Basic science studies have advanced our understanding of the role of key enzymes in the steroidogenesis pathway and those that affect the pathophysiology of PCOS. Studies with ovarian theca cells taken from women with PCOS have demonstrated increased androgen production due to increased CYP17A1 and HSD3B2 enzyme activities. Furthermore, overexpression of DENND1A variant 2 in normal theca cells resulted in a PCOS phenotype with increased androgen production. Notably, cellular steroidogenesis models have facilitated the understanding of the mechanistic effects of pharmacotherapies, including insulin sensitizers (e.g., pioglitazone and metformin) used for the treatment of insulin resistance in PCOS, on androgen production. In addition, animal models of PCOS have provided a critical platform to study the effects of therapeutic agents in a manner closer to the physiological state. Indeed, recent breakthroughs have demonstrated that natural derivatives such as the dietary medium-chain fatty acid decanoic acid (DA) can restore estrous cyclicity and lower androgen levels in an animal model of PCOS, thus laying the platform for novel therapeutic developments in PCOS. This chapter reviews the current understanding on the pathways modulating androgen biosynthesis, and the cellular and animal models that form the basis for preclinical research in PCOS, and sets the stage for clinical research.
Highlights
- •
Ovarian and adrenal steroidogenesis and hyperandrogenism in association with polycystic ovarian syndrome (PCOS) are elucidated.
- •
Cellular models of steroidogenesis are presented.
- •
The animal models of PCOS are discussed.
Introduction
Following on from the previous chapter, where the diagnostic criteria of PCOS and the difficulty of establishing a defined set of criteria is discussed, it is apparent that PCOS is not just a clinical conundrum but also one with a pathophysiological origin that remains unclear. Basic science studies based upon cellular models of steroidogenesis and animal models of PCOS have been fundamental to the current understanding of the PCOS pathophysiology. This chapter reviews the current understanding on the pathways modulating androgen biosynthesis and the cellular and animal models useful for preclinical research in PCOS.
Regulation of androgen production
Basic science studies have advanced our understanding of the mechanisms of androgen secretion, a process fundamental to the pathogenesis of PCOS. Androgens are produced by the adrenal and gonads and are involved in the regulation of numerous developmental and physiological processes in women. Two major axes control androgen production in women: the hypothalamic–pituitary–adrenal axis and the hypothalamic–pituitary–ovarian axis ( Fig. 1 ).
In the hypothalamic–pituitary–adrenal axis, the hypothalamus secretes corticotrophin-releasing hormone (CRH), which stimulates the production of adrenocorticotropic hormone (ACTH) by the anterior pituitary ( Fig. 1 , left panel). ACTH after being released into the systemic circulation binds to specific receptors on adrenocortical cells in the adrenal gland to stimulate the production of adrenal androgens, mineralocorticoids, and glucocorticoids . Conversely, in the hypothalamic–pituitary–ovarian axis, the gonadotropin-releasing hormone (GnRH) is secreted by the hypothalamus in a pulsatile fashion , and an increased pulse frequency of hypothalamic GnRH stimulates secretion of luteinizing hormone (LH) by the anterior pituitary ( Fig. 1 , right panel). Ovarian theca cells respond to LH in the circulation by increasing biosynthesis and secretion of androgens . Both ACTH and LH bind to specific G-protein-coupled receptors, which stimulates adenylyl cyclase activity and increases the production of the intracellular second messenger cyclic adenosine monophosphate (cAMP) ( Fig. 2 ). The cAMP activates protein kinase A and increases the steroidogenic capacity of the cell via a cascade of events leading to an increased expression of the steroidogenic enzymes (e.g., StAR, CYP11A1, CYP17A1, and HSD3B2) that produce androgens from cholesterol ( Fig. 2 ). The steroidogenesis pathway is illustrated in detail in Fig. 2 .
Androgens are derived primarily from dietary cholesterol that circulates in the form of low-density lipoproteins (LDL) in the plasma . These plasma LDL cholesterols can be taken up by LDL receptor-mediated endocytosis into steroidogenic cells to synthesize androgens ( Fig. 2 ). In the steroid-producing cells, LDL cholesterol can either be stored in lipid droplets for future use or be converted into free cholesterol and utilized immediately for androgen biosynthesis .
