Jean-Claude Emperaire1
(1)
Bordeaux, France
The principal objective of mono- or paucifollicular ovarian stimulation is to improve or restore ovulation for a patient in which the process is occurring improperly or not at all, or to rescue additional follicles from atresia. On the other hand, a multifollicular stimulation is by its nature disruptive to normal ovarian function because it promotes maturation and development of abnormally numerous follicles. For this reason, it would appear useful to review briefly a few essential concepts of physiology as a way to orient thoughts on the logic, consistency and protocols of different approaches to ovulation stimulation.
At the heart of this process lies the relationship between the ovary and the anterior pituitary gland. The pituitary itself is controlled from the hypothalamus through its gonadotropin releasing hormone (GnRH, gonadorelin) that is in turn subject to influences from multiple areas of the central nervous system that express varied forms of external stimulation upon the pituitary-ovarian axis. Ovarian follicles secrete steroid hormones that act on their respective peripheral hormonal receptors, and they additionally exert feedback actions on the hypothalamic-pituitary complex.
2.1 The Anterior Pituitary-Ovarian Axis
Throughout her reproductive life, a woman’s reserve of primordial follicles declines with advancing age, from some 200,000 at puberty to less than 1,000 at menopause. Each primordial follicle contains an oocyte resting in meiotic prophase I. The fate of a follicle destined for ovulation is actually decided 3 months earlier, when it is “awakened” from this primordial pool. Together with several others, it commences a slow maturation process under the influence of paracrine factors until arriving at the pre-antral stage, a crossing point controlled by anti-Müllerian hormone (AMH). At this point the targeted follicles acquire FSH receptors for the first time. Whereas previous growth was FSH-independent, the structures now become hormonally dependent. By the start of a menstrual cycle, several pre-antral follicles, each 2–5 mm in diameter, have survived the sequence of events during this “awakening” period (Fig. 2.1).
Fig. 2.1
Development of the ovarian follicle from the primordial stage through ovulation
As the menstrual cycle terminates, corpus luteal decay results in a fall of estrogen and progestin secretion that triggers a chain of endometrial changes leading to menstruation. This decline of steroids and inhibin A removes their inhibitory effects on the hypothalamic-pituitary axis and permits a rise of FSH that is perceptible in plasma even before the first day of the menstruation.
Enhanced secretion of FSH, at the end of the previous cycle and during the subsequent early follicular phase, stimulates pre-antral follicles inconsistently because the sensitivity of each follicle is correlated with its respective endowment of FSH receptors. As the growing wave of FSH surpasses the threshold of the most hormone-sensitive follicles, a period of rapid growth ensues for those recruited early, and an “FSH window” opens. With concomitant stimulation by pituitary LH, cells of the theca interna deliver an androgen substrate to the aromatase complex of adjacent granulosa cells that will synthetize increasing amounts of estradiol and also the protein hormone inhibin B. Estradiol in turn furthers the synthesis of additional FSH receptors, with assistance from FSH itself. This “snowballing” effect amplifies follicular sensitivity to both gonadotropins and enhances growth of follicles to the antral stage that is evidenced by accumulating follicular fluid. At the same time, estradiol and FSH induce the synthesis of LH receptors on the granulosa cells, thereby promoting still more estrogen secretion.
The few follicles recruited at the beginning of the cycle are asynchronous with the others as a result of their unique sensitivity to FSH. Although initially indistinguishable morphologically from other follicles, the one whose FSH threshold is lowest begins to grow before the others. Its aromatase complex is the first to be aroused by FSH, and it begins to secrete estrogens before the others. This follicle is then enabled to enter the “snowball” phase ahead of the remaining cohort and is becoming the “selected one” to yield a gamete. As the other follicles join to secrete increasing amounts of estradiol and inhibin B, a negative feedback begins to impact pulsatile FSH release at the hypothalamic-pituitary level. On about the seventh day of the menstrual cycle, FSH secretion falls and the “FSH window” closes. Only the selected follicle continues growing, thanks to its rich population of FSH receptors that permits development even in the face of declining FSH. Whereas growth of all the remaining less sensitive follicles diminishes or stops altogether during the late follicular phase, the lead follicle, now 1 cm in diameter, expands due to additional stimulation from an FSH-like effect of LH. The aromatase complex also becomes functionally coupled to LH receptors, permitting additional LH participation in the selected follicle’s primacy and its late follicular phase maturation. By this time all of the other irregularly stimulated follicles that were initially recruited at the beginning of the cycle have undergone atresia and vanish within the ovarian stroma.
