|
|
||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Mechanisms of Hormone Action |
Departments of Physiology3
Obstetrics and Gynaecology4, University of Adelaide, Adelaide, South Australia 5000, Australia
| ABSTRACT |
|---|
|
|
|---|
adrenal cortex, adrenocorticotropic hormone receptor, anterior pituitary, pregnancy, steroid hormones
| INTRODUCTION |
|---|
|
|
|---|
-hydroxylase (CYP17) mRNA [6]. Conversely, the expression of these key branch-point enzymes in adrenal steroidogenesis is reduced in the adrenals of fetuses after fetal hypophysectomy and is restored in hypophysectomized fetuses treated with ACTH [7]. These data suggest that cortisol biosynthesis in the fetal adrenal is dependent on ACTH derived from the fetal pituitary. This is further supported by studies showing that intact neurovascular connections between the fetal hypothalamus and pituitary are required for the ontogenetic increase in ACTH (1-39) with adrenal functional development [5, 8]. After disconnection of the fetal pituitary from the hypothalamus, no increase was noted in circulating ACTH (1-39) concentrations or in adrenal CYP11A1, CYP17, and 3ß-hydroxysteroid dehydrogenase (3ßHSD) mRNA levels in the adrenal during late gestation [5]. Recent evidence also indicates that an upregulation of expression of mRNA levels of both the ACTH receptor (melanocortin type 2 receptor [MC2-R]) [9, 10] and steroidogenic acute regulatory protein (StAR) in the fetal adrenal [11] may also play a role in augmenting the actions of ACTH on adrenal steroidogenesis in late gestation. In the adult adrenal, StAR acts to transport cholesterol from the outer to the inner mitochondrial membrane, where CYP11A1 is localized [12], and StAR has been localized within the zona glomerulosa and fasciculata of the fetal sheep adrenal throughout late gestation [11]. In the fetal sheep adrenal, ACTH binding, ACTH-induced adenylate-cyclase activity, and expression of StAR mRNA and mature StAR protein each increase coincident with the prepartum increase in adrenal responsiveness to ACTH and steroid output [11, 13].
Additionally, recent studies have shown that the action of cortisol in fetal tissues may be regulated via the type 2 isoform of the intracellular microsomal enzyme, 11ß-hydroxysteroid dehydrogenase (11ßHSD2), which is a unidirectional, NAD-dependent enzyme that catalyzes the conversion of the biologically active cortisol to the inert cortisone [14]. Interestingly, a decrease is observed in 11ßHSD2 expression in the fetal sheep adrenal between 125 and 141 days of gestation, which is parallel with the increase in circulating cortisol concentrations [15]. Furthermore, intrafetal infusion of cortisol before 125 days of gestation also results in a decrease in 11ßHSD2 mRNA levels in the fetal adrenal [16]. This suggests that cortisol can act through regulation of this enzyme to enhance intracellular exposure to cortisol within the fetal adrenal during late gestation [16]. It is not clear, however, if an increase in intra-adrenal cortisol is required to mediate the actions of ACTH on adrenal growth and steroidogenesis during the prepartum stimulation of the fetal hypothalamo-pituitary-adrenal axis.
We have used an in vivo model to investigate the effects of high fetal ACTH concentrations on fetal adrenal MC2-R, StAR, 11ßHSD2, and steroidogenic enzyme mRNA expression in the presence of an inhibitor of endogenous cortisol biosynthesis. We have infused metyrapone, a competitive inhibitor of the steroidogenic enzyme, 11ß-hydroxylase, which catalyzes the formation of cortisol from 11-desoxycortisol, into fetal sheep between 125 and 140 days of gestation and measured fetal plasma cortisol, 11-desoxycortisol, and ACTH concentrations; pituitary pro-opiomelanocortin (POMC) mRNA and adrenal mRNA expression of MC2-R, StAR, 11ßHSD2, CYP11A1, CYP17, 3ßHSD, and cytochrome P450 21-hydroxylase (CYP21A1); and StAR protein in the fetal adrenal gland.
| MATERIALS AND METHODS |
|---|
|
|
|---|
All experiments were carried out according to the guidelines of the Standing Committee of Ethics and Animal Experimentation, The University of Adelaide Animal Ethics Committee. Twenty-four pregnant Merino ewes were used in the present study. Ewes were housed in single crates and kept under a 12L:12D photoperiod. Animals were fed once daily and were given water ad libitum throughout the entire length of the protocol. Ewes were fasted for 24 h before surgery. Surgery was performed to insert vascular catheters into the fetus and ewe as described previously [17].