The steroidogenic acute regulatory protein (StAR) facilitates the transportation of cholesterol from the outer mitochondrial membrane to the inner mitochondrial membrane . This step limits the availability of cholesterol for steroidogenesis. Initiation of steroidogenesis occurs within the mitochondria where the cholesterol side-chain cleavage enzyme encoded by CYP11A1 catalyzes the conversion of cholesterol to pregnenolone ( Fig. 2 ). This is the rate-limiting step in steroidogenesis. In the adrenal gland, pregnenolone can be metabolized to mineralocorticoids, glucocorticoids, or adrenal androgens, depending on the relative expression and activities of the steroidogenic enzymes . Conversely, in the ovarian theca cells, pregnenolone can only be metabolized via the androgen biosynthesis pathway .
Dehydroepiandrosterone (DHEA) synthesis from pregnenolone is mediated by CYP17A1, which encodes a single enzyme cytochrome P450 17A1, with both 17α-hydroxylase and 17, 20-lyase activities . CYP17A1 is considered to be the rate-limiting enzyme in androgen biosynthesis . Dehydroepiandrosterone sulfate (DHEA-S) is a metabolite of DHEA that is produced exclusively in the adrenal glands by the addition of a sulfate group, catalyzed by the sulfotransferase enzymes . DHEA is a Δ5 steroid and a weak androgen. Most of the biologically active androgens are Δ4 steroids. DHEA can be converted to androstenedione, a Δ4 steroid, by the enzyme 3β-hydroxysteroid dehydrogenase. Two active isoforms of 3β-hydroxysteroid dehydrogenase have been found in humans, namely 3β-hydroxysteroid dehydrogenase type 1 (HSD3B1) and type 2 (HSD3B2). HSD3B1 is mainly active in the placenta, while it is also found in breast, liver, and brain . However, HSD3B2 is the primary isoform found in the adrenals and gonads and is involved in multiple steps of the androgen biosynthesis in these glands. It catalyzes the biosynthesis of progesterone from pregnenolone, 17α-hydroxyprogesterone from 17α-hydroxypregnenolone, and androstenedione from DHEA .
Androstenedione is converted by 17β-hydroxysteroid dehydrogenase (HSD17B) to the more potent androgen testosterone. There are at least 14 types of isoforms of HSD17B in humans and they differ in their substrate specificity and sites of expression . The predominant isoform that converts androstenedione to DHEA is HSD17B3, although HSD17B1 and HSD17B5 have also been found to catalyze the reaction .
At the end of the androgenic steroidogenesis pathway, testosterone can be either reduced to the non-aromatizable dihydrotestosterone (DHT) by 5α-reductase (SRD5A2) or aromatized to estrogen in the ovarian granulosa cells by the enzyme aromatase (CYP19A1) ( Fig. 2 ), which is a member of the cytochrome p450 family. Compared to testosterone, DHT has a greater binding affinity for the androgen receptor (AR) and a slower dissociation rate, thus making DHT the most potent androgen . The relative potencies of the androgens are illustrated in Table 1 .
| Androgen | Relative potency (%) |
|---|---|
| DHT | 300 |
| Testosterone | 100 |
| Androstenedione | 10 |
| DHEA | 5 |
Besides the adrenal glands and ovaries, peripheral tissues including adipose tissues, hair follicles, and genital skin also contribute to androgen biosynthesis and circulating androgens . Human adipose tissues express HSD3B2, HSD17, and SRD5A2 and are thus capable of converting circulating plasma DHEA-S, DHEA, androstenedione, and testosterone to more potent forms of androgens . Recent findings further suggest that the presence of StAR and CYP11A1 may facilitate de novo steroidogenesis from cholesterol in adipose tissues . The relative contributions of the major androgens by the ovary, adrenal gland, and peripheral tissues are illustrated in Table 2 .
| Androgen | Source | ||
|---|---|---|---|
| Ovary | Adrenal gland | Peripheral tissues | |
| Testosterone | 25% | 25% | 50% |
| Androstenedione | 30% | 30% | 40% |
| DHEA | 20% | 50% | 30% |
| DHEAS | – | >95% | – |
Regulation of androgen production
Basic science studies have advanced our understanding of the mechanisms of androgen secretion, a process fundamental to the pathogenesis of PCOS. Androgens are produced by the adrenal and gonads and are involved in the regulation of numerous developmental and physiological processes in women. Two major axes control androgen production in women: the hypothalamic–pituitary–adrenal axis and the hypothalamic–pituitary–ovarian axis ( Fig. 1 ).
