| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLE |
Department of Physiology and Functional Genomics, University of Florida College of Medicine (C.E.W., K.E.G.), and Department of Pharmacodynamics, University of Florida College of Pharmacy (M.K.-W.), Gainesville, Florida 32610-0274
Address all correspondence and requests for reprints to: Charles E. Wood, Ph.D., Department of Physiology and Functional Genomics, P.O. Box 100274, University of Florida College of Medicine, Gainesville, Florida 32610-0274. E-mail: cwood{at}phys.med.ufl.edu.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
We have demonstrated that estrogen has a strong stimulatory effect on the fetal HPA axis (3, 4, 5). Elevation of the fetal plasma estradiol concentration to well within the physiological range increases both basal and stimulated fetal ACTH secretion and elevates the fetal plasma cortisol concentration (3). We have proposed that the interplay between placental estrogen production and the activity of the fetal HPA axis constitutes a positive feedback loop (3). According to this view, increased activity of the fetal HPA axis increases placental production of estradiol, which, in turn, further increases the activity of the fetal HPA axis. The culmination of this process is labor and delivery of the fetus.
Fetal plasma is rich in sulfoconjugated estrogens. The abundance of estrone-3-sulfate in fetal and maternal plasma, for example, has been appreciated for several years (6, 7). We have proposed that sulfoconjugated estrogens are biologically active in the fetal brain and anterior pituitary (8). We have demonstrated the presence of steroid sulfatase (STS; estrogen sulfatase) in high abundance and activity in fetal brain regions, and we have localized the enzyme to both neurons and vascular endothelial cells in regions of the brain involved in control of the fetal HPA axis (8). Interestingly, little is known about the presence of 17ß-estradiol-3-sulfate in ovine fetal plasma. If this steroid did circulate in fetal plasma, it would provide a ready source of estradiol within the regions in the fetal brain that control the HPA axis. The present study was designed, therefore, to test the hypothesis that this steroid circulates in fetal plasma and that it is biologically active with regard to stimulation of the fetal HPA axis. We also report the results of experiments designed to test the hypothesis that sulfoconjugated estrogens are taken up by the fetal brain.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Study 1: fetal responses to estradiol-3-sulfate infusion
Each fetus used in the first study was chronically catheterized using surgical methods that have been described previously (9). Briefly, we used aseptic technique to chronically implant polyvinyl chloride catheters into the vasculature of the fetus. Fetuses were 113–119 d gestation at the time of surgery. Catheters (outside diameter, 0.090 in.; inside diameter, 0.050 in.) were placed in the fetal saphenous veins, and the tips of the catheters were advanced to the abdominal inferior vena cava. Smaller catheters (outside diameter, 0.050 in.; inside diameter, 0.030 in.) were placed in the fetal tibial arteries, and the tips were advanced to the abdominal aorta. A catheter (outside diameter, 0.090; inside diameter, 0.050) was sutured to the fetal skin for access to the amniotic fluid. Fetal catheters exited the ewe at the flank and were maintained in a disposable pocket that was held to the skin of the ewe with a commercially available bandage material (Spandage, Medi-Tech International Corp., Brooklyn, NY). Antibiotic (ampicillin, 750 mg, sc, twice daily) was administered to ewes (750 mg, sc, twice daily) and to fetuses (750 mg, intraamniotically, twice daily) during the first 5 d postoperatively. Fetuses were not studied within 5 d of surgery.
Each fetus was subjected to an iv infusion of saline (n = 5; Baxter Healthcare, Deerfield, IL) or estradiol-3-sulfate (Sigma-Aldrich, St. Louis, MO) at a rate of 0.25 (n = 4) or 1.0 (n = 5) mg/d for 5 d. Immediately before and at the end of 5 d of infusion, each fetus was subjected to a 1-h period of blood sampling in which 13 arterial blood samples (3 ml each) were withdrawn at 5-min intervals. Blood samples were drawn at relatively frequent intervals because of the known ultradian rhythm in fetal plasma cortisol concentrations in fetal sheep (10) and because of studies in this laboratory that demonstrated variations in fetal plasma concentrations of both ACTH and cortisol that are consistent with a robust ultradian rhythm (3). An additional arterial blood sample (1 ml) was withdrawn at the beginning of the blood-sampling period to assess the fetus. In some cases an additional blood sample was withdrawn at the end of the blood-sampling period to justify the assumption that the blood sampling alone did not produce any fetal hypoxia, hypercapnia, or acidemia. At the end of the experiment, the ewe and the fetus were humanely euthanized using an overdose of sodium pentobarbital administered to the ewe iv. The brain of the fetus was rapidly removed and dissected, and brain regions of interest were rapidly frozen in liquid nitrogen, then stored at -80 C until used.