Experimental Protocol
At 125 days of gestation, metyrapone (4.8 mmol/day in 0.6 M tartaric acid, n = 10), a competitive inhibitor of 11ß-hydroxylase, or vehicle (tartaric acid, 0.6 M; n = 11; saline control, n = 3), were infused (0.4 ml/h) into the fetal jugular vein. The infusion was maintained continuously for 15 days until postmortem at 140 days of gestation. On the first day of the infusion, fetal arterial blood (2.5 ml) was collected at 1000 h (designated as 60 min) and 1100 h (designated as 0 min) before the start of infusion. Throughout the infusion period, fetal arterial blood samples (2.5 ml) were collected daily, between 0900 and 1100 h, for blood gas and plasma hormone analysis. Blood samples (1.0 ml) were aliquoted into EDTA (10 µl, 100 000 kIU/ml; Sigma Chemical Co., St Louis, MO) or heparin-coated tubes (125 IU). Blood samples were centrifuged at 1500 x g for 10 min; the plasma was separated and frozen at 20°C until analysis. Blood samples (0.5 ml) were also collected for the measurement of fetal arterial PO2, PCO2, pH, SO2, and hemoglobin using an ABL 520 blood gas analyzer (Radiometer, Copenhagen, Denmark).
Radioimmunoassays
ACTH
Fetal plasma immunoreactive ACTH concentrations were measured using a radioimmunoassay kit from ICN Biomedicals (Seven Hills, NSW, Australia), which has been validated previously for use in fetal sheep plasma [18]. According to the manufacturer, the cross-reactivity of the rabbit anti-human ACTH (139) is less than 1% with ß-endorphin,
-melanocyte-stimulating hormone,
-lipotrophin, and ß-lipotrophin. The sensitivity of the assay was 9 pg/ml. The intra-assay coefficient of variation was less than 5%, and the interassay coefficient of variation was less than 15%.
11-Desoxycortisol Fetal plasma 11-desoxycortisol concentrations were measured using a radioimmunoassay kit (ICN Biomedicals). Because of the higher concentration of 11-desoxycortisol in the fetal plasma samples collected from metyrapone-infused animals, these samples were diluted in assay buffer (1:10, 0.1 M PBS; Sigma Chemical Co.; and 0.1% BSA and 0.1% NaN3; BDH Laboratories, Poole, Dorset, U.K.) before extraction, whereas samples from vehicle-infused animals, did not require dilution before extraction. 11-Desoxycortisol was extracted from standard solutions (10 µl) and plasma samples (2100 µl) using dichloromethane [19], and the efficiency of extraction was greater than 95%. Extracted standards and samples were then reconstituted in buffer (10 µl), and then 100 µl of buffer were added in place of steroid globulin-binding inhibitor and the remainder of the assay performed according to the manufacturer's directions (ICN Biomedicals). Extracted 11-desoxycortisol from increasing volumes of fetal sheep plasma was quantitatively recovered (105.4% ± 5.6%). The relationship between the expected concentration of 11-desoxycortisol (x) and the observed concentration of 11-desoxycortisol (y) in increasing volumes of fetal sheep plasma was described by the equation y = 1.2 · 2.6 (r = 0.999, P < 0.001). The rabbit anti-human 11-desoxycortisol was specified by the manufacturer to have a cross-reactivity of less than 0.3% with progesterone and pregnenolone sulfate. The cross-reactivity of the anti-11-desoxycortisol with cortisol was determined to be 0.25%. The mean binding of the anti-11-desoxycortisol to [125I]11-desoxycortisol in the absence of antigen was 67.4% ± 1.1%. The sensitivity of the assay was 0.5 nmol/L. The intraassay coefficient of variation was less than 5%, and the interassay coefficient of variation was less than 15%.
Cortisol
Fetal plasma samples from metyrapone- and vehicle-infused fetuses were extracted and assayed in duplicate as described previously [19, 20]. The efficiency of the recovery was greater than 85%. The cross-reactivity of the anticortisol with 11-desoxycortisol (Sigma Chemical Co.) was determined to be 3.7%. The intraassay coefficient of variation was less than 5%, and the interassay coefficient of variation was less than 20%. In addition, we have validated our cortisol assay for use in extracts of plasma from metyrapone-infused fetuses. Recovery of cortisol (0.00210.0 pmol) added to extracts of fetal plasma was 105.6% ± 1.7% (y = 0.06 + 1.15x, r = 0.998), and increasing volumes of extracted plasma resulted in a displacement curve that was parallel to the cortisol standard curve (observed/expected concentrations = 99.5% ± 8.4%; y = 0.06 + 0.8x, r = 0.99). For all plasma samples, the amount of cortisol and 11-desoxycortisol in each sample was then calculated using the following equations:
![]() |
Autopsy and Tissue Collection
At 140 days, ewes were killed using a lethal injection of sodium pentobarbitone (Lethabarb, 25 ml at 325 mg/ml; Vibrac Australia, Peakhurst, NSW, Australia). The uterus was removed via hysterectomy and the fetus by hysterotomy. Fetal weight and crown-rump length were measured. Fetuses were then decapitated, and fetal organs, including the adrenals and pituitary, were removed and weighed. The whole left adrenal, half the right adrenal, and the anterior pituitary (separated from the neurointermediate lobe) were snap-frozen in liquid nitrogen and stored at 80°C until Northern and Western blot analyses. The other half of the right adrenal was fixed in 4% paraformaldehyde in 0.1 M phosphate buffer (BDH Laboratories) for 24 h at 4°C before rinsing in 0.1 M PBS (twice for 30 min each time; Sigma Chemical Co.) and dehydration in 70% ethanol before processing for embedding in paraffin wax for histological analysis.