When plasma estradiol levels have been maintained in excess of 100 pg/ml for more than 48 h, and when the selected follicular diameter begins to exceed 16 mm, a large release of hypothalamic GnRH elicits a surge of pituitary FSH and LH. This final gonadotropin wave initiates a stream of follicular events leading to an initial erosion of the follicular wall accompanied by an increase of follicular fluid osmotic pressure that provokes a triad of events: a follicular rupture at the apex, a differentiation of the cumulus, and a final maturation of the oocyte itself. The oocyte’s meiotic process is awakened from prophase I through metaphase II, where maturation is again halted to await arrival of a haploid spermatozoon.
Following expulsion of the ovum, the follicle’s remaining granulosa cells become heavily vascularized and join with remaining theca interna to be transformed into luteinized cells. The corpus luteum actively secretes estradiol accompanied by a large output of progesterone, all being directed by a pulsatile signal from pituitary LH. Luteal lifespan is already programmed for 12–14 days unless its function can be rescued and maintained by exponentially rising chorionic gonadotropin (hCG) levels from trophoblast of an implanting embryo. While the menstrual cycle progressively develops under the influence of two pituitary gonadotropins, FSH and LH, the third gonadotropin, hCG, of placental origin, appears once the embryo implants. This signals the transformation of a menstrual cycle into a beginning pregnancy.
2.2 Follicle-Stimulating Hormone (FSH)
Throughout the menstrual cycle, follicle-stimulating hormone is secreted in a pulsatile manner by gonadotrophs in the adenohypophysis, a process that is itself controlled through a pulsatile release of hypothalamic gonadorelin (GnRH). FSH is delivered through blood to specific receptors on the surface of the ovarian granulosa cells and act to promote follicular growth and development. Despite a modest degradation by the liver, and elimination from the body by the kidneys, urinary FSH retains its biological activity (i.e., it is still able to bind to, and activate, ovarian receptors).
The evolution of ovarian follicles is directed by substantial changes of plasma FSH levels throughout the menstrual cycle. Levels are highest during the end of the luteal phase and early follicular phase (the recruitment period). Then FSH begins to decrease in mid-follicular phase (the selection-leadership period) due to the negative feedback effect of rising levels of estradiol and inhibin B (Fig. 2.2). There is a pre-ovulatory surge of FSH, synchronous with the LH surge, and then levels enter a declining period, only to rise again at the end of the luteal phase.
Fig. 2.2
Patterns of plasma FSH and LH throughout the menstrual cycle
The physiologic role of FSH is directed solely toward proliferation and development of ovarian granulosa cells, thereby to control follicular growth. FSH also induces secretion of inhibin B and, together with estradiol, the appearance and proliferation of LH receptors in the follicle. Finally the pre-ovulatory FSH peak participates importantly in the maturation of the cumulus-oocyte complex.
2.2.1 FSH: A Complex Molecule
Follicle-stimulating hormone is a heterodimer composed of alpha and beta subunits linked noncovalently. The alpha subunit is a chain of 92 amino acids folded by five disulfide bonds, and is identical to the alpha subunits of other glycoprotein hormones: thyrotropin (TSH), luteotropin (LH), and chorionic gonadotropin (hCG).
The beta subunit of FSH is composed of 111 amino acids folded by six disulfide bonds. However, this structure, as with each of the glycoprotein hormones, is unique to its respective hormone. Neither subunit of FSH has biological activity; only the conjoined heterodimer is capable of expressing a hormonal response.
FSH molecules are also characterized by glycosylation: each subunit bears two oligosaccharide chains (glycans) linked to asparagine residues (Asn 52 and Asn 78 on the alpha subunit, Asn7 and Asn 24 on the beta-subunit). These saccharide chains are constructed of N-acetyl-glucosamine, monosaccharides (mannose, galactose), and terminal sialic acids that become more numerous through side-chain branching. Varied oligosaccharide branching results in a number of unique FSH isoforms, also known as isohormones or glycoforms [1]. The pattern of branching is responsible for their richness of sugars (glycosylation) and sialic acid (sialylation) (Fig. 2.3). The glycosylated chains are essential for the full expression of hormonal activity, and cleavage leads to a progressive decline of activity in target cells. The alpha subunit saccharide chains appear to be most important for FSH receptor activation and initiation of the intracellular signal transduction process (adenylate cyclase), whereas the beta subunit chains are known to play a major role in protecting the FSH molecule against degradation and elimination.
