Understanding the mechanisms linking obesity with hypertension is important in the current obesity epidemic as it may improve therapeutic interventions. Plasma aldosterone levels are positively correlated with body mass index and weight loss in obese patients is reported to be accompanied by decreased aldosterone levels. This suggests a relationship between adipose tissue and the production/secretion of aldosterone. Aldosterone is synthesized principally by the adrenal glands, but its production may be regulated by many factors, including factors secreted by adipocytes. In addition, studies have reported local synthesis of aldosterone in extra-adrenal tissues, including adipose tissue. Experimental studies have highlighted a role for adipocyte-secreted aldosterone in the pathogenesis of obesity-related cardiovascular complications via the mineralocorticoid receptor. This review focuses on how aldosterone secretion may be influenced by adipose tissue and the importance of these mechanisms in the context of obesity-related hypertension.
Aurelie Nguyen Dinh Cat, Malou Friederich-Persson, Anna White and Rhian M Touyz
A Jamieson, J M C Connell and R Fraser
Glucocorticoid-suppressible hyperaldosteronism (GSH), first described in 1966 (Sutherland et al. 1966), is a rare cause of familial hypertension. It presents in young adults with hypertension, hypokalaemia and suppressed plasma renin activity (features caused by the excess activity of aldosterone secretion), and is distinguished from other forms of primary hyperaldosteronism by its autosomal dominant mode of inheritance and the reversal of all its clinical and biochemical abnormalities by the administration of small doses of the synthetic glucocorticoid dexamethasone (Connell et al. 1986). GSH is also characterized by abnormally elevated levels of 18-hydroxycortisol and 18-oxocortisol, the excretion of which also falls to normal following dexamethasone administration (Chu & Ulick, 1982; Ulick et al. 1983; Gomez-Sanchez et al. 1984). The study of the production of these unusual 18-hydroxylated steroids has led to a reappraisal of the late reactions in aldosterone and cortisol synthesis by the adrenal cortex,
A J Conley, W E Rainey and J I Mason
This study examined fetal steroidogenic enzyme expression and function during pregnancy in the pig. Northern and Western analyses were performed to detect the cytochrome P450 enzyme 17α-hydroxylase/17–20 lyase (P450c17) and that for cholesterol side-chain cleavage (P450scc), as well as 3β-hydroxysteroid dehydrogenase (3β-HSD) expression in several porcine fetal tissues. The data demonstrate higher steroidogenic enzyme expression in the fetal adrenal glands and testes than in the placenta at all stages of development examined. Although steroidogenic enzyme expression was maintained throughout gestation in both the fetal adrenals and the testes, adrenal P450c17 expression was higher in the early and late stages when compared with the intermediate stages of fetal development. The stimulation of fetal adrenal steroidogenic enzyme expression in the later stage fetuses was accompanied by increased expression of P450c17 in both the fetal testes and placenta. The expression of 3β-HSD by porcine fetal testes was low compared with that of the fetal adrenal gland at all stages of development. Adrenal explants and cultured cells secreted cortisol and androstenedione but much lower amounts of corticosterone, dehydroepiandrosterone and aldosterone. Secretion of cortisol and androstenedione by adrenal explants was maintained by ACTH for 5 days of culture but declined in controls. In cultured porcine fetal adrenal cells, ACTH and angiotensin II stimulated the secretion of multiple steroids. Porcine fetal testis explants and cultured cells secreted testosterone, dehydroepiandrosterone and androstenedione, but were only moderately responsive to trophic stimulation by LH. In general, the data suggest that the fetal adrenal glands and the fetal testes have the potential to contribute significantly to the production of steroids during pregnancy in pigs.