Immunohistochemical Localization of 3ßHSD
To determine the proportion of the adrenal gland comprised of the steroidogenic cells of the adrenal cortex, sections (35 µm) from the midglandular region of the right adrenal were stained with antisera raised against 3ßHSD. The 3ßHSD was localized using a HistoPlus immunostaining kit (Zymed, South San Francisco, CA), which uses a biotinylated secondary antibody and a streptavidin-horseradish peroxidase conjugate. The 3ßHSD primary antibody was raised in rabbits against human 3ßHSD [21] and was used at a concentration of 1:2000. An immunopure, metal-enhanced diaminobenzidine substrate (Pierce, Rockford, IL) was used as the chromogen to identify positive staining, and the sections (35 µm) were counterstained (Mayers Hematoxylin; Sigma Diagnostics, St. Louis, MO). Slides were then dehydrated, mounted, and coverslipped.
Morphometric Analysis
Morphometric analysis was performed on sections taken from the midglandular region of the right adrenal of metyrapone- and vehicle-infused animals. Serial sections of each adrenal were cut (thickness, 35 µm) and stained with hematoxylin and eosin to identify cell nuclei and cytoplasm, respectively. The density of cell nuclei in the adrenal cortex (zona glomerulosa and zona fasciculata) and medulla were measured in vehicle- and metyrapone-infused fetuses using the following method: First, the width and length of a defined area were measured in pixels on the image-analysis system at 40x magnification on an Olympus BHS microscope (Olympus, Richmond, SA, Australia) using a Panasonic KR222 camera (Panasonic, Welland, SA, Australia) connected to VideoPro imaging software (Leading Edge, Adelaide, SA, Australia). The width and length were converted to micrometers by multiplication with 0.2604 (40x magnification), as determined by the use of a hemocytometer to calculate the area. The total number of cell nuclei was counted in a defined area (686017 750 µm2) in 10 random fields of view at least 1 mm apart at a magnification of 40x and analyzed using VideoPro image-analysis software. For each animal, the number of cell nuclei per square micrometer was calculated by the total number of nuclei in 10 defined areas divided by the total area.
The width of the fetal adrenals from vehicle- and metyrapone-infused fetuses was measured in four separate sections using the VideoPro image-analysis system. To measure the diameter of the total adrenal, the distance from the edge of the adrenal capsule to the middle of the adrenal vein was determined and then multiplied by two. The width of the adrenal cortex was determined by measuring the distance from the interface of the adrenal capsule and zona glomerulosa to the interface of the zona fasciculata and adrenal medulla. Finally, the width of the adrenal medulla was determined by measuring the distance from the interface between the zona fasciculata and adrenal medulla to the edge of the adrenal vein. All data were collected as pixels. The data of the adrenal widths were transformed from pixels to micrometers through multiplication by 5.1282 (2x magnification), as determined by the use of a hemocytometer.
RNA Extraction
Total RNA was extracted from approximately 100 mg of tissue from each fetal adrenal gland and from the anterior lobe of each fetal pituitary gland using Sigma TriReagent method (TriReagent, Sigma-Aldrich Corp., Castle Hill, NSW, Australia) as described previously [11, 15].