Fetal blood was collected in chilled glass tubes containing sodium EDTA and kept on ice until centrifugation at 3000 x g for separation of plasma and red blood cells. Plasma was kept frozen in aliquots at -20 C until analysis by RIA. ACTH was measured by RIA after extraction of the peptide from plasma as previously described (11). The cortisol concentration was measured using RIA after extraction of the steroid from plasma with ethanol, also as previously described (12). The estradiol concentration was measured using enzyme-linked immunoassay (Oxford Chemical Co., Oxford, MI; catalog no. EA70) after extraction of plasma using hexane/ethyl acetate (3:2, vol/vol) (3). Recovery was 80–90%; calculated plasma concentrations were not adjusted for recovery. The antiserum used in the estradiol assay cross-reacted less than 1% with all other major estrogens in ovine plasma, but cross-reacted 100% with estradiol-3-sulfate. We tested possible cross-reactivity with estradiol-3-glucuronide and found that it cross-reacted less than 1% with this steroid. We therefore used the same antiserum to measure estradiol-3-sulfate concentrations. To measure the sulfoconjugated form of the steroid, we extracted the plasma with ethanol. This extraction procedure deproteinizes the plasma, but preserves the water-soluble steroid in solution (the diethyl ether, used in the estradiol extraction, excludes the sulfoconjugated form of the steroid). The recovery of the estradiol-3-sulfate was 95–100%; calculated plasma concentrations were not adjusted for recovery. The intraassay coefficient of variation in this enzyme-linked immunoassay was 8.1%.
As a marker of estrogen action in the fetal brain (Giroux, D., and C. E. Wood, unpublished observations), we measured immunoreactive Fos abundance in the cerebellum. Fetal brain tissue was homogenized in 5 vol boiling lysis buffer (1% sodium dodecyl sulfate; 1.0 mM sodium orthovanadate; and 10 mM Tris, pH 7.4), boiled, centrifuged to remove particulates, aliquoted, then stored at -80 C until assayed. The protein content of the supernatant was measured with a modified Bradford method (Bio-Rad Laboratories, Inc., Hercules, CA) using BSA as the standard (SigmaAldrich). For analysis, aliquots were thawed on ice, boiled, and electrophoresed. Electrophoresis (10–40 µg/lane, depending upon brain region) was performed using a Criterion gel and transfer apparatus (Bio-Rad Laboratories, Inc.) and precast 7.5% gels. The electrophoresed proteins were electroblotted onto nitrocellulose membranes (0.45 µm pore size; Bio-Rad Laboratories, Inc.). After transfer to the nitrocellulose membrane, the blot was probed with polyclonal anti-Fos antiserum (Oncogene Research Products, San Diego, CA; catalog no. PC05T). Molecular weight was calibrated using Rainbow7 molecular weight markers (Amersham Pharmacia Biotech, Arlington Heights, IL). All blots were probed with peroxidase-conjugated goat antirabbit IgG (Sigma-Aldrich) and visualized with a chemiluminescence reagent and film (Fuji Photo Film Co., Ltd., Tokyo, Japan; and Kodak, Rochester, NY). The density of the Fos immunostaining was quantified using Quantity One densitometer and software (Bio-Rad Laboratories, Inc.). ODs were corrected by subtraction of background.
Study 2: uptake of unconjugated and sulfoconjugated estrogen by the fetal brain
In the second study, we performed acute studies in anesthetized fetuses to measure uptake of estrone and estrone-3-sulfate. A surgical plane of anesthesia was induced and maintained in pregnant ewes using halothane (maintenance dose, 0.5–2% in oxygen). Through a midline abdominal incision, the uterus was exposed, and a hysterotomy was performed over one fetal head. The fetal head was delivered through the opening in the uterus, and the uterine wall was marsupialized to the skin of the fetal neck. Through a single midline incision in the fetal neck, the left common carotid artery was exposed and isolated.
The uptake of sulfoconjugated estrogen by the fetal brain was estimated using a modification of the method described by Stonestreet and co-workers (13). We injected a solution containing 5–10 µCi [14C]polyethylene glycol (molecular weight, 4000 g/mol; Amersham Pharmacia Biotech, catalog no. CFA508) and 10 µCi of either [3H]estrone (Amersham Pharmacia Biotech, catalog no. TRK321) or [3H]estrone-3-sulfate [Perkin-Elmer, NEN Life Science Products (Boston, MA), catalog no. NET2032] in approximately 0.5 ml normal saline. Thirty seconds after this injection, the fetus was delivered and killed with an overdose of sodium pentobarbital injected into the umbilical vein, and the fetal head was immediately removed for recovery of brain tissues. The brain of the fetus was quickly dissected for recovery of brainstem, cerebellum, hippocampus, hypothalamus, and cerebral cortex. In addition, we collected the fetal pituitary. Brain regions were isolated and divided into approximately 200- to 500-mg samples. These samples of fetal brain tissue were dissolved in a commercially available alkaline tissue solubilizer (Solvable, Packard Instruments, Downers Grove, IL). The solubilized tissue was analyzed for 3H and 14C counts using a scintillant appropriate for solutions with high ionic strength (Hionic Fluor, Packard Instruments). After correction for channel spillover, the ratio of 3H/14C was computed for each tissue sample as well as for each injectate.