Fig. 2.3
Linkage of the polysaccharide groups to the two FSH subunits. The groups extend outside the molecular axis formed by the two interlaced subunits, and thereby increase molecular bulk. The number of radicals linked to the beta subunit distinguishes human FSH tetra- (a), and bi- (b) sialylated molecules, as well as tri-sialylated equine FSH (c), and human recombinant FSH (d)
Some 20–30 FSH isoforms have been identified with distinctive oligosaccharide chain complexes that determine their ability to be maintained in circulation, to bind to the FSH receptor, and to induce a biological response [2]. Observations that the mix of isoforms found in pituitary tissue and in plasma are identical suggest that circulating hormone is not simply the degradation product of a single secreted molecule [3]. Physicochemical characteristics of each isoform are determined by three parameters which affect their physiological properties (Fig. 2.4):
Fig. 2.4
Isoforms of human FSH (Redrawn from Ref. [1])
· Their degree of sialylation in relation to the number of the terminal sialic acid residues, in part a function of the oligosaccharide chain complexity.
· Their degree of sulfonation (sulfonic acid radicals): there is significant competition between galactosamine transferase and galactotransferase at the terminal N-acetylglucosamine residue of the common mucopolysaccharide core. The former attaches an N-acetylgalactosamine residue, followed by a terminal sulfonate, whereas the latter enzyme attaches a galactose residue followed by a terminal sialic acid. In contrast to pituitary LH, no FSH isoform contains more than two sulfonate radicals.
· Their degree of glycosaccharidic chain intricacy, in relation to the richness in chain branching and complexity, and particularly the presence of a bisecting N-acetylglucosamine radical between two other common core segments.
As the branch complexity of FSH isoforms increases, the molecules become more acidic, due to the lateral chains acquiring two to four terminal sialate radicals (or “antennas”). Isoforms are distinguishable by their sialic acid content which proportionately lowers the isoelectric point (pI). Natural FSH becomes a mixture of isoforms with isoelectric points between three (most acidic) and six (least acidic). Nevertheless, mere recognition and identification of the varied FSH isoforms based solely on pI is not in itself sufficient; the richness of internal branching and the number of “antennas” also plays an important part in bioactivity. Isoforms with varied oligosaccharide complexity or degree of sulfonation can share the same level of overall sialylation yet express distinct physiological properties.
Glycosylation of the FSH beta subunit is a separate issue. A portion of the subunit molecules have no glycosylation at either Asn 7 or Asn 24 which results in a mixture of tetra- (MW 24,000) and bi- (MW 21,000) sialylated molecules. This mixture is the result of the hormonal milieu in blood: estradiol decreases the rate of N-glycosylation, most likely through inhibition of oligosaccharyl transferase. Bi-glycosylated forms comprise 75–95 % of the molecular population in postmenopausal women, but only 35–40 % of plasma FSH in menstruating women [4]. This shift of proportion exerts important effects upon the 3-dimensional configuration of the FSH molecules, and hence upon plasma clearance. Both of the N-glycans of the alpha subunit are aligned along the molecule’s axis, whereas the N-glycans of the beta subunit extend outward, effectively doubling the molecular diameter and reducing its rate of glomerular filtration. Studies of a mutant di-glycosylated recombinant FSH showed an elimination rate from blood to be twice as fast as that of normal tetra-glycosylated rFSH [5]. This “all or nothing” process of beta subunit glycosylation provides yet another control mechanism over bioactivity, plus it alters the electrical charge of the molecule as the proportion of the 2- and 4-glycan isoforms shifts. A final point: 3-glycosylated isoforms, having a single radical on the beta subunit (Asn 24), may be found in natural equine FSH or in recombinant human FSH (on Asn 7) but this is never found in naturally occurring human FSH.
The degree of sialylation, and of possible sulfonation, exerts a considerable influence on metabolism of the various isoforms. FSH metabolism begins with removal of hormone from plasma by binding to receptors for specific sialic acid or sulfated residues in hepatic epithelial cells, and followed by renal excretion. More numerous sialic residues on a molecule hinder its hepatic capture and slow its glomerular filtration due to the expanded molecular size. Consequently, the more sialylated and acidic isoforms persist in blood for a longer period. An isohormone with a pI of 4.27 has a half-life of 24 h, but the duration falls to 12 h for isoforms at a pI of 5.49 [6]. On the other hand, for the same degree of sialylation, the addition of sulfone radicals enhances the rate of extraction and elimination of the glycoforms, in direct proportion with the number of SO radicals [7]. Thus the pituitary actually secretes a mixture of isoforms, and each interacts uniquely with the ovarian FSH receptor.