cDNA and Oligonucleotide Probes
The POMC mRNA was detected using a 400-base pair (bp) ovine POMC cDNA probe [22]. Adrenal steroidogenic enzymes were detected using cDNA probes for human CYP11A1 (1.82 kilobase [kb]) [23], bovine CYP17 (1.2 kb) [24], human 3ßHSD (435 bp) [25], and human CYP21A1 (1.14 kb) [26]. The 11ßHSD2 mRNA was detected using a 45-mer oligonucleotide probe for ovine 11ßHSD2, complementary to nucleotides 10661110 [27]. The StAR mRNA was detected by an ovine StAR cDNA probe (404 bp) [28], and the MC2-R mRNA was detected using an ovine MC2-R cDNA (670 bp) generously donated by Dr Kathleen Mountjoy (University of Auckland, Auckland, New Zealand). The cDNA probes were radiolabeled by the random priming method with [
-P32]deoxycytidine triphosphate (3000 Ci/mmol; Perkin-Elmer Life Sciences, Rowville, VIC, Australia) and Klenow fragment (6.4 U/µl1) using an oligolabeling kit (Pharmacia, North Ryde, NSW, Australia). An antisense oligonucleotide probe complementary to the coding nucleotides of 151 180 of rat 18S rRNA [29] was used to confirm equal loading of RNA into each lane. The 11ßHSD2 and 18S probes were end-labeled with [
-32P]ATP (4000 Ci/mmol; Perkin-Elmer Life Sciences) using T4 polynucleotide kinase (7.9 U/µl; Pharmacia). The cDNA and oligonucleotide probes were purified using a NICK Sephadex G-50 column (Pharmacia).
Northern Blot Analysis
Total RNA (20 µg) was denatured by the addition of 2.2 M deionized formaldehyde and 50% deionized formamide and incubation at 55°C for 15 min. Northern blot analysis was performed as previously described [5, 11, 15, 16, 18]. After washing, membranes were allowed to dry, sealed in a plastic bag, and exposed to a blank Fuji Bas-IIIs PhosphorImager plate (Fuji Photo Co., Tokyo, Japan) from 30 min to 2 days. Autoradiographs of membranes were visualized using a Fuji-Bas PhosphorImager (Fuji-Bas, Tokyo, Japan), and intensity of the signal was quantified using Fuji Image Gauge software (Version 3.46; Fuji Photo Co., Tokyo, Japan).
Western Blot Analysis for StAR Protein
The content of StAR protein in fetal adrenal extracts from metyrapone- and vehicle-infused fetuses was determined by Western blot analysis essentially as described previously [30]. The StAR antibody was generously provided by Dr. D.B. Hales (University of Illinois, Chicago, IL) and has been fully characterized and shown to detect the mature, 30-kDa form of StAR protein in a range of species, including sheep, human, rat, and mouse adrenals and/or gonads [28, 31, 32]. In brief, the Western blot analysis was performed on adrenal extracts (50 µg of protein) using a rabbit polyclonal mouse StAR antibody (1:500) overnight at 4°C followed by a horseradish peroxidase-labeled rabbit immunoglobulin G (1:1000) and detected using an amplified Opti-CN kit (Bio-Rad, Richmond, CA). The Western blot membrane was analyzed using a Densitometer (GS-710 Calibrated Imaging Densitometer; Bio-Rad), quantified using Image-Analysis Software (Quantity-One 4.2.1; Bio-Rad), and data expressed as arbitrary densitometric units (AU) of StAR protein per microgram of total protein.
Statistical Analyses
All data are expressed as the mean ± SEM. The ratios of plasma cortisol to ACTH and of plasma cortisol to 11-desoxycortisol at each gestational age were used as a marker to determine the effectiveness of the metyrapone suppression of the 11ß-hydroxylase enzyme. Hormone data were log-transformed, when required, to reduce heterogeneity of variance as determined by the use of the Bartlett and Cochran tests. Plasma ACTH, 11-desoxycortisol, and cortisol values and the ratios of plasma cortisol to ACTH and of plasma cortisol to 11-desoxycortisol between 125 and 140 days were analyzed using a multifactorial ANOVA with repeated measures, with age and treatment as the specified variables using the Statistical Package for Social Sciences (SPSS, Inc., Chicago, IL) on a VAX mainframe computer system. When a significant interaction between major factors was identified by ANOVA, the data were split on the basis of the interacting factor and reanalyzed. For all analyses in which ANOVA identified significant differences between groups, a Duncan multiple-range post-hoc test was used to identify the differences between mean values. The morphometric measurements of the adrenal glands (cell density and adrenal, adrenocortical, and adrenomedullary widths) were compared between the metyrapone- and vehicle-treated groups using a Student unpaired t-test. The ratios of pituitary POMC mRNA to 18S rRNA and of adrenal MC2-R, StAR, 11ßHSD2, CYP11A1, CYP17, 3ßHSD, and CYP21A1 mRNAs to 18S rRNA as well as the amount of adrenal StAR protein were also compared between the vehicle- and metyrapone-infused groups using a Student unpaired t-test. The relationships of plasma ACTH, 11-desoxycortisol, and cortisol concentrations (measured on the day closest to autopsy) with adrenal CYP11A1 mRNA and CYP17 mRNA expression were determined using linear-regression analysis (SPSS, Inc.). A probability of 5% was taken to be significant (P < 0.05).