R Sirianni, BR Carr, S Ando and WE Rainey
A unique characteristic of the primate adrenal is the ability to produce 19-carbon steroids, often called the adrenal androgens. Although it is clear that the major human adrenal androgens, dehydroepiandrosterone (DHEA) and DHEA sulfate (DHEA-S), are produced almost solely in the adrenal reticularis, the mechanisms regulating production are poorly understood. Herein, we tested the hypothesis that the Src family of tyrosine kinases are involved in the regulation of adrenal androgen production. The NCI-H295R human adrenal cell line and primary human adrenal cells in culture were used to study adrenal androgen production and expression of enzymes involved in steroidogenesis. To examine the role of Src tyrosine kinase, cells were treated with PP2, a specific Src inhibitor. Alternatively, adrenal cells were transfected with an expression vector containing a dominant-negative form of Src. PP2 treatment inhibited basal cortisol production while significantly increasing the production of DHEA and DHEA-S (together referred to as DHEA(S)) in both adrenal cell models. The effect of PP2 on steroidogenesis occurred along with a rapid induction of steroidogenic acute regulatory (StAR) protein synthesis as revealed by Western analysis. Treatment with PP2 also increased mRNA levels for StAR, and cholesterol side-chain cleavage (CYP11A) and 17alpha-hydroxylase/17,20-lyase (CYP17) enzymes. Treatment of adrenal cells with the cAMP agonist dibutyryladenosine cyclic monophosphate (dbcAMP), stimulated the production of cortisol and DHEA(S). However, treatment of adrenal cells with a combination of PP2 and dbcAMP enhanced the production of DHEA(S) while inhibiting cortisol production. During dbcAMP treatment PP2 was able to augment the expression of CYP17 and to inhibit the induction of 3beta-hydroxysteroid dehydrogenase type 2 (HSD3B2) levels. Increasing the CYP17 to HSD3B2 ratio is likely to promote the use of steroid precursors for the production of DHEA(S) and not for cortisol. Taken together these data suggest that the inhibition of Src tyrosine kinases causes adrenal cells to adopt a reticularis phenotype both by the production of DHEA(S) and by the steroidogenic enzymes expressed.
Yewei Xing, C Richard Parker, Michael Edwards and William E Rainey
The adrenal glands are the primary source of minerocorticoids, glucocorticoids, and the so-called adrenal androgens. Under physiological conditions, cortisol and adrenal androgen synthesis are controlled primarily by ACTH. Although it has been established that ACTH can stimulate steroidogenesis, the effects of ACTH on overall gene expression in human adrenal cells have not been established. In this study, we defined the effects of chronic ACTH treatment on global gene expression in primary cultures of both adult adrenal (AA) and fetal adrenal (FA) cells. Microarray analysis indicated that 48 h of ACTH treatment caused 30 AA genes and 84 FA genes to increase by greater than fourfold, with 20 genes common in both cell cultures. Among these genes were six encoding enzymes involved in steroid biosynthesis, the ACTH receptor and its accessory protein, melanocortin 2 receptor accessory protein (ACTH receptor accessory protein). Real-time quantitative PCR confirmed the eight most upregulated and one downregulated common genes between two cell types. These data provide a group of ACTH-regulated genes including many that have not been previously studied with regard to adrenal function. These genes represent candidates for regulation of adrenal differentiation and steroid hormone biosynthesis.
M Shimojo, C B Whorwood and P M Stewart
11β-Hydroxysteroid dehydrogenase (11β-HSD) catalyses the interconversion of biologically active cortisol to inactive cortisone in man, and corticosterone to 11-dehydrocorticosterone in rodents. As such, this enzyme has been shown to confer aldosterone-selectivity on the mineralocorticoid receptor and to modulate cortisol/corticosterone access to the glucocorticoid receptor (GR). Two kinetically distinct isoforms of this enzyme have been characterized in both rodents and man; a low-affinity NADP(H)-dependent enzyme (11β-HSD1) which predominantly acts as an oxo-reductase and, more recently, a high-affinity NAD-dependent uni-directional dehydrogenase (11β-HSD2). In this study we have analysed the expression of both 11β-HSD1 and 11β-HSD2 isoforms in rat adrenal cortex and medulla and have investigated their possible roles with respect to glucocorticoid-regulated enzymes mediating catecholamine biosynthesis in adrenal medullary chromaffin cells.