In addition, the structural mix of secreted molecules is not static, and it changes throughout one’s life. The mix also changes throughout the menstrual cycle in reproductive-age women, in accordance with GnRH pulsatility that is principally controlled by the plasma levels of estradiol. The more acidic FSH isoforms prevail during lower estrogen environments, such as before puberty and after menopause. Similarly, more acidic isoforms dominate at the beginning and the end of the menstrual cycle, and less acidic isohormones prevail during the pre-ovulatory period to produce a more homogenous distribution (Fig. 2.5) [8]. Individual patients secrete a uniquely tailored isoform profile that is already distinguishable on cycle day 3. The same occurs for LH isoforms [9].
Fig. 2.5
Profile of FSH isoform secretion during the menstrual cycle (Redrawn from Ref. [2])
Bioactivity of the isoforms also varies according to the model studied. Whereas in vivo activity is principally related to the molecular half-life, in vitro activity becomes more a function of its affinity for the cellular receptors. The relative activity of FSH isoforms is affected by their degree of glycosylation, in an apparently conflicting manner [10]:
· The more acidic glycoforms show a higher bioactivity in vivo, for example, in the Steelman-Pohley bioassay that measures a dose–response variation of immature rat ovarian weight in relation to the injected FSH doses.
· On the contrary, the less acidic isoforms show a higher affinity in vitro for the receptor, and thus a higher bioactivity that stimulates an earlier, more rapid and intense estradiol secretion by the granulosa cells, that may also be enhanced by a stronger postreceptor activity.
This outward inconsistency is explained, at least in part, by the clearance of the different isohormones, related itself to their degree of glycosylation. The more acidic forms are picked up and removed from the plasma more slowly, because of the sialic acid inhibition of FSH binding to the hepatic asialo-glycoproteic receptors: their prolonged existence within the blood stream partly compensates for their lower bioactivity.
It should be noted that both in vivo and in vitro assays measuring the FSH activity have their own limits. In vivo models fail to account for interspecies variations of FSH clearance, and for the possible bias introduced by a bolus injection of hormone that contrasts to the normal pulsatile secretion of FSH. On the other hand, in vitro assays, while avoiding the hormone half-life issues, measure only a single type of response, e.g., androgen aromatization by rat granulosa or Sertoli cells, or the quantity of AMP produced by cell lines expressing the human FSH receptor. Reliance on a single response parameter is unable to account for the intricacy of the complete follicular response to FSH, that includes follicular development, sequential as well as synergetic differentiation of granulosa and theca cells, antrum formation and antral fluid secretion. These concerns have raised awareness of working toward in vitro testing models that utilize intact ovarian follicles of mice capable of being grown to full maturity. This would lead to an improved evaluation of follicle and oocyte quality with respect to the various hormonal isoforms being tested.
It should be further noted that bioactivity differences between the different isoforms of FSH are not solely explained by their isoelectric point, but are also subject to individual properties that were revealed by in vitro studies of intact mouse follicles. Results from this model lend strength to the critical importance of FSH in antral follicular development, and also illustrate how all follicular cells – granulosa, theca and oocyte – are responding to increasing FSH dosages [11]:
· A rather precise threshold dose stimulates follicular antrum formation, which supports the concept of an FSH threshold often observed in clinical practice; no FSH effects are detectible below the threshold dose.
· A maximal dose above which degenerative modifications appear histologically in granulosa and theca cells, as well as in the oocyte. There appears to be a “ceiling” dose, over which granulosa cell death and oocyte damage has been observed, which characterizes a true FSH overdose, at least at the follicle level.
· An FSH “efficacy range” that occurs between the threshold and ceiling doses.
Importantly, these three parameters are known to vary widely according to the molecular isoform pI being tested. Less acidic glycohormones can provoke follicular growth at concentrations as low as 1.5 μg/ml, and can stimulate a follicular pool, even those less sensitive, yet their “ceiling” can be as low as 5 μg/ml. The more acidic isoforms, in contrast, are effective only at concentrations above 40 μg/ml, to an activity ceiling of 400 μg/ml. The latter isoform induces a more selective follicular growth. Similarly, a less acidic mixture of FSH isohormones stimulates a more rapid follicular growth in vitro. Additionally, use of this model makes it possible to follow different isoform effects up through the overall embryo quality, beginning with the oocyte of a chosen follicle.