| RESULTS |
|---|
|
|
|---|
Before the start of the infusion, no significant difference was found in plasma ACTH, 11-desoxycortisol, and cortisol concentrations between fetuses assigned to the vehicle- and metyrapone-infused groups. During the period of infusion, the mean plasma ACTH concentrations were significantly higher (P < 0.05) in metyrapone (156.8 ± 134.4 pg/ml) than in vehicle-infused fetuses (81.0 ± 40.8 pg/ml) between 126 and 140 days. After the start of the infusion, plasma 11-desoxycortisol concentrations were significantly higher (P < 0.0001) in the metyrapone-infused compared to the vehicle-infused fetuses. In vehicle-infused fetuses, a significant (P < 0.001) and progressive increase was observed in plasma 11-desoxycortisol concentrations with increasing gestational age whereby, in metyrapone-infused fetuses, plasma 11-desoxycortisol concentrations were significantly higher (P < 0.001) at 136140 days of gestation than at any other time during the infusion period (Fig. 1A).
|
During the infusion period, no significant difference was found between plasma cortisol concentrations in the metyrapone- and vehicle-infused fetuses between 126 and 140 days of gestation. Plasma cortisol concentrations, however, increased with increasing gestational age in both the metyrapone- and vehicle-infused groups (Fig. 1B).
The ratio of plasma cortisol to 11-desoxycortisol concentrations was significantly lower (P < 0.0001) in the metyrapone-infused fetuses than in the vehicle-infused group between 126 and 140 days of gestation (Table 1). In the vehicle-infused group, a significant (P < 0.001) increase in the ratio of plasma cortisol to 11-desoxycortisol was found during the infusion period, in which this ratio was highest at 136140 days than at any other time during the infusion (Table 1). In addition, the ratio of plasma cortisol to 11-deoxycortisol was also significantly higher at 131135 days than at 126130 days in the vehicle-infused group (Table 1). In the metyrapone-infused group, the ratio of cortisol to 11-desoxycortisol was significantly (P < 0.005) higher at 130140 days compared with earlier in gestation (Table 1).
|
No difference was found in the ratio of the plasma cortisol to ACTH concentrations between the metyrapone- and vehicle-infused fetuses during the infusion period. The ratio of plasma cortisol to ACTH concentrations was significantly higher (P < 0.0001) between 136 and 140 days than at any time between 120 and 135 days of gestation in both metyrapone- and vehicle-infused fetuses (Fig. 2).
|
Fetal Outcome, Adrenal Weight, and Adrenal Morphometry
No significant difference in mean fetal body weight or crown-rump length was found between the metyrapone-infused and vehicle-infused fetal sheep (Table 2). The combined adrenal weight was significantly greater (P < 0.001) in the metyrapone-infused than in the vehicle-infused fetuses (Table 2). The adrenal cortex, but not the adrenal medulla, was significantly thicker (P < 0.05) in the metyrapone-infused than in the vehicle-infused fetuses (Table 2). Significantly fewer (P < 0.01) cell nuclei per square micrometer were found in the zona fasciculata of adrenals from the metyrapone-infused group (n = 6) compared with the vehicle-infused controls (n = 6) (Fig. 3). No difference, however, was found in the density of cell nuclei per square micrometer in either the zona glomerulosa or the adrenal medulla between the metyrapone- and vehicle-infused groups (Fig. 3).
|
|
Pituitary POMC mRNA
A single transcript (1.4 kb) was detected for POMC mRNA in the anterior pituitary from metyrapone-infused and control fetuses. The relative expression of POMC mRNA versus that of 18S rRNA was four- to fivefold higher (P < 0.005) in the anterior pituitary of metyrapone-infused fetuses when compared to vehicle-infused controls (Table 2).
Adrenal StAR mRNA and Protein Content
A single transcript (3.0 kb) for StAR mRNA was detected in all fetal adrenals, and the relative expression of StAR mRNA was significantly lower (P < 0.05) in the metyrapone-infused group (n = 8) than in the vehicle-infused controls (n = 6) (Fig. 4A, B). The mature, 30-kDa StAR protein band was detected by Western blot analysis in the protein extracts of adrenal glands from metyrapone- and vehicle-infused fetal sheep (Fig. 4C). The amount of the mature, 30-kDa StAR protein in fetal adrenals was significantly greater (P < 0.000) in the adrenal glands from metyrapone (n = 7, 1.06 ± 0.08 AU/µg adrenal protein) and vehicle (n = 7, 0.47 ± 0.06 AU/µg adrenal protein) infused fetal sheep (Fig. 4D).