Uptake of the sulfoconjugated estrone was calculated as a dimensionless index (fold enrichment) whose value can theoretically vary between 1 (no enrichment) and 
(infinite enrichment). In tissue with no uptake, the value of this index would be 1, and increasing uptake would produce values increasingly greater than 1. Fold enrichment of the steroid in brain tissue with respect to the amount remaining in plasma was calculated in the following way: Enrichment Index = (3H/14Cunknown)/(3H/14Cinjectate).
A value of the enrichment index that is significantly greater than 1 was interpreted as evidence that the injected steroid was concentrated in fetal brain tissue relative to its concentration in plasma. [14C]Polyethylene glycol serves as a marker of plasma water, as it does not cross the blood-brain barrier (13). This value does not include any assumptions about degradation or enzymatic conversion (e.g. from estrone-3-sulfate to estrone) after uptake into the tissue.
Statistical analysis
Plasma hormone concentrations were analyzed by three-way ANOVA in which the main treatment effects were day, estradiol sulfate dose, and time relative to the start of blood sampling. Fos abundance in cerebellum was analyzed using one-way ANOVA. Because of heteroscedasicity, the Fos data were subjected to logarithmic transformation before statistical analysis. Values of uptake index and enrichment index were analyzed by two-way ANOVA in which main treatment effects were brain region and sulfoconjugation state (estrone vs. estrone-3-sulfate). The criterion used for assessment of statistical significance was P < 0.05. All data were analyzed using SPSS version 11.0 (SPSS, Inc., Chicago, IL) for Windows (Microsoft Corp., Redmond, WA).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
1–5 ng/ml vs. 50 pg/ml, respectively). As shown in Fig. 1
, infusion of estradiol-3-sulfate iv increased fetal arterial plasma concentrations of the steroid in a dose-dependent manner (P < 0.05 for main effect of estradiol sulfate dose and for the interaction between day and dose effects when analyzed by three-way ANOVA). The results of the ANOVA revealed a significant cubic interaction of time with day, and a significant quadratic three-way interaction of time with day and infusion rate. The infusions produced a clearly dose-related effect (P < 0.05 for interaction between day and infusion rate) on the plasma concentrations of unconjugated estradiol (Fig. 2).
|
|
and 4. Fetal plasma cortisol (Fig. 3) was increased in response to the estradiol-3-sulfate infusions, and the effect on cortisol was also dose related (P < 0.001 for the main effect of day and for the interaction between day and infusion rate). The effect of estradiol-3-sulfate infusions on fetal ACTH (Fig. 4) was similar to that on cortisol, although the increase in immunoassayable ACTH was only apparent at the higher rate of infusion. There was a significant quadratic interaction between time and day (P < 0.05) and between time, day, and infusion rate (P < 0.01). Plasma concentrations of cortisol, ACTH, and estradiol-3-sulfate were somewhat variable within the serial sampling periods.
|
|
). The infusions stimulated an increase in the abundance of Fos in homogenates of cerebellum (P < 0.05, by one-way ANOVA after logarithmic transformation).
|
). For estrone-3-sulfate, the value of this index varied from 1.5 ± 0.4 in pituitary to 2.9 ± 0.3 in hypothalamus. For estrone, the value of this index varied from 3.6 ± 1.2 in pituitary to 29.3 ± 5.0 in hypothalamus and 30.8 ± 5.7 in hippocampus. Both steroids were taken up by brain regions more efficiently than in pituitary. In the brain regions studied, the calculated mean enrichment of estrone was 5.0–11.0 times that of estrone-3-sulfate. On the other hand, the enrichment of estrone was more variable than that of estrone-3-sulfate. When 95% confidence intervals were calculated, the enrichment of estrone was statistically significant (different from a value of 1) only in hypothalamus, hippocampus, and cortex, whereas the enrichment of estrone-3-sulfate was statistically significant in all regions. The uptake of both estrone and estrone-3-sulfate was significantly greater in some brain regions than in others (P < 0.001 for main effect of brain region, but no significant interaction between brain region and sulfoconjugation state). The results of statistical analysis confirm that overall the uptake of estrone was more efficient than that of estrone-3-sulfate (P < 0.001 for main effect of sulfoconjugation state). Nevertheless, estrone-3-sulfate was significantly and consistently concentrated in the fetal brain.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
There is much evidence demonstrating the existence of estrone sulfate in fetal plasma, but there is little published evidence of estradiol sulfate in fetal plasma. Published data demonstrate estradiol-3-sulfate in the plasma of the pregnant woman (20). In women, circulating concentrations of estradiol sulfate in plasma increase as a function of fetal gestational age (20), but there are no reports of estradiol sulfate measurements in plasma of the human fetus. To our knowledge, this is the first report of estradiol sulfate measurements in plasma of the fetal sheep, and this is the first report that exogenous estradiol sulfate is biologically active in the fetus. These experiments were designed with multiple samples (at 5-min intervals) throughout each sampling period to allow calculation of mean plasma hormone concentrations upon which ultradian variations in both cortisol (10) and ACTH (3) were superimposed. Although we expected short-term fluctuations in both cortisol and ACTH plasma concentrations, the variability in measured concentrations of estradiol-3-sulfate was not expected. The apparent rhythmicity in plasma estradiol-3-sulfate concentrations is identifiable, even in the summary data (Fig. 1
). Nevertheless, the results of the present experiments do not identify the origin of these short-term fluctuations. The variations in plasma estradiol-3-sulfate concentrations bore no obvious relationship to the minute to minute variations in plasma ACTH or cortisol.
