| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Article |
a Department of Physiology, The University of Adelaide, Adelaide, South Australia 5000, Australia
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Cortisol administration to hypophysectomized fetuses resulted in an enhanced degree of cytodifferentiation of adrenocortical cells [2], and Lye and Challis [8] also demonstrated that concurrent infusion of metopirone with ACTH(1-24) in intact fetuses prevented the cAMP accumulation measured following ACTH stimulation in fetal sheep adrenal cells in vitro. These authors concluded that cortisol may mediate the increase in adrenal responsiveness to subsequent ACTH stimulation in vitro that results after pulsatile ACTH(1-24) infusion in vivo [8]. While the actions of cortisol on intracellular events distal to cAMP generation are unknown, it appears that during a number of different conditions, intra-adrenal cortisol may play a role in mediating the actions of ACTH on fetal adrenal growth and steroidogenesis in late gestation.
Recent studies have shown that the action of cortisol in fetal tissues may be regulated via the two isoforms of the intracellular microsomal enzyme, 11ß hydroxysteroid dehydrogenase (11ßHSD). The reversible NADP(H)-dependent isoform, 11ßHSD type 1, can act either as a dehydrogenase or as a reductase. 11ßHSD type 2, however, is a unidirectional NAD-dependent enzyme that catalyses the conversion of the biologically active cortisol to the inert cortisone [9]. Thus the level of expression and direction of activity of 11ßHSD within a tissue may regulate tissue exposure to glucocorticoid action.
We have demonstrated in the fetal adrenal that the levels of 11ßHSD type 2 mRNA decrease during the last 10 days of gestation concomitantly with the prepartum increase in fetal plasma cortisol concentrations [10]. We have hypothesized that this decreased expression of 11ßHSD type 2 mRNA may be a consequence of the increase in cortisol; this could result in a positive feedback system operating within the fetal adrenal in the prepartum period such that there is increasing tissue exposure to any maturational actions of the endogenously generated glucocorticoids. In the present study we have therefore investigated the effect of intrafetal cortisol administration, before the normal prepartum cortisol surge, on the expression of 11ßHSD type 2 mRNA in the fetal adrenal. We also determined whether at this stage of gestation, increased fetal cortisol concentrations can stimulate growth of the fetal adrenal gland or increase the expression of key adrenal steroidogenic enzymes.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
All procedures were approved by the University of Adelaide Standing Committee on Ethics in Animal Experimentation. Twenty-five pregnant Border-Leicester x Merino ewes and their singleton fetuses were used in this study. Surgery was carried out at either 103 days or 104 days of gestation under general anesthesia and using aseptic techniques [4]. Catheters were implanted into the fetal carotid artery and jugular vein, filled with 50 IU/ml heparinized saline (heparin sodium; David Bull Laboratories, Mulgrave, Australia; saline: 0.9% NaCl solution; Baxter Healthcare, Old Toongabbie, Australia) and exteriorized via an incision in the ewe's flank. There was a recovery period of at least 3 days after surgery before fetal blood samples were collected, and all ewes were fed once daily between 0900 and 1300 h. At 116 days of gestation, ewes were killed with an overdose of sodium pentobarbitone (Lethabarb: 25 ml at 325 mg/ml; Vibrac Australia, Peakhurst, Australia); and fetal sheep were removed, weighed, and killed by decapitation. One adrenal gland from each fetus was quickly removed, weighed, snap frozen in liquid N2, and stored at -80°C until total RNA was extracted.
Infusion Regime and Blood Sampling Protocol
Infusion regime Cortisol (hydrocortisone succinate: 2.0–3.0 mg in 4.4 ml/24 h; SoluCortef; Upjohn, Rydalmere, Australia) was infused into fetal sheep between 109 and 116 days of gestation (cortisol-infused group; n = 12 fetuses). Saline (4.4 ml/24 h) was administered to control animals between 109 and 116 days (saline-infused group; n = 13 fetuses).