|
Adrenal MC2-R, 11ßHSD2, CYP11A1, CYP17, 3ßHSD, and CYP21A1 mRNA
Both a major (3.5-kb) and a minor (2.0-kb) transcript were detected for MC2-R mRNA. No difference was found in the relative expression of the major MC2-R mRNA:18S rRNA transcript between the metyrapone-infused (72.4 ± 13.8, n =8) and vehicle-infused fetuses (53.2 ± 6.8, n = 6). A single transcript (1.8 kb) was detected for 11ßHSD2 mRNA, and the relative expression of 11ßHSD2 mRNA: 18S rRNA was significantly higher (P < 0.05) in metyrapone-infused fetuses (n =8) compared with vehicle-infused fetuses (n =6) (Fig. 5A). Single transcripts for adrenal CYP11A1 mRNA (2.0 kb), CYP17 mRNA (2.2 kb), and 3ßHSD mRNA (1.8 kb) were detected, and two major transcripts (2.2 and 1.8 kb) were detected for adrenal CYP21A1 mRNA. The relative expression of adrenal CYP11A1 mRNA:18S rRNA and of CYP17 mRNA:18S rRNA were significantly higher (P < 0.05) in the metyrapone-infused fetuses than in vehicle-infused controls (Fig. 5, B and C, respectively). No difference, however, was found in the relative expression of 3ßHSD mRNA versus that of 18S rRNA or in that of CYP21A1 (2.2 + 1.8 kb) mRNA versus that of 18S rRNA (P = 0.09) between metyrapone-infused (3ßHSD, 73.8 ± 3.3, n = 7; CYP21A1, 639 ± 80.4, n = 6) and vehicle-infused fetuses (3ßHSD, 69.9 ± 6.0, n = 7; CYP21A1, 451.7 ± 49.1, n = 8).
|
When the groups were combined, a significant positive correlation was found between plasma ACTH concentrations and the relative expression of CYP11A1 mRNA (CYP11A1 mRNA = 4.95 [ACTH] + 1397.8; P < 0.0001, r = 0.861) and that of CYP17 mRNA (CYP17 mRNA = 5.77 [ACTH] + 781.8; P < 0.005, r = 0.713).
| DISCUSSION |
|---|
|
|
|---|
Negative Feedback Between the Fetal Pituitary and Adrenal
In the metyrapone-infused fetuses, an increase was observed in fetal plasma ACTH concentrations, whereas circulating cortisol concentrations were maintained at values similar to those in the vehicle-infused group. One possibility is that as ACTH increases to the level required to overcome the steroidogenic enzyme block within the adrenal, fetal cortisol concentrations increase and act to suppress the central stimulation of ACTH synthesis and secretion. In the presence of metyrapone, however, any fall in ACTH would be followed by a fall in adrenal cortisol output, a restimulation of the fetal pituitary and consequent increase in fetal ACTH, and a restoration of fetal cortisol concentrations. These findings suggest that negative feedback is operating between the fetal adrenal and pituitary throughout late gestation and that increase occurs in the central stimulation of the pituitary in metyrapone-infused fetuses to maintain circulating cortisol concentrations at levels similar to those in the vehicle-infused fetuses during late gestation. The greater variability of plasma cortisol concentrations in the metyrapone-infused group may indicate that the interaction between the fetal pituitary and adrenal is not in steady state during the metyrapone-infusion period. Previously, we and others have also proposed that metyrapone may act directly on the fetal hypothalamus and/or pituitary rather than modulate cortisol negative feedback to stimulate the increase in fetal plasma ACTH concentrations [33, 34]. However, as we observed no significant effect of metyrapone treatment on the ratio of cortisol to ACTH throughout the infusion period, it would appear likely that the stimulation of fetal ACTH secretion is predominantly a result of the removal of cortisol negative feedback.