All these varied data confirm a higher bioactivity of the less acidic isoform, and explain further the relevance of the spectrum of composition changes in the FSH secreted throughout the menstrual cycle. Follicular recruitment occurs under the effect of relatively high quantities of more acidic isoforms that have a higher activity threshold and larger efficacy range. Dominance of less acidic isoforms in the middle and late follicular phases serves to increase estrogen secretion and stimulate granulosa cell mitosis, which enhances the overall FSH effect on antral development.
This mouse model also underscores the concept that the different source-related bioactivities of FSH cannot be attributed solely to variations of isoelectric point. When the in vitro responses of an intact follicle to natural pituitary FSH are compared to those of recombinant FSH normalized to the same pI (3.5–5.3), the threshold dose is lower for rFSH. This higher in vitro bioactivity cannot be explained by charge discrepancies between the isoforms of the two preparations. Perhaps the use of a nonhuman cell line for rFSH synthesis results in other subtle consequences for bioactivity that would be located within the specific glycosaccharidic chain structure and composition.
There is also growing evidence that, in addition to pI differences, various FSH isoforms might play a specific role in the recruitment, and then development, of the dominant follicle up to the moment of ovulation, as is known to be the case for LH, and that this has some noticeable outcomes on the embryo quality. In vitro, some glycoforms show a particular ability to stimulate estradiol secretion, or granulosa cell proliferation, or even synthesis of plasminogen that is crucial in the optimal maturation of the pre-ovulatory follicle and its wall dehiscence. Other isohormones are endowed with paradoxical properties, such as an LH-like effect or a FSH antagonist activity. At some future time therapeutic FSH molecules may be tailored by procedures that shorten one or several carefully selected isoforms [12].
2.2.2 The FSH Receptor
FSH exerts its activity through specific receptors located on the surface of ovarian granulosa cells. These guanine nucleotide-binding, signal-transducing proteins (often called G protein-coupled receptors or GPR) are constructed of extracellular, transmembrane, and intracellular domains (Fig. 2.6). A receptor’s primary sequence is susceptible to genomic variations. Aside from uncommon mutations that result in loss of sensitivity to FSH, more numerous variants can have significant implications for clinical practice. Many receptor polymorphisms are as simple as an alteration of a single DNA nucleotide (single nucleotide polymorphism, or SNP), and can occur in at least 1 % of all people. Among the more clinically significant polymorphisms, allelic combinations at position 680 are of clinical importance; the heterozygous asparagin/serin (Asn/Ser) mutation, however, is more widespread than the homozygous variants Ser/Ser and Asn/Asn [13]:
Fig. 2.6
Structure of the FSH receptor (Redrawn from Ref. [1])
· The Ser/Ser genotype produces a receptor having decreased sensitivity to FSH that leads to a slightly elevated plasma level, and the necessity for higher FSH doses to drive ovarian stimulation [14];
· The Asn/Asn genotype, on the contrary, can induce an exaggerated sensitivity to FSH that could enhance the risk for severe ovarian hyperstimulation [15].
Taken together, normally-occurring polymorphisms of both the FSH molecular isoforms and of the FSH receptor are likely responsible for many of the varied responses of the same patient to the different available FSH preparations, as well as the different outcomes between patients receiving the same FSH preparation.
2.2.3 Mechanisms of Action
Overall, pituitary FSH is the main hormone responsible for ovarian follicular development. There are four aspects to its action:
1.
2.
3.
4.
2.3 Luteinizing Hormone (LH)
Like FSH, LH is a heterodimer. It is composed of a 92-amino acid alpha subunit that is identical to FSH-α, and a covalently bound unique 121 amino acid beta subunit (Fig. 2.9). LH is much less sialylated than FSH or chorionic gonadotropin (hCG) and has a less acidic isoelectric point in the range of 7.1–9.5. Both di- and tri- glycosylated isoforms are present in serum [16]. Some 17 LH glycoforms have been identified by pI from pituitary tissue. As is the case for FSH, the clearance rate from blood of the different LH isoforms is hindered in proportion to their degree of sialylation, and accelerated by the degree of sulfonation. Also similarly to FSH, LH binding affinity to its receptor, and resulting in vitro bioactivity, decreases, and plasma half-life increases, as the pI declines. The least acidic isoforms are some 16 times more bioactive than the most acidic isoforms, due to enhanced receptor binding, and despite a slower clearance of the latter.