Blood sampling protocol Fetal (2 ml) arterial blood samples were collected into chilled collection tubes daily from fetuses in the saline- and cortisol-infused groups between 107 days and 116 days of gestation for cortisol RIA. Blood for cortisol assay was collected into tubes containing 125 IU lithium heparin (Sarstedt Australia, Inglefarm, Australia). Blood for immunoreactive (ir)-ACTH assay was collected into plain tubes containing EDTA (18.6 g/L of whole blood; Sarstedt) and aprotinin (1000 kallikrein inhibitor units in 100 µl/ml of whole blood; Sigma-Aldrich, St. Louis, MO). Blood samples were centrifuged at 1800 x g for 10 min at 4°C before separation and storage of plasma at -20°C for subsequent assay. Fetal arterial blood (0.5 ml) was collected on alternate days for measurement of whole blood paO2, paCO2, pH, O2 saturation, and hemoglobin content using an ABL 550 acid base analyzer and OSM2 hemoximeter (Radiometer Pacific, Findon, Australia). Fetal blood gas values remained in the normal range reported for healthy fetal sheep in late gestation [6] throughout the infusion period.
RIAs
Cortisol RIA
Cortisol concentrations were measured in fetal plasma samples from the saline-infused group (n = 13 fetuses; n = 125 samples) and cortisol-infused group (n = 12 fetuses; n = 107 samples). Total cortisol concentrations in fetal sheep plasma were measured using an RIA kit, validated for fetal sheep plasma (Orion Diagnostica, Turku, Finland). Prior to assay, cortisol was extracted from plasma with dichloromethane (BDH Laboratory Supplies, Poole, Dorset, UK) using a method described previously [11]. The efficiency of recovery of 125I-cortisol from fetal plasma using this method was >90%. The sensitivity of the assay was 0.78 nM/L, and the cross-reactivity of the rabbit anti-cortisol antibody was <1% with cortisone and 17
-hydroxyprogesterone and <0.001% with pregnenolone, aldosterone, progesterone, and estradiol. The inter- and intraassay coefficients of variation were <10%.
ACTH RIA
ir-ACTH concentrations were measured in fetal plasma samples from a subset of the saline-infused group (n = 6 fetuses; n = 34 samples) and cortisol-infused group (n = 9 fetuses; n = 54 samples). The concentrations of ir-ACTH were measured using an RIA kit (ICN Biomedicals Australasia, Seven Hills, Australia) [12]. The sensitivity of the assay was 7 pg/ml, and the rabbit anti-human ACTH(1-39) had a cross-reactivity of <0.1% with ß-endorphin, -melanocyte-stimulating hormone, 
-lipotropin, and ß-lipotropin. The interassay coefficient of variation (COV) was 14.6% and the intraassay COV was <10%.
Complementary DNA and Antisense Oligonucleotide Probes and Probe Labeling
Complementary DNA probes
Human (h) cytochrome P450 cholesterol side chain cleavage (CYP11A1) [13], cytochrome P450 17 hydroxylase (hCYP17) [14], and cytochrome P450 21 hydroxylase (hCYP21A1) [15] cDNA probes were generously provided by Professor W. Miller (Department of Pediatrics, UCSF, San Francisco, CA). A 3ß hydroxysteroid dehydrogenase (h3ßHSD) cDNA probe was donated by Dr. R. Rodgers (Department of Medicine, Flinders University, Australia). Complementary DNAs were radiolabeled with [-32P]dCTP (3000 Ci/mmol; GeneWorks, Adelaide, Australia) by the random priming oligomer method to a specific activity of 109 cpm/µg or greater, using a random primer kit (Pharmacia, North Ryde, Australia).
Oligonucleotide probes
A 45-mer antisense oligonucleotide probe for ovine 11ßHSD-2, complementary to nucleotides 1066–111, was synthesized (GeneWorks). A 30-mer antisense oligonucleotide probe for rat 18S rRNA, complementary to nucleotides 151–180, was also synthesized. Oligonucleotide probes were end-labeled by T4 polynucleotide kinase (Pharmacia, North Ryde, Australia) using [-32P]ATP (4000 Ci/mmol; GeneWorks) as substrate.