We propose that estradiol sulfate, circulating in high concentrations, is deconjugated by STS in the fetal brain, and that the liberated estradiol is biologically active at the estrogen receptor. Several reports demonstrate that sulfoconjugated estrogens do not bind the estrogen receptor and therefore have no direct action within the cell (21, 22). Because of the abundant activity of STS within the fetal brain, it is logical to propose that the deconjugation step (which obligatorily precedes estradiol action) occurs within the brain regions important for HPA axis control (e.g. hypothalamus, brainstem, or hippocampus, or other brain regions with high STS activity). It is possible, however, that the deconjugation occurs at a peripheral site, such as the fetal liver. With the present experimental design, we cannot exclude this as a possibility. Indeed, we did measure an increase in fetal plasma estradiol (unconjugated) concentrations. Nevertheless, sulfoconjugated estrogen is directly available to the fetal brain, and it seems logical to propose that uptake by fetal brain and local deconjugation might mediate the action on the fetal HPA axis.
We measured the uptake of sulfoconjugated estrogen by the fetal brain using estrone sulfate, rather than estradiol sulfate. We used this steroid because of its commercial availability and because the physicochemical properties of estrone sulfate are not unlike those of estradiol sulfate. The mechanism by which the sulfoconjugated estrogen gains access to the fetal brain is unknown, because there is a functioning blood-brain barrier at this time in fetal development in the sheep (13). It is possible that the sulfoconjugated estrogen crosses the blood-brain barrier with the help of a transporter. A likely candidate for this action is one or more members of the family of organic acid transporters, which are known to transport sulfoconjugated steroids (23, 24). Although this is a possible mechanism for transport across the blood-brain barrier, there is no information available concerning the expression of the organic acid transporter proteins in the ovine fetus. The access of the sulfoconjugated estrogen to the fetal brain is not, however, as free as the access of unconjugated estrogen. The results of our experiments indicate that estrone crosses the blood-brain barrier 5–11 times as efficiently as estrone sulfate. On the other hand, the circulating concentration of estradiol-3-sulfate is approximately 20 times the concentration of estradiol in the fetus. Making the assumption that the kinetics of estradiol-3sulfate uptake into the fetal brain are similar to those of estrone-3-sulfate uptake, it is possible to estimate that the total flux of sulfoconjugated estradiol might equal or even exceed that of estradiol.
The fetal HPA axis and placental estrogen biosynthesis form a sort of positive feedback cycle in the ovine fetus. Increasing concentrations of cortisol in fetal plasma stimulate the activity of cytochrome P450c17, which, in turn, increases the rate of estrogen biosynthesis and decreases the rate of progesterone secretion into plasma (25). The increased estrogen concentration in fetal plasma increases fetal HPA axis activity (3, 5). The activity of estrogen on the fetal HPA axis is accounted for by an action at the fetal brain, rather than at the pituitary (5). Ontogenetic profiles of the circulating concentrations of cortisol and estradiol both reveal exponential increases before labor and delivery (26, 27). We envision this hormonal interaction as being a true positive feedback cycle, one that culminates in the birth of the fetus (3). However, we do not envision this as the sole trigger to parturition in this species. The participation of sulfoconjugated estrogens in this process is most likely the addition of a large pool of heretofore unmeasured and unconsidered estrogen precursors.
There are several variables that could influence the biological activity of estradiol sulfate as a stimulator of the fetal HPA axis. One would expect that the availability of estradiol sulfate to the fetal brain would depend upon the circulating concentrations as well as the ability of the fetal brain to deconjugate the estradiol sulfate. The high abundance of STS in the fetal brain compared with