19-hydroxy Steroids in the Aromatase Reaction: Review on Expression and Potential Functions

Scientific evidence related to the aromatase reaction in various biological processes spanning from mid-1960 to today is abundant; however, as our analytical sensitivity increases, a new look at the old chemical reaction is necessary. Here, we review an irreversible aromatase reaction from the substrate androstenedione. It proceeds in 3 consecutive steps. In the first 2 steps, 19-hydroxy steroids are produced. In the third step, estrone is produced. They can dissociate from the enzyme complex and either accumulate in tissues or enter the blood.

In this review, we want to highlight the potential importance of these 19-hydroxy steroids in various physiological and pathological conditions. We focus primarily on 19-hydroxy steroids, and in particular on the 19-hydroxyandrostenedione produced by the incomplete aromatase reaction. Using a PubMed database and the search term “aromatase reaction,” 19-hydroxylation of androgens and steroid measurements, we detail the chemistry of the aromatase reaction and list previous and current methods used to measure 19-hydroxy steroids.

We present evidence of the existence of 19-hydroxy steroids in brain tissue, ovaries, testes, adrenal glands, prostate cancer, as well as during pregnancy and parturition and in Cushing’s disease. Based on the available literature, a potential involvement of 19-hydroxy steroids in the brain differentiation process, sperm motility, ovarian function, and hypertension is suggested and warrants future research.

We hope that with the advancement of highly specific and sensitive analytical methods, future research into 19-hydroxy steroids will be encouraged, as much remains to be learned and discovered.

Adrenal androgens, adrenarche, and zona reticularis: A human affair?

In humans, reticularis cells of the adrenal cortex fuel the production of androgen steroids, constituting the driver of numerous morphological changes during childhood. These steps are considered a precocious stage of sexual maturation and are grouped under the term "adrenarche". This review describes the molecular and enzymatic characteristics of the zona reticularis, along with the possible signals and mechanisms that control its emergence and the associated clinical features. We investigate the differences between species and discuss new studies such as genetic lineage tracing and transcriptomic analysis, highlighting the rodent inner cortex's cellular and molecular heterogeneity. The recent development and characterization of mouse models deficient for Prkar1a presenting with adrenocortical reticularis-like features prompt us to review our vision of the mouse adrenal gland maturation. We expect these new insights will help increase our understanding of the adrenarche process and the pathologies associated with its deregulation.

Progestin withdrawal at parturition in the mare

Mammalian pregnancies need progestogenic support and birth requires progestin withdrawal. The absence of progesterone in pregnant mares, and the progestogenic bioactivity of 5α-dihydroprogesterone (DHP), led us to reexamine progestin withdrawal at foaling. Systemic pregnane concentrations (DHP, allopregnanolone, pregnenolone, 5α-pregnane-3β, 20α-diol (3β,20αDHP), 20α-hydroxy-5α-dihydroprogesterone (20αDHP)) and progesterone) were monitored in mares for 10days before foaling (n=7) by liquid chromatography-mass spectrometry. The biopotency of dominant metabolites was assessed using luciferase reporter assays. Stable transfected Chinese hamster ovarian cells expressing the equine progesterone receptor (ePGR) were transfected with an MMTV-luciferase expression plasmid responsive to steroid agonists. Cells were incubated with increasing concentrations (0-100nM) of progesterone, 20αDHP and 3α,20βDHP. The concentrations of circulating pregnanes in periparturient mares were (highest to lowest) 3α,20βDHP and 20αDHP (800-400ng/mL respectively), DHP and allopregnanolone (90 and 30ng/mL respectively), and pregnenolone and progesterone (4-2ng/mL). Concentrations of all measured pregnanes declined on average by 50% from prepartum peaks to the day before foaling. Maximum activation of the ePGR by progesterone occurred at 30nM; 20αDHP and 3α,20βDHP were significantly less biopotent. At prepartum concentrations, both 20αDHP and 3α,20βDHP exhibited significant ePGR activation. Progestogenic support of pregnancy declines from 3 to 5days before foaling. Prepartum peak concentrations indicate that DHP is the major progestin, but other pregnanes like 20αDHP are present in sufficient concentrations to play a physiological role in the absence of DHP. The authors conclude that progestin withdrawal associated with parturition in mares involves cessation of pregnane synthesis by the placenta.

Equine 5α-reductase activity and expression in epididymis

The 5α-reductase enzymes play an important role during male sexual differentiation, and in pregnant females, especially equine species where maintenance relies on 5α-reduced progesterone, 5α-dihydroprogesterone (DHP). Epididymis expresses 5α-reductases but was not studied elaborately in horses. Epididymis from younger and older postpubertal stallions was divided into caput, corpus and cauda and examined for 5α-reductase activity and expression of type 1 and 2 isoforms by quantitative real-time polymerase chain reaction (qPCR). Metabolism of progesterone and testosterone to DHP and dihydrotestosterone (DHT), respectively, by epididymal microsomal protein was examined by thin-layer chromatography and verified by liquid chromatography tandem mass spectrometry (LC-MS/MS). Relative inhibitory potencies of finasteride and dutasteride toward equine 5α-reductase activity were investigated. Pregnenolone was investigated as an additional potential substrate for 5α-reductase, suggested previously from in vivo studies in mares but never directly examined. No regional gradient of 5α-reductase expression was observed by either enzyme activity or transcript analysis. Results of PCR experiments suggested that type 1 isoform predominates in equine epididymis. Primers for the type 2 isoform were unable to amplify product from any samples examined. Progesterone and testosterone were readily reduced to DHP and DHT, and activity was effectively inhibited by both inhibitors. Using epididymis as an enzyme source, no experimental evidence was obtained supporting the notion that pregnenolone could be directly metabolized by equine 5α-reductases as has been suggested by previous investigators speculating on alternative metabolic pathways leading to DHP synthesis in placenta during equine pregnancies.