Total RNA Isolation
Total RNA was extracted from one adrenal from each of nine saline-infused (n = 9 fetuses) and nine cortisol-infused (n = 9 fetuses) fetal sheep by homogenization in 4 M guanidine hydrochloride solution (4 M guanidinium thiocyanate, Merck, Kilsyth, Australia; 25 mM sodium citrate, APS Ajax Finechem, Auburn, Australia; 0.5% w/v sodium laurylsarcosine, Sigma-Aldrich; 3.3 µl/ml of Sigma antifoam A, Sigma-Aldrich; and 1 µl/ml of ß-mercaptoethanol, BDH) and ultracentrifugation overnight at 36 000 rpm, through a cushion of 5.7 M CsCl (Boehringer Mannheim Australia, Castle Hill, Australia) [4]. Total RNA was reconstituted in sterile deionized distilled water, and nucleic acid purity and concentrations were quantified using a Beckman DU-50 spectrophotometer (Beckman Coulter Australia, Gladesville, Australia). Prior to Northern blot analysis, the integrity of the total RNA preparations was verified by subjecting 1 µl of each RNA sample to 1% agarose gel electrophoresis using molecular biology grade agarose (BDH) in single-strength Tris-acetate EDTA (40 mM Tris-acetate, 1 mM EDTA, pH 8.0; BDH), and staining with ethidium bromide (BDH). Total RNA preparations were stored at a concentration of around 5 µg/ml at -80°C until required for use.
Northern Blot Analysis
The radiolabeled hCYP11A1, hCYP17, hCYP21A1, and h3ßHSD cDNAs and the antisense o11ßHSD-2 oligonucleotide probes were used to probe two Northern blots of fetal adrenal total RNA from a subset of the saline-infused (n = 9 fetuses) and cortisol-infused (n = 9 fetuses) groups. One Northern blot membrane was hybridized sequentially with the hCYP11A1, hCYP17, and hCYP21A1 cDNAs, while the second membrane was hybridized sequentially with the h3ßHSD cDNA and o11ßHSD-2 oligonucleotide probes. The Northern blots were prepared as described previously [4]. In brief, total RNA samples (20 µg of adrenal RNA) were denatured by incubation in 2.2 M formaldehyde (APS Ajax Finechem) and 50% v/v formamide (BDH) at 55°C for 10 min and separated by electrophoresis in 1% agarose gels containing 2.2 M formaldehyde; they were then transferred by capillary blotting to Zetaprobe nylon membranes (Bio-Rad, Richmond, CA). Membranes were washed in 10-strength SSC (single-strength SSC is 0.15 M sodium chloride and 0.015 M sodium citrate), 0.1% SDS for 10 min at room temperature and baked for 1 h at 80°C, prior to prehybridization at 42°C. Membranes were hybridized sequentially for 20 h at 50°C in 30 ml of fresh hybridization buffer containing either 1–2 x 106 cpm/ml of the cDNA probe or 5 x 105 cpm/ml of the o11ßHSD-2 oligonucleotide probe, or the 18S rRNA oligonucleotide probe. Membranes were washed once (10 min) at room temperature in single-strength SSC, 0.1% SDS; then twice (10 min each time) in 0.1-strength SSC, 0.1% SDS at 42°C; they were then briefly air dried and sealed in a plastic bag. Membranes were exposed to phosphorimager plates in BAS 2040 cassettes (Berthold Australia, Bundoora, Australia) and then quantified on a Fuji-BAS 1000 phosphorimager scanner using Fuji MacBAS software (MacBas 2.2) (Berthold). Complementary DNA and antisense oligonucleotide probes were stripped from membranes between hybridizations by washing in 0.01-strength SSC, 0.5% SDS for 10 min at 80°C. Consistency of lane loading for each Northern gel was verified by a final hybridization of each membrane with the 30-mer antisense 18S rRNA oligonucleotide probe and exposure to phosphorimager plates in BAS 2040 cassettes. A ratio of the density of each specific band to the density of the corresponding 18S rRNA band was calculated before comparisons were made.
Statistical Analysis
All data are presented as the mean ± SEM. The ratios of 11ßHSD-2, CYP11A1, CYP17, CYP21A1, and 3ßHSD mRNA to 18S rRNA were compared between the saline- and cortisol-infused groups using two-tailed, unpaired Student's t-tests. Student's t-tests were also used to compare fetal body weight, adrenal weight, and the ratio adrenal:fetal body weight between saline- and cortisol-infused fetuses. Plasma hormone concentrations were compared using a two-way ANOVA with repeated measures, with treatment group (i.e., saline and cortisol infusion) and time (i.e., preinfusion and infusion) as the specified factors. A probability of <5% (P < 0.05) was considered to be significant.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Fetal plasma cortisol concentrations Cortisol infusion resulted in a significant increase in plasma cortisol concentrations from preinfusion values (preinfusion: 1.5 ± 0.1 nM/L; postinfusion: 39.3 ± 2.8 nM/L). Plasma cortisol concentrations were significantly higher in the cortisol-infused fetuses than in the saline-infused group (1.6 ± 0.1 nM/L) throughout the infusion period. Plasma ir-ACTH concentrations did not change during infusion of either saline (preinfusion: 41.35 ± 4.23 pg/ml; infusion: 32.68 ± 2.48 pg/ml) or cortisol (preinfusion: 35.28 ± 3.94 pg/ml; infusion: 30.76 ± 3.07 pg/ml). There was no significant difference in the plasma ir-ACTH concentrations between the two treatment groups.