In the hypothalamic–pituitary–adrenal axis, the hypothalamus secretes corticotrophin-releasing hormone (CRH), which stimulates the production of adrenocorticotropic hormone (ACTH) by the anterior pituitary ( Fig. 1 , left panel). ACTH after being released into the systemic circulation binds to specific receptors on adrenocortical cells in the adrenal gland to stimulate the production of adrenal androgens, mineralocorticoids, and glucocorticoids . Conversely, in the hypothalamic–pituitary–ovarian axis, the gonadotropin-releasing hormone (GnRH) is secreted by the hypothalamus in a pulsatile fashion , and an increased pulse frequency of hypothalamic GnRH stimulates secretion of luteinizing hormone (LH) by the anterior pituitary ( Fig. 1 , right panel). Ovarian theca cells respond to LH in the circulation by increasing biosynthesis and secretion of androgens . Both ACTH and LH bind to specific G-protein-coupled receptors, which stimulates adenylyl cyclase activity and increases the production of the intracellular second messenger cyclic adenosine monophosphate (cAMP) ( Fig. 2 ). The cAMP activates protein kinase A and increases the steroidogenic capacity of the cell via a cascade of events leading to an increased expression of the steroidogenic enzymes (e.g., StAR, CYP11A1, CYP17A1, and HSD3B2) that produce androgens from cholesterol ( Fig. 2 ). The steroidogenesis pathway is illustrated in detail in Fig. 2 .
Androgens are derived primarily from dietary cholesterol that circulates in the form of low-density lipoproteins (LDL) in the plasma . These plasma LDL cholesterols can be taken up by LDL receptor-mediated endocytosis into steroidogenic cells to synthesize androgens ( Fig. 2 ). In the steroid-producing cells, LDL cholesterol can either be stored in lipid droplets for future use or be converted into free cholesterol and utilized immediately for androgen biosynthesis .
The steroidogenic acute regulatory protein (StAR) facilitates the transportation of cholesterol from the outer mitochondrial membrane to the inner mitochondrial membrane . This step limits the availability of cholesterol for steroidogenesis. Initiation of steroidogenesis occurs within the mitochondria where the cholesterol side-chain cleavage enzyme encoded by CYP11A1 catalyzes the conversion of cholesterol to pregnenolone ( Fig. 2 ). This is the rate-limiting step in steroidogenesis. In the adrenal gland, pregnenolone can be metabolized to mineralocorticoids, glucocorticoids, or adrenal androgens, depending on the relative expression and activities of the steroidogenic enzymes . Conversely, in the ovarian theca cells, pregnenolone can only be metabolized via the androgen biosynthesis pathway .
Dehydroepiandrosterone (DHEA) synthesis from pregnenolone is mediated by CYP17A1, which encodes a single enzyme cytochrome P450 17A1, with both 17α-hydroxylase and 17, 20-lyase activities . CYP17A1 is considered to be the rate-limiting enzyme in androgen biosynthesis . Dehydroepiandrosterone sulfate (DHEA-S) is a metabolite of DHEA that is produced exclusively in the adrenal glands by the addition of a sulfate group, catalyzed by the sulfotransferase enzymes . DHEA is a Δ5 steroid and a weak androgen. Most of the biologically active androgens are Δ4 steroids. DHEA can be converted to androstenedione, a Δ4 steroid, by the enzyme 3β-hydroxysteroid dehydrogenase. Two active isoforms of 3β-hydroxysteroid dehydrogenase have been found in humans, namely 3β-hydroxysteroid dehydrogenase type 1 (HSD3B1) and type 2 (HSD3B2). HSD3B1 is mainly active in the placenta, while it is also found in breast, liver, and brain . However, HSD3B2 is the primary isoform found in the adrenals and gonads and is involved in multiple steps of the androgen biosynthesis in these glands. It catalyzes the biosynthesis of progesterone from pregnenolone, 17α-hydroxyprogesterone from 17α-hydroxypregnenolone, and androstenedione from DHEA .
Androstenedione is converted by 17β-hydroxysteroid dehydrogenase (HSD17B) to the more potent androgen testosterone. There are at least 14 types of isoforms of HSD17B in humans and they differ in their substrate specificity and sites of expression . The predominant isoform that converts androstenedione to DHEA is HSD17B3, although HSD17B1 and HSD17B5 have also been found to catalyze the reaction .