MC2-R and Steroidogenic Enzyme mRNA Expression
Interestingly, we found a gestational increase in the ratio of plasma cortisol to ACTH concentrations in both the metyrapone- and vehicle-infused fetal sheep after 136 days of gestation. This indicates an increase in adrenal responsiveness to the steroidogenic actions of ACTH in both treatment groups. Studies in cultured bovine adrenocortical cells have shown that ACTH stimulates the expression of its own receptor [35], and increased ACTH binding and ACTH-induced adenylate-cyclase activity occur in the fetal sheep adrenal gland with increasing gestational age [13]. In the present study, whereas plasma ACTH concentrations were higher in metyrapone fetuses between 125 and 140 days of gestation, no difference was found in adrenal MC2-R mRNA levels between metyrapone- and vehicle-infused fetuses. Simmonds et al. [7] demonstrated that the expression of MC2-R mRNA in the fetal sheep adrenal was unchanged after fetal hypophysectomy or intrafetal ACTH infusion. In a previous study, we also showed that no significant change occurred in the expression of MC2-R between 125 and 140 days of gestation [10]. Whereas an increase in adrenal expression of MC2-R mRNA in both metyrapone- and vehicle-infused fetuses may precede the increase in adrenal responsiveness to ACTH from 136 days of gestation, it appears that such an increase is not directly related to circulating ACTH concentrations once they exceed those in the circulation of the vehicle-infused group in late gestation. It has been reported that administration of metyrapone to the human adrenal cortical cell line NCI-295 resulted in decreased expression of the MC2-R mRNA [36], but we found no evidence for such a decrease in these in vivo studies. Significant increases, however, were found in the expression of adrenal CYP11A1 and CYP17 mRNA, but not in those of 3ßHSD or CYP21A1 mRNA, in the metyrapone-infused fetuses. Previously, it has been shown that adrenal CYP11A1, CYP17, and 3ßHSD mRNA levels decrease after fetal hypophysectomy and are restored following ACTH replacement [6, 7]. In the present study, we also showed that plasma ACTH concentrations were correlated significantly with adrenal expression of CYP11A1 and CYP17 mRNA. The increases in the mRNA levels of adrenal CYP11A1 and CYP17 in the metyrapone-infused fetuses therefore likely are a consequence of the increase in fetal ACTH concentrations. These in vivo findings are in contrast to previous in vitro studies in which it was shown that metyrapone directly inhibits CYP11A1 and CYP17 activity in the NCI-295 adrenocortical cell line [37, 38]. The impact of an increase in circulating ACTH likely may override any residual negative actions of metyrapone on the adrenal. The lack of a difference between 3ßHSD mRNA expression in the metyrapone- and vehicle-infused fetuses may reflect that maximal stimulation of this enzyme has occurred at the levels of ACTH present in the vehicle-infused fetuses during late gestation. This is consistent with evidence that intrafetal ACTH infusion increases the expression of 3ßHSD mRNA at 132 days but not at term [7].
We also found no significant difference in the expression of CYP21A1 in the adrenals from the metyrapone- and vehicle-infused fetuses. Intrafetal ACTH infusion does not stimulate adrenal expression of CYP21A1 mRNA, indicating that this enzyme may not be responsive to the increase in circulating ACTH that occurs during late gestation [6] or after metyrapone infusion. Thus, in the metyrapone-infused group, the fetal hypothalamo-pituitary-adrenal axis is reset to maintain the increase in circulating ACTH concentrations that is required to stimulate fetal adrenal steroidogenesis to overcome the metyrapone-induced block of cortisol biosynthesis.
Adrenocortical Growth
In the metyrapone-infused fetal sheep, a doubling in the thickness of the adrenal cortex occurred, along with an approximately 20% increase in cell size within the zona fasciculata. Metyrapone treatment for 37 days similarly resulted in an increase in both adrenal growth and adrenocortical cell size in the primate fetus [39]. The most likely factor responsible for the increased adrencortical growth in the metyrapone-infused fetuses is either ACTH or another trophic peptide derived from POMC. We have previously shown that infusion of POMC (177) stimulated adrenocortical growth, but not steroidogenesis, in late gestation [40]. Interestingly, infusion of POMC (177) also resulted in a decrease in StAR mRNA expression in the fetal sheep adrenal [41]. Whereas adrenocortical growth was increased markedly in the metyrapone-infused group, plasma cortisol concentrations were not significantly different between metyrapone- and vehicle-infused fetuses. This suggests that cortisol production may be relatively decreased from the adrenocortical cells of the enlarged adrenal in the metyrapone-treated group.
11ßHSD2 mRNA Expression
The relative expression of 11ßHSD2 mRNA was higher in metyrapone-infused fetuses compared with that in vehicle-infused controls. Previously, we reported that 11ßHSD2 mRNA expression in the fetal adrenal decreased coincident with the prepartum increase in adrenocortical growth and steroid output, and we suggested that this would result in an increasing adrenocortical exposure to endogenously generated glucocorticoids [15]. Furthermore, infusion of cortisol into fetal sheep at a stage in gestation when fetal ACTH and cortisol concentrations are normally low (i.e., between 109 and 116 days of gestation) resulted in a specific decrease in 11ßHSD2 mRNA expression in the fetal adrenal, which suggests that cortisol may act to decrease the expression of the enzyme that regulates its metabolism within the fetal adrenal [16]. Previous studies have found no evidence for a direct action of metyrapone on 11ßHSD2 activity in vitro [42]. Taken together, these studies would therefore suggest that the higher 11ßHSD2 mRNA levels in the adrenals of metyrapone-infused fetuses compared with the control group would be consistent with a reduction in intracellular cortisol concentrations within the adrenocortical cells of the metyrapone-infused fetuses.