Using a rat 11β-HSD1 probe and a recently cloned in-house mouse 11β-HSD2 cDNA probe, Northern blot analyses revealed expression of mRNA species encoding both 11β-HSD1 (1·4kb) and 11β-HSD2 (1·9kb) in the whole adrenal. Consistent with this, 11β-dehydrogenase activity (pmol 11-dehydrocorticosterone formed/mg protein per h, mean ± s.e.m.) in adrenal homogenates, when incubated with 50 nm corticosterone in the presence of 200 μm NAD, was 97·0 ± 9·0 and with 500 nm corticosterone in the presence of 200 μm NADP, was 98·0 ± 1·4 11-Oxoreductase activity (pmol corticosterone formed/mg protein per h) with 500 nm 11-dehydrocorticosterone in the presence of 200 μm NADPH, was 187·7 ± 31·2. In situ hybridization studies of rat adrenal cortex and medulla using 35S-labelled antisense 11β-HSD1 cRNA probe revealed specific localization of 11β-HSD1 mRNA expression predominantly to cells at the corticomedullary junction, most likely within the inner cortex. In contrast, 11β-HSD2 mRNA was more abundant in cortex versus medulla, and was more uniformly distributed over the adrenal gland. Negligible staining was detected using control sense probes.
Ingestion of the 11β-HSD inhibitor, glycyrrhizic acid (>100mg/kg body weight per day for 4 days) resulted in significant inhibition of adrenal NADP-dependent (98·0 ± 1·4 vs 42·5 ± 0·4) and NAD-dependent (97·0 ± 9·0 vs 73·2 ± 6·7) 11β-dehydrogenase activity and 11-oxoreductase activity (187·7 ± 31·2 vs 67·7 ± 15·3). However, while levels of 11β-HSD1 mRNA were similarly reduced (0·85 ± 0·07 vs 0·50 ± 0·05 arbitrary units), those for 11β-HSD2 remained unchanged (0·44 ± 0·03 vs 0·38 ± 0·01). Levels of mRNA encoding the glucocorticoid-dependent enzyme phenylethanolamine N-methyltransferase which catalyses the conversion of noradrenaline to adrenaline, were also significantly reduced in those rats given glycyrrhizic acid (1·12 ± 0·04 vs 0·78 ± 0·04), while those for the glucocorticoid-independent enzyme tyrosine hydroxylase (1·9 kb), which catalyses the conversion of tyrosine to DOPA, were unchanged (0·64 ± 0·04 vs 0·61 ± 0·04).
In conclusion, the rat adrenal gland expresses both 11β-HSD1 and 11β-HSD2 isoforms. 11β-HSDl gene expression is localized to the adrenal cortico-medullary junction, where it is ideally placed to regulate the supply of cortex-derived corticosterone to the medullary chromaffin cells. This, together with our in vivo studies, suggests that 11β-HSD1 may play an important role with respect to adrenocorticosteroid regulation of adrenaline biosynthesis. The role of 11β-HSD2 in the adrenal remains to be elucidated.
Fabio Luiz Fernandes-Rosa, Sheerazed Boulkroun and Maria-Christina Zennaro
Primary aldosteronism (PA), the most common form of secondary hypertension, is caused in the majority of cases by unilateral aldosterone-producing adenoma (APA) or bilateral adrenal hyperplasia. Over the past few years, somatic mutations in KCNJ5, CACNA1D, ATP1A1 and ATP2B3 have been proven to be associated with APA development, representing more than 50% of sporadic APA. The identification of these mutations has allowed the development of a model for APA involving modification on the intracellular ionic equilibrium and regulation of cell membrane potential, leading to autonomous aldosterone overproduction. Furthermore, somatic CTNNB1 mutations have also been identified in APA, but the link between these mutations and APA development remains unknown. The sequence of events responsible for APA formation is not completely understood, in particular, whether a single hit or a double hit is responsible for both aldosterone overproduction and cell proliferation. Germline mutations identified in patients with early-onset PA have expanded the classification of familial forms (FH) of PA. The description of germline KCNJ5 and CACNA1H mutations has identified FH-III and FH-IV based on genetic findings; germline CACNA1D mutations have been identified in patients with very early-onset PA and severe neurological abnormalities. This review summarizes current knowledge on the genetic basis of PA, the association of driver gene mutations and clinical findings and in the contribution to patient care, plus the current understanding on the mechanisms of APA development.