Adrenal and Fetal Body Weights
There was no significant effect of cortisol infusion on fetal body weight (cortisol: 2.21 ± 0.08 kg; saline: 2.17 ± 0.13 kg) or on total adrenal weight (cortisol: 223.0 ± 11.2 mg; saline: 238.2 ± 15.5 mg). There was also no effect of cortisol on adrenal weight when expressed relative to fetal body weight (cortisol: 101.7 ± 5.3 mg/kg; saline: 108.2 ± 4.3 mg/kg).
11ßHSD-2 mRNA Expression
A single 11ßHSD-2 mRNA transcript of ~2.0 kilobases (kb) was present in the Northern blot analyses of the total RNA extracted from fetal adrenals. The ratio adrenal 11ßHSD-2 mRNA:18S rRNA expression was significantly lower in fetal sheep infused with cortisol (0.75 ± 0.02) compared with saline (1.65 ± 0.14) (Fig. 1).
|
Steroidogenic Enzyme mRNA Expression
CYP11A1 (1.9 kb), CYP17 (1.7 kb), CYP21A1 (1.8 and 2.2 kb), and 3ßHSD (1.6 kb) mRNA transcripts were present in the Northern blot analyses of total RNA (Fig. 2). There was no significant effect of the cortisol infusion on the relative abundance of any of the steroidogenic enzyme mRNAs (Fig. 3).
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
In the adult sheep, 11ßHSD-2 mRNA is localized exclusively to the adrenal cortex and is expressed highly in the zona fasciculata and reticularis with relatively low expression in the zona glomerulosa [16]. We have shown previously that 11ßHSD-2 mRNA levels in the fetal sheep adrenal decrease after 125 days gestation concomitantly with the prepartum cortisol surge [10]. The findings of the present study indicate that a physiological elevation in fetal cortisol concentrations at 116 days gestation can result in a premature decrease in the expression of adrenal 11ßHSD-2 mRNA. It appears that the decrease in 11ßHSD-2 mRNA levels in response to cortisol is not through an indirect action of cortisol at the fetal pituitary, as there was no change in circulating ACTH in response to the intrafetal cortisol infusion. It has been reported that intrafetal infusion of cortisol before the prepartum activation of the fetal pituitary-adrenal axis inhibits fetal ACTH responses to stimulation by corticotrophin releasing factor [17] or acute stressors such as hypoxemia [18], but does not suppress basal ACTH secretion or steady-state expression of the ACTH precursor, proopiomelanocortin (POMC), in the fetal pituitary [19]. In contrast, cortisol infusion after 140 days gestation suppresses basal and stimulated ACTH concentrations [17, 20]. Thus at around 110 days of gestation, the fetal hypothalamo-pituitary-adrenal axis appears relatively insensitive to negative feedback by physiological elevations in glucocorticoid concentrations in fetal sheep at this gestational age range under basal conditions.