Fig. 2.9
Essential features of primary structures of the common alpha subunit and unique beta subunits of FSH, LH, and hCG
Less acidic LH isoforms have fewer oligosaccharide moieties, typically containing only one “antenna” with a terminal mannose residue; the more acidic glycoforms have more complex oligosaccharide chains, often with two “antennae” having sialic acid or N-galactosamine sulfate terminals. Desialylation of the LH molecule reduces in vivo bioactivity by over 100-fold, but without affecting immunoactivity [17].
The mix of plasma LH isoforms also changes throughout the menstrual cycle, being less acidic during the pre-ovulatory surge peak (pI of 8.0–9.0) than during either the follicular or luteal phases (7.5–8.9) [2]. Di-glycosylated LH circulates at lower concentrations than the triglycosylated form throughout the menstrual cycle, except at midcycle when the opposite occurs [16]. Isoform profiles are also very distinctive for each person [18].
The mix of LH isohormones in urine contrasts with plasma, due to the variable rates of renal clearance as well as to the effects of chemical purification procedures used in hormone extraction. When two pharmaceutical preparations of HMG were on the market, the Humegon® isoform mix was much less acidic than the Pergonal® mix.
LH acts through a specific receptor, also activated by hCG (LH/hCG receptor, LHr), located on cells of the ovarian theca interna, granulosa and in the corpus luteum. This receptor belongs to the same G-coupled superfamily as the FSH receptor, and has the typical composition of extra-cellular, intra-cellular, and transmembrane domains. It shares approximately a 50 % homology with the FSH receptor. Primordial and primary follicles are devoid of LHr, which begin to appear on granulosa and theca cells when antral follicles reach a 0.3 mm size. Receptor population then increases as the follicles mature to the pre-ovulatory stage, and LHr protein appears also to be expressed in the cumulus cells. LHr protein is ubiquitously present in the corpus luteum. Receptor development does not seem to occur simultaneously, but gradually at different times and in varied follicular sizes. Only a subset of follicles produce LHr at sufficient levels to enable response to administered hCG, which may explain why all of the oocytes are not equally mature at the time of aspiration for IVF [19].
The physiologic role of LH in ovarian function varies according to the target:
· In the theca interna of growing follicles, LH stimulates the synthesis of androgens. While some product is secreted into the bloodstream, most of the androgen production moves toward the granulosa cells to become a substrate for estrogen biosynthesis by the aromatase complex.
· In granulosa cells of the growing follicle, expression of the LH/hCG receptor is apparently controlled by FSH action. In essence LH and FSH synergistically promote final maturation of the dominant follicle.
· In mature follicles, the pre-ovulatory surge of LH initiates a chain of irreversible events within the follicle and its contents, leading to its rupture and expulsion of its gamete. Pre-ovulatory transformations of the cumulus cells are provoked together with the concomitant FSH surge.
· In the corpus luteum, a pulsatile secretion of pituitary LH supports the postovulatory release of estradiol and progesterone for 12–14 days.
In addition, alterations of the LH molecule or of its secretion patterns may serve to explain certain clinical abnormalities
· LH levels are frequently elevated in the polycystic ovary syndrome, due to a basal hypersecretion through an increased pulse frequency and amplitude, that is only partly compensated by a preferential secretion of less acidic LH isoforms having a shorter half-life [20].
· One of the genetic variants of LH has two mutations in the beta subunit, one of which contains an additional glycosylated radical that increases its half-life. This tends to diminish LH pulsatility, which could explain the excessive rate of ovulatory disorders and idiopathic infertility that occurs more commonly in these patients [21].
· A common polymorphic allele of the LH beta-subunit (V-beta LH) is associated with higher FSH consumption during controlled ovarian stimulation for assisted reproductive technology [22].
Overall, LH has an essential role in the ovarian steroidogenesis, as well as a secondary role in the final maturation of the follicle. Unfortunately, the fact that hCG has the same biological profile as LH in the most essential clinical indications has retarded important advances in understanding the properties and possible clinical advantages of the different LH isoforms.
2.4 Human Chorionic Gonadotropin (hCG)
hCG is secreted by the placenta throughout the entire pregnancy, from implantation to delivery. There are also reports of minute, pulsatile secretions of pituitary hCG together with pulses of LH. However, these are detectible only with high plasma LH concentration, such as during the pre-ovulatory gonadotropin surge and after menopause. Indeed, the gene coding for the LH beta subunit is located among the seven genes coding for hCG beta. In contrast to FSH and LH, hCG is not produced as a single molecule, but as several molecules; some variants have no hormonal activity. The total product secreted by placental trophoblast and by nontrophoblastic tumors, includes five distinct hCG molecules [23]:
· Complete hCG, a heterodimer of two covalently-bound subunits.