SM MacKenzie, CJ Clark, R Fraser, CE Gomez-Sanchez, JM Connell and E Davies
The terminal stages of cortisol and aldosterone production in the human adrenal gland are catalysed by the enzymes 11beta-hydroxylase and aldosterone synthase, which are encoded by the CYP11B1 and CYP11B2 genes respectively. Recent studies have suggested that aldosterone and cortisol are also made in other tissues such as the brain, heart and vascular system and may play a role in cardiovascular homeostasis. The aim of this study was to confirm the presence of these enzymes and localise them precisely in the rat brain. Reverse transcription-polymerase chain reaction (RT-PCR)/Southern blotting confirmed transcription of CYP11B1 and CYP11B2 in whole brain and hypothalamus minces from Wistar-Kyoto rats. 11beta-Hydroxylase and aldosterone synthase were immunolocalised in paraffin-embedded rat adrenal and brain sections using mouse monoclonal antibodies. Negative controls utilised a mouse monoclonal antibody raised against a non-mammalian epitope. In the brain, 11beta-hydroxylase and aldosterone synthase were detected in the cerebellum, especially the Purkinje cells, as well as the hippocampus. The specificities of the 11beta-hydroxylase and aldosterone synthase antibodies were confirmed by positive immunostaining of the relevant regions of the adrenal cortex. This is the first direct evidence that steroid hydroxylases involved in the final stages of corticosteroid biosynthesis are present in specific regions of the central nervous system.
The nuclear receptor family responds to a diverse group of ligands, including steroids, retinoids, thyroid hormone, prostaglandins and fatty acids. Previous sequence analyses of adrenal and sex steroid receptors indicate that they form a clade separate from other nuclear receptors. However, the relationships of adrenal and sex steroid receptors to each other and to their ancestors are not fully understood. We have used new information from androgen, estrogen, mineralocorticoid and progesterone receptors in fish to better resolve the phylogeny of adrenal and sex steroid receptors. Sequence divergence between fish and mammalian steroid receptors correlates with differences in steroid specificity, suggesting that phylogeny needs to be considered in evaluating the endocrine effects of xenobiotics. Among the vertebrate steroid receptors, the most ancient is the estrogen receptor. The phylogeny indicates that adrenal and sex steroid receptors arose in a jawless fish or a protochordate and that changes in the sequence of the hormone-binding domain have slowed considerably in land vertebrates. The retinoid X receptor clade is closest to the adrenal and sex steroid receptor clade. Retinoid X receptor is noteworthy for its ability to form dimers with other nuclear receptors, an important mechanism for regulating the action of retinoid X receptor and its dimerization partners. In contrast, the adrenal and sex steroid receptors bind to DNA as homodimers. Moreover, unliganded adrenal and sex steroid receptors form complexes with heat shock protein 90. Thus, the evolution of adrenal and sex steroid receptors involved changes in protein-protein interactions as well as ligand recognition.
S. M. Laird, J. P. Hinson, G. P. Vinson, N. Mallick, S. Kapas and R. Teja
The involvement of the calcium messenger system in the control of steroidogenesis in the rat and bovine adrenal cortex has been studied extensively. However the role of these second messengers in the control of human adrenocortical function is not established. This was therefore studied by incubating collagenase-dispersed human adrenocortical cells with the calcium ionophore A23187 and the protein kinase C activator phorbol 12-myristate 13-acetate (TPA). The effects of the calcium channel blocker verapamil on basal and stimulated steroidogenesis were also studied.
Both TPA (1 pmol/l–10 μmol/l) and A23187 (1 nmol/l–10 μmol/l) caused a dose-dependent increase in cortisol, aldosterone and corticosterone production. Verapamil (10 μmol/l) inhibited the increase in aldosterone, corticosterone and cortisol produced in response to ACTH(1–24), potassium, and desacetyl-αMSH. Unlike previous results in the rat, these effects were not specific for aldosterone secretion.
The results suggest that, as in other species, calcium mobilization and protein kinase C activation have a role in the control of steroidogenesis in the human adrenal cortex. However, in contrast to the rat, these mechanisms appear to be involved in the control of steroidogenesis in both the zona glomerulosa and inner zone cells.