It is possible that 11ßHSD-2 within the fetal sheep adrenal protects the adrenocortical cells from the high levels of locally produced glucocorticoids up until around 125 days gestation. The decrease in adrenal 11ßHSD-2 mRNA levels in late gestation is, however, coincident with the prepartum increase in adrenocortical growth and steroid output. This prepartum increase in fetal adrenocortical function may therefore require enhanced intra-adrenal exposure to glucocorticoids. In the present study, however, there was no increase in fetal adrenal weight or in steroidogenic enzyme expression in association with the decrease in adrenal 11ßHSD-2 mRNA levels at 116 days gestation. In contrast to the current findings, we have previously reported that cortisol infusion in HPD fetuses at 135–140 days gestation did stimulate an increase in fetal adrenal weight to values comparable with those in intact fetal sheep of the same age [7]. One possibility is that surgical disconnection of the fetal hypothalamus results in a decrease in the circulating concentrations of a pituitary-derived inhibitory factor that is present and blocks the action of cortisol on fetal adrenal growth in fetal sheep with an intact hypothalamic-pituitary axis. It has been proposed that there are inhibitory factors normally present in the fetal circulation that limit the responsiveness of the fetal adrenal to ACTH until late in gestation, when increasing ACTH concentrations override the tonic pituitary inhibition of the adrenal [21]. Potential candidates for adrenal inhibitory factors include the ACTH precursor POMC and pro-ACTH. POMC and pro-ACTH are present in high concentrations in the fetal circulation [21] and inhibit the ACTH-induced secretion of cortisol from ovine fetal adrenal cells in vitro [22]. ACTH precursors are also present in the circulation of the HPD fetus [4], but it is unclear whether the biological activity of these hormones is similar in HPD and intact animals. It should be noted, however, that cortisol does not stimulate adrenal growth in hypophysectomized fetuses [2], which makes it unlikely that a pituitary-derived inhibitor is the sole explanation for the lack of effect of cortisol on adrenal growth in intact fetal sheep.
In the present study, cortisol infusion did not alter the expression of adrenal steroidogenic enzyme mRNA levels in the fetal sheep at 116 days gestation. We have reported that while cortisol infusion increased fetal adrenal weight in HPD fetal sheep, there was no effect of cortisol on adrenal steroidogenic enzyme expression in these animals. It has been demonstrated in studies in vitro that prior treatment of fetal adrenal cells with dexamethasone enhanced the corticosteroid response to ACTH(1-24), forskolin, and cAMP [23]. Exposure of adult ovine adrenal cells to dexamethasone in vitro also increased ACTH receptor expression [24] and enhanced ACTH-stimulated translocation of cholesterol from the cytoplasm into the mitochondria [25]. A key difference between the steroidogenic effects of glucocorticoids in vivo and in vitro may relate to a requirement for synergy of action between glucocorticoids and ACTH that is present only when ACTH concentrations are relatively high as occurs either in vitro or in vivo after 135 days gestation.
In summary, we have found that a premature elevation in cortisol resulted in a suppression of 11ßHSD-2 mRNA levels in the fetal adrenal at 116 days gestation. Cortisol infusion had no effect, however, on fetal adrenal weight or on adrenal steroidogenic enzyme expression. Increased intra-adrenal exposure to cortisol at this stage of gestation is therefore not sufficient to promote adrenal growth or steroidogenic enzyme gene expression. Under conditions in which the fetal adrenal is stimulated by increased fetal ACTH as in the week before birth [1], or during chronic intrauterine stress [26], a fall in adrenal 11ßHSD-2 expression and increased intra-adrenal exposure to glucocorticoids may be important in enhancing the local growth factor and steroidogenic responses to the prevailing high ACTH concentrations.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
1 We acknowledge financial support from the National Heart Foundation of Australia and the Australian Research Council. 
2 Correspondence: I.C. McMillen, Department of Physiology, The University of Adelaide, Fourth Floor North Wing, Medical School Building, Frome Road, Adelaide, SA 5000, Australia. FAX: 08 8303 3356; caroline.mcmillen{at}adelaide.edu.au
Accepted: December 16, 1999.
Received: September 9, 1999.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  K.E. Warnes, I.C. McMillen, J.S. Robinson, and C.L. Coulter Differential Actions of Metyrapone on the Fetal Pituitary-Adrenal Axis in the Sheep Fetus in Late Gestation Biol Reprod, August 1, 2004; 71(2): 620 - 628. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A L Fowden and A J Forhead Endocrine mechanisms of intrauterine programming Reproduction, May 1, 2004; 127(5): 515 - 526. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. J. W. Fletcher, X. H. Ma, W. X. Wu, P. W. Nathanielsz, H. H. G. McGarrigle, A. L. Fowden, and D. A. Giussani Antenatal glucocorticoids reset the level of baseline and hypoxemia-induced pituitary-adrenal activity in the sheep fetus during late gestation Am J Physiol Endocrinol Metab, February 1, 2004; 286(2): E311 - E319. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