· The alpha subunit (identical to FSH and LH α) is a single 92-amino acid chain with two glycosaccharidic chains bound to Asn residues, and with two or three “antennae.”
· The beta subunit, specific to hCG and composed of 155 amino acids, has two Asn-bound glycosaccharidic chains having two or three “antennae” each. The primary sequence of the first 121 N-terminal amino acids is very close to that of the LH-ß subunit, but hCG-β has an additional 34 amino acids at the C-terminal end (carboxyl terminal peptide, CTP). This additional sequence also contains four tri- or hexa-saccharide chains O-linked to serine residues. The presence of these residues results in a higher volume of distribution for hCG, and a prolonged presence in blood.
· Hyperglycosylated hCG: This is the complete hCG molecule bearing additional trisaccharide N-linked chains and hexasaccharide O-linked chains. Its molecular weight reaches 40,000 (versus 36,000 for the complete hCG), of which 25–30 % of the total MW is oligosaccharide.
· Three unpaired subunits: Also secreted are free alpha and beta hyperglycosylated subunits and a free O-glycosylated alpha subunit. These unpaired alpha subunits are devoid of biological activity.
Placental syncytiotrophoblast secretes the whole hCG and the free alpha subunits, whereas the hyperglycosylated hCG and the free O-glycosylated alpha subunits are produced by the cytotrophoblast cells.
The degradation of the whole hCG begins in plasma under the effect of circulating macrophage proteases. Alpha and beta subunits are first separated, and then each is further cleaved. One large cleavage unit is the unique CTP chain at amino acids 93–155 on the beta subunit. Freed subunits and hydrolysis products are cleared from the plasma ten times faster than the whole dimeric hCG, particularly when the CTP group has been cleaved. Plasma of pregnant woman may contain up to ten degradation products in addition to the five complete hCG molecules originally synthesized. None of the degradation products is biologically active.
As a result, extracted urinary hCG is a highly heterogeneous mix of 15 distinct molecules resulting from renal degradation. A commonly found structure is a core fragment of the beta subunit containing two sequences (amino acids 6–40 and 55–92) still bound together by five disulfide bridges. The proportion of this beta-core fragment rises throughout pregnancy and by the seventh week of gestation can actually exceed levels of the whole bioactive hCG in urine. This biologically inactive material is typically the dominant hCG-related molecule in urine by the time of parturition.
The physiologic action of hCG is notable by having a higher, more stable affinity for the LH/hCG receptor than does LH, because of its enhanced glycosylation. In addition, the prolonged half-life (24–33 h for the slow elimination period, versus 10–12 h for LH) explains the more robust in vivo biological effect that is, for example, seven times higher in the rat seminal vesicle assay (a target for Leydig cell androgen production); it also explains the unique physiologic roles of each hormone [24].
It has always been believed that the physiological role of hCG is relegated to stimulation of progesterone production by ovarian luteal cells during early pregnancy, but it has never seemed logical that a hormone produced for 37 weeks should exist for only a 3-week span until the placenta’s own progesterone production becomes adequate. Indeed, numerous additional basic actions of hCG have been highlighted during the past decade, including support for angiogenesis of the myometrial spiral cells, and fusion of the syncytio- and cytotrophoblast cells. Both actions are critical functions for a normal placentation that could have become involved in the brain development throughout human evolution [23].
References
1.
Ulloa-Aguirre A, Timossi C (1998) Structure-function relationship of follicle-stimulating hormone and its receptor. Hum Reprod Update 4:260–283PubMedCrossRef
2.
Stanton PG, Burgon PG, Hearn MTW, Mendez JP (1996) Structural and functional characterization of FSH and LH isoforms. Mol Cell Endocrinol 125:133–141PubMedCrossRef
3.
Ulloa-Aguirre A, Timossi C, Mendez JP (2001) Is there any physiological role for gonadotropin oligosaccharide hetrogenicity in human? I. Gonadotropins are synthesized and released in multiple molecular forms. A matter of fact. Hum Reprod 16:599–604PubMedCrossRef
4.
Bousfield GR, Butnev VY, Walton W et al (2007) All-or-none N-glycosylation in primate follicle stimulating hormone ß-subunits. Mol Cell Endocrinol 262:40–48CrossRef
5.