At the end of the androgenic steroidogenesis pathway, testosterone can be either reduced to the non-aromatizable dihydrotestosterone (DHT) by 5α-reductase (SRD5A2) or aromatized to estrogen in the ovarian granulosa cells by the enzyme aromatase (CYP19A1) ( Fig. 2 ), which is a member of the cytochrome p450 family. Compared to testosterone, DHT has a greater binding affinity for the androgen receptor (AR) and a slower dissociation rate, thus making DHT the most potent androgen . The relative potencies of the androgens are illustrated in Table 1 .
| Androgen | Relative potency (%) |
|---|---|
| DHT | 300 |
| Testosterone | 100 |
| Androstenedione | 10 |
| DHEA | 5 |
Besides the adrenal glands and ovaries, peripheral tissues including adipose tissues, hair follicles, and genital skin also contribute to androgen biosynthesis and circulating androgens . Human adipose tissues express HSD3B2, HSD17, and SRD5A2 and are thus capable of converting circulating plasma DHEA-S, DHEA, androstenedione, and testosterone to more potent forms of androgens . Recent findings further suggest that the presence of StAR and CYP11A1 may facilitate de novo steroidogenesis from cholesterol in adipose tissues . The relative contributions of the major androgens by the ovary, adrenal gland, and peripheral tissues are illustrated in Table 2 .
| Androgen | Source | ||
|---|---|---|---|
| Ovary | Adrenal gland | Peripheral tissues | |
| Testosterone | 25% | 25% | 50% |
| Androstenedione | 30% | 30% | 40% |
| DHEA | 20% | 50% | 30% |
| DHEAS | – | >95% | – |
Ovarian and adrenal hyperandrogenism in PCOS
The ovaries and adrenal glands are major contributors to circulating androgens and are both likely to cause androgen excess in PCOS. As PCOS is seen primarily as an ovarian disorder, the conventional view is that the ovary is the main source of excess androgens. Indeed, women with PCOS have raised LH levels, a key stimulus leading to an increased ovarian androgen production . Notably, theca cells taken from the ovaries of women with PCOS have increased CYP17A1 and HSD3B2 enzyme activities and increased production of androgens in both basal and gonadotropin-stimulated state . More recently, genome-wide association studies in Asian and Caucasian cohorts established the role of DENND1A gene in PCOS . Increased expression of DENND1A variant 2 was detected in theca cells taken from women with PCOS. Moreover, overexpression of DENND1A variant 2 in normal theca cells resulted in a PCOS phenotype with increased CYP17A1 expression and DHEA production . Regarding adrenal hyperandrogenism, elevated levels of DHEA-S, an androgen exclusively secreted by the adrenal gland, have also been noted in women with PCOS .
Cellular models of steroidogenesis
In women, androgens are secreted in almost equal quantities by the ovaries and adrenal glands , and the enzymes involved in the steroidogenesis pathway are similar . Thus, both human adrenal- and ovarian-derived cell lines have been employed for basic research examining the regulation of steroidogenesis and are useful tools for studying compounds that may modulate these processes. Two of these well-characterized in vitro models of steroidogenesis are discussed later.
Adrenocortical cell lines
The NCI-H295 cell line was originally derived from an invasive human adrenocortical tumour and has been established as a permanent functioning culture . It expresses multiple pathways of adrenal steroidogenesis, secreting corticosteroids, mineralocorticoids, androgens, and estrogens, in response to ACTH, forskolin, and cAMP . Detailed characterization of the NCI-H295 cell line revealed expression of all the key steroidogenic enzymes, including StAR, CYP11A1, CYP17A1, HSD3B2, aromatase, and CYP21A1 , thereby making it an excellent model to study mechanisms controlling steroid production.
From the NCI-H295 cell line, two sub-strains, NCI-H295A and NCI-H295R, were derived . Differences in the expression of HSD3B2 and CYP17A1 between the two sub-strains have led to markedly different steroid production profiles, with NCI-H295A cells secreting mainly mineralocorticoids and glucocorticoids, while the NCI-H295A cells produce androgens (DHEA, androstenedione, and testosterone) preferentially . Selection of an appropriate sub-strain based on these unique characteristics is therefore critical when employing either NCI-H295A or NCI-H295R cells in basic research.
Ovarian cell lines
There is a paucity of ovarian cell lines suitable for investigation on PCOS. The KGN cell line is of human ovarian granulosa origin and it was derived from a patient with invasive ovarian granulosa cell carcinoma . With the expression of functional follicle-stimulating hormone (FSH) receptors, it is responsive to both FSH and cAMP stimulation and retains the steroidogenic characteristics of granulosa cells. It secretes pregnenolone and progesterone readily, but not DHEA, androstenedione, or estradiol. Although aromatase expression and activity is high, the absence of androstenedione prevents the synthesis of estradiol. However, when androstenedione is provided exogenously, estradiol is readily produced and secreted by the KGN cells .
