

23. Knauff EA, Eijkemans MJ, Lambalk CB, ten Kate-Booiji MJ, Hoek A, Beerendonk CC, Laven JS, Goverde AJ, Broekmans FJ,

et al. Anti-Mullerian hormone, inhibin B and antral follicle count in young women with ovarian failure. J Clin Endocrinol Metab 2009; 94(3): 786 - 792.