D’Antonio M, Borrelli F, Datola A et al (1999) Biological characterization of recombinant human follicle stimulating hormone isoforms. Hum Reprod 14:1160–1167PubMedCrossRef
6.
Timossi CM, Barrios-di-Tomassi J, Gonzalez-Suarez R et al (2000) Differentiated effects of the charge variants of human follicle stimulating hormone. J Endocrinol 165:193–205PubMedCrossRef
7.
Wide L, Eriksson K, Sluss PM, Hall JE (2009) Serum half-life of pituitary gonadotropins is decreased by sulfonation and increased by sialylation in women. J Clin Endocrinol Metab 94:958–964PubMedCentralPubMedCrossRef
8.
Anobile CJ, Talbot JA, McCann SJ et al (1998) Glycoform composition of serum gonadotropins through the normal menstrual cycle and in the post-menopause state. Mol Hum Reprod 4:631–639PubMedCrossRef
9.
Phillips DJ, Albertson-Xikland K, Eriksson K et al (1997) Changes in the isoform of luteinizing hormone and follicle stimulating hormone during puberty in normal children. J Clin Endocrinol Metab 82:3103–3106PubMedCrossRef
10.
Ryan RJ, Charlesworth MC, McCormick DJ, Milius RP et al (1982) The glycoprotein hormones: recent studies of structure-function relationship. FASEB J 2:2661–2669
11.
Nayudu P, Vitt UA, Barrios de Tomasi J et al (2002) Intact follicle culture: what it can tell us about the roles of glycoforms during the follicle development. Reprod Biomed Online 5:240–253PubMedCrossRef
12.
Emperaire JC (2011) Propriétés et intérêt clinique des isoformes de la FSH. MT Med Reprod Gynecol Endocrinol 13:49–60
13.
Hughes JN (2011) Récepteur de la FSH et stimulation ovarienne. In: Franchin R (ed) La stimulation ovarienne: concepts et réalités. MED’COM, Paris, p 41
14.
Perez Mayorga M, Gromoll J, Behre HM et al (2000) Ovarian response to follicle stimulating hormone (FSH) stimulation depends on the FSH receptor genotype. J Clin Endocrinol 85:3365–3369
15.
Daelmans C, Smits G, de Maertelaer V et al (2004) Prediction of severity of symptoms in iatrogenic ovarian hyperstimulation syndrome by follicle stimulating hormone receptor Ser680Asn polymorphism. J Clin Endocrinol Metab 89:6310–6315CrossRef
16.
Wide L, Eriksson K (2013) Dynamic changes in glycosylation and glycan composition of human FSH and LH during natural ovarian stimulation. Ups J Med Sci 118:153–164PubMedCentralPubMedCrossRef
17.
Stanton PG, Pozvek G, Burgon PG et al (1993) Isolation and characterization of human LH isoforms. J Endocrinol 138:529–543PubMedCrossRef
18.
Burgon PG, Stanton PG, Robertson DM (1996) In vivo bioactivities and clearance patterns of highly purified human luteinizing hormone isoforms. Endocrinology 137:4827–4836PubMed
19.
Yung Y, Aviel-Ronen S, Maman E et al (2014) Localization of luteinizing hormone receptor protein in the human ovary. Mol Hum Reprod 20:844–849PubMedCrossRef
20.
Ropelato MG, Garcia-Rudaz MC et al (1999) A preponderance of basic luteinizing (LH) isoforms accompanies inappropriate hypersecretion of both basal and pulsatile LH in adolescents with polycystic ovarian syndrome. J Clin Endocrinol Metab 84:4629–4636PubMed
21.
Wide L, Eriksson K, Sluss PM, Hull JE (2010) The common genetic variant of luteinizing hormone has a longer half life than the wild type in heterozygous women. J Clin Endocrinol 95:383–389CrossRef
22.
Alviggi C, Petersson K, Longobardi S et al (2013) A common polymorphic allele of the beta-LH subunit gene is associated with higher exogenous FSH consumption during controlled ovarian stimulation for assisted reproductive technology. Reprod Biol Endocrinol 11:51PubMedCentralPubMedCrossRef
23.
Cole LA (2009) New discoveries on the biology and detection of human chorionic gonadotropin. Reprod Biol Endocrinol 7:1–47CrossRef
24.
Choi J, Smitz J (2014) Luteinizing hormone and human chorionic gonadotropin: distinguishing unique physiologic roles. Gynecol Endocrinol 30:174–181PubMedCentralPubMedCrossRef