Suitability of the cellular model for studying human androgen biosynthesis
NCI-H295R
The human adrenal-derived cell line, NCI-H295R, is a well-established model for studying mechanisms controlling human androgen production and steroidogenesis . When cultured under the recommended growth conditions including serum and insulin, NCI-H295R cells produce all three types of adrenal steroid hormones, namely mineralocorticoids, glucocorticoids, and androgens . Interestingly, under serum-starved conditions, NCI-H295R cells adopt a hyperandrogenic steroid profile . Stimulation of the serum-starved NCI-H295R cells with cAMP increases androgen production further through transcriptional regulation and increased expression of essential steroidogenic enzymes including HSD3B2 and CYP17A1 . With the availability of molecular tools to enhance androgen production in these cells, serum-starved NCI-H295R cells provide a suitable in vitro model to study the regulation of androgen biosynthesis . In this regard, a dietary medium-chain fatty acid, decanoic acid (DA), which was previously shown to improve glucose tolerance and lipid profile in a mouse model of diabetes, was tested for its ability to modulate androgen biosynthesis, like metformin, using NCI-H295R cells. Notably, it was reported for the first time that DA, like metformin, can inhibit androgen biosynthesis in NCI-H295R steroidogenic cells by regulating the enzyme HSD3B2 in a cAMP stimulation-dependent manner. Specifically, both DA and metformin were shown to reduce cAMP-enhanced recruitment of Nur77 to the HSD3B2 promoter, coupled with decreased transcription and protein expression of HSD3B2 .
Insulin resistance and hyperinsulinemia have been shown to modulate molecular mechanisms implicated in androgenic hypersecretion, as seen in PCOS . As such, several studies have employed the NCI-H295R steroidogenesis model to study the mechanistic effects of insulin sensitizers (e.g., pioglitazone and metformin) on androgen production . Notably, using this cellular model, Kempná et al. reported that pioglitazone inhibits gene expression of CYP17A1 and HSD3B2 and represses basal and cAMP-stimulated activities of the CYP17A1 and HSD3B2 promoter reporters . More recently, metformin was found to inhibit androgen production in NCI-H295R cells by decreasing HSD3B2 expression and regulating complex I of the mitochondrial respiratory chain .
KGN
At first glance, selection of the KGN cell line to study human steroidogenesis in PCOS may seem more appropriate, as PCOS is primarily seen as an ovarian disorder. However, the KGN cells, being of granulosa origin, produce estrogens but not androgens , and their utility in the study of steroidogenesis is limited to estrogen production instead.
The KGN cell line is frequently employed to assess the signaling pathways , growth, and function of human granulosa cells . Studies on the effect of insulin sensitizers have also been conducted in the KGN cell line to assess the direct effects of the drugs on FSH action, aromatase activity, insulin signaling pathway, and glucose transport . Rice et al. demonstrated that metformin treatment of KGN cells enhanced the insulin-stimulated translocation of glucose transporter type 4 from the cytosol to the cell membrane . Metformin also repressed FSH-stimulated aromatase expression and activity in the KGN cells, an effect attributed to the inhibition of FSH receptor expression by metformin and independent of AMPK . Vistatin, a cytokine hormone and an enzyme involved in metabolic disorders (obesity, type II diabetes), was demonstrated to increase insulin-like growth factor 1-induced steroidogenesis in the KGN cells .
The cellular models of steroidogenesis have provided in-depth insight into the numerous pathways regulating steroid biosynthesis. They have also been useful tools to screen drug candidates and tease out mechanisms in a simple and cost-effective manner. In future, establishing and commercializing well-characterized cell lines, especially of ovarian theca origin, would contribute further to the study of androgen biosynthesis specifically in the ovaries.
Animal models of PCOS
The lack of a “gold standard” in animal models poses a key challenge in preclinical research on PCOS. Animal models of PCOS have been developed in a wide variety of species, including rhesus monkeys, sheep, and rats . Each of these species offers different benefits as a preclinical research model in terms of developmental time, cost-effectiveness, and translational relevance to humans.
Rhesus monkey PCOS models
With a genome that shares 93% similarity with humans, rhesus monkey models indisputably provide the greatest translational relevance for human disease . Their close resemblance to humans in terms of reproductive biology , metabolic physiology , developmental traits , and aging make them an optimal model for preclinical research, especially for studying PCOS where reproductive physiology and ovarian function are the focus of the investigations. In addition, their larger size compared to rodent models allows for the collection of larger blood volumes for sequential hormonal profiling and repeated ultrasound monitoring of ovarian follicular dynamics.
Prenatal exposure to testosterone produces a comprehensive adult PCOS-like phenotype in female rhesus monkeys . Daily subcutaneous injections of rhesus monkey dams have consistently resulted in hormonal, ovarian, and metabolic dysfunction that closely mimics PCOS in women, with the monkeys exposed to testosterone during early gestation presenting with the most severe PCOS-like phenotype ( Table 3 ).
Full access? Get Clinical Tree