Follicle growth, ovulation, and luteal formation in primates and rodents: A comparative perspective

Ovarian function has a great deal of functional overlap between species; antral follicles grow in response to FSH, ovulation involves proteolysis, and the steroidogenic pathway is largely the same. However, embedded in these similarities are important differences that reflect the evolutionary and natural history of species and may focus future research into these critical areas. This review compares ovarian function of rats and mice with primates, focusing on estradiol and follicle growth, steroidogenesis and rupture during the periovulatory interval, and the formation of a functional corpus luteum, drawing the conclusion that careful comparison of species yields more functional information about both than studying them in isolation.

The role of enzyme compartmentalization on the regulation of steroid synthesis

Steroidogenic enzymes can be compartmentalized at different levels, some by virtue of being membrane bound in specific intra-cellular compartments. Although both 3β-hydroxysteroid dehydrogenase/Δ(5)-Δ(4) isomerase (3β-HSD) and 17α-hydroxylase/17,20-lyase cytochrome P450 (P450c17) are expressed in the endoplasmic reticulum (ER) membrane, these proteins may still be spatially separated within this membrane system. Side chain cleavage cytochrome P450 (P450scc) is anchored to the inner mitochondrial membrane and this organelle is the major source of pregnenolone (P5) feeding steroidogenesis. Furthermore, steroidogenic enzymes can also be partitioned in different cells. Although well recognized, the effect of enzyme compartmentalization on the rate of steroid production and the balance of different steroids is unclear. This study uses mathematical modeling to investigate the effect of enzyme compartmentalization on steroid synthesis in a human-ovine-bovine model of steroid synthesis. The study shows that the spatial separation of steroidogenic enzymes within the ER has a minimal effect on the rate of steroid synthesis. The compartmentalization of the enzymes into different organelles of a cell creates cellular steroid gradients and can affect the balance of the different steroid products. The partitioning of steroidogenic enzymes in different cells reduces the rate of steroid synthesis. The greater is the distance between the cells that contain different enzymes, the more the rate of steroid synthesis is reduced. Additionally, when 3β-HSD is not in the same cell with P450scc (the P5 source) and P450c17, the ratio of the Δ(5)-pathway products' concentrations to the Δ(4)-pathway products' concentrations is increased. However, none of these levels of compartmentalization of steroidogenic enzymes alter the qualitative behaviors of steroid synthesis in response to variation in an enzyme activity or P5 supply.

Variation in 3β-hydroxysteroid dehydrogenase activity and in pregnenolone supply rate can paradoxically alter androstenedione synthesis

The 3β-hydroxysteroid dehydrogenase/Δ(5)-Δ(4) isomerase (3β-HSD) and 17α-hydroxylase/17,20-lyase cytochrome P450 (P450c17) enzymes are important in determining the balance of the synthesis of different steroids such as progesterone (P4), glucocorticoids, androgens, and estrogens. How this is achieved is not a simple matter because each of the two enzymes utilizes more than one substrate and some substrates are shared in common between the two enzymes. The two synthetic pathways, Δ(4) and Δ(5), are interlinked such that it is difficult to predict how the synthesis of each steroid changes when any of the enzyme activities is varied. In addition, the P450c17 enzyme exhibits different substrate specificities among species, particularly with respect to the 17,20-lyase activity. The mathematical model developed in this study simulates the network of reactions catalyzed by 3β-HSD and P450c17 that characterizes steroid synthesis in human, non-human primate, ovine, and bovine species. In these species, P450c17 has negligible 17,20-lyase activity with the Δ(4)-steroid 17α-hydroxy-progesterone (17OH-P4); therefore androstenedione (A4) is synthesized efficiently only from dehydroepiandrosterone (DHEA) through the Δ(5) pathway. The model helps to understand the interplay between fluxes through the Δ(4) and Δ(5) pathways in this network, and how this determines the response of steroid synthesis to the variation in 3β-HSD activity or in the supply of the precursor substrate, pregnenolone (P5). The model simulations show that A4 synthesis can change paradoxically when 3β-HSD activity is varied. A decrease in 3β-HSD activity to a certain point can increase A4 synthesis by favouring metabolism through the Δ(5) pathway, though further decrease in 3β-HSD activity beyond that point eventually limits A4 synthesis. The model also showed that due to the competitive inhibition of the enzymes' activities by substrates and products, increasing the rate of P5 supply above a certain point can suppress the synthesis of A4, DHEA, and 17OH-P4, and consequently drive more P5 towards P4 synthesis.