StAR mRNA Expression and Protein Content
A differential effect of metyrapone administration on adrenal StAR mRNA and protein levels was observed. The regulation of StAR mRNA expression is complex and determined by the balance between transcription and mRNA turnover, each of which are regulated in turn by multiple factors [43]. Previous studies have shown that ACTH stimulates the expression of StAR mRNA in adrenal cells both in vitro and in vivo via a cAMP-dependent mechanism [35, 4446], and it has also been shown that an ontogenetic increase occurs in StAR mRNA and protein in the fetal sheep adrenal between 125 and 140 days of gestation, coincident with the prepartum increase in fetal ACTH and cortisol [11]. In vitro studies have shown, however, that glucocorticoids have a direct role in enhancing the action of phorbol ester on StAR mRNA levels in adrenocortical cell lines [47]. Interestingly, concurrent infusion of metyrapone with ACTH (124) in fetal sheep prevented the cAMP accumulation measured following ACTH stimulation in fetal sheep adrenal cells in vitro [48]. One possibility therefore is that whereas circulating ACTH concentrations are higher in the metyrapone-treated group, the actions of ACTH on adrenal StAR mRNA expression are limited by relatively lower intracellular cortisol concentrations within the fetal adrenal cortex in this group. Alternatively, as discussed above, metyrapone infusion may also result in increased secretion of other POMC-derived peptides, such as POMC (177), which have been shown to decrease the StAR mRNA expression in the fetal sheep adrenal [41]. It is also possible that metyrapone may directly affect StAR expression; however, data from in vitro studies using the adrenal H295R adrenal cell line reveal no evidence that StAR mRNA expression was modulated by metyrapone [49]. Whereas a decrease occurred in the steady-state levels of StAR mRNA levels, a significant increase occurred in the 30-kDa, mature form of the StAR protein in the adrenals of the metyrapone-infused fetuses. The p30 protein is derived from a p37 precursor, which is phosphorylated in response to cAMP and then processed in the mitochondria to the phosphorylated p30 form [43]. The observation of the dissociation of the effects of metyrapone on StAR mRNA and protein expression in vivo is novel, and this may indicate that the stimulatory actions of ACTH on StAR transcription, mRNA stability, and posttranslational modification may, in part, depend on intracellular cortisol concentrations within the adrenocortical cell.
In summary, metyrapone administration in the late-gestation sheep fetus results in an increase in circulating ACTH, adrenal growth, and expression of adrenal steroidogenic enzymes. Plasma cortisol concentrations, however, are not different between metyrapone and vehicle-infused fetuses throughout late gestation. We propose that cortisol production by the adrenocortical cells is relatively decreased in this model and that low intracellular cortisol concentrations are associated with the observed increase in adrenal 11ßHSD2 mRNA and decrease in adrenal StAR mRNA expression. These findings suggest that the actions of both ACTH and cortisol are required to generate the increases in adrenal growth (ACTH), steroidogenesis (ACTH), and StAR (cortisol) and the decrease in 11ßHSD2 (cortisol) that result in generation of the prepartum cortisol increase, which is essential for a successful transition at birth. Thus, infusion of metyrapone may represent a unique model that allows the dissociation of the relative actions of ACTH and cortisol on fetal adrenal steroidogenesis and growth during late gestation.
| ACKNOWLEDGMENTS |
|---|
| FOOTNOTES |
|---|
2 Correspondence: C.L. Coulter, Discipline of Physiology, School of Molecular and Biomedical Sciences Research Centre for the Early Origins of Adult Health, University of Adelaide, Adelaide, SA 5000, Australia. FAX: 61 8 8303 3356; catherine.coulter{at}adelaide.edu.au ![]()
Received: 5 November 2003.
First decision: 28 November 2003.
Accepted: 30 March 2004.
| REFERENCES |
|---|
|
|
|---|
-hydroxylase in the sheep fetus during late gestation. Endocrinology 2000 141:2153-2158This article has been cited by other articles:
![]() |
J. T. Ross, I. C. McMillen, F. Lok, A. G. Thiel, J. A. Owens, and C. L. Coulter Intrafetal Insulin-Like Growth Factor-I Infusion Stimulates Adrenal Growth But Not Steroidogenesis in the Sheep Fetus during Late Gestation Endocrinology, November 1, 2007; 148(11): 5424 - 5432. [Abstract] [Full Text] [PDF] |
||||
![]() |
L. C. Carey, Y. Su, N. K. Valego, and J. C. Rose Infusion of ACTH stimulates expression of adrenal ACTH receptor and steroidogenic acute regulatory protein mRNA in fetal sheep Am J Physiol Endocrinol Metab, August 1, 2006; 291(2): E214 - E220. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. A. Myers, K. Hyatt, M. Mlynarczyk, I. M. Bird, and C. A. Ducsay Long-term hypoxia represses the expression of key genes regulating cortisol biosynthesis in the near-term ovine fetus Am J Physiol Regulatory Integrative Comp Physiol, December 1, 2005; 289(6): R1707 - R1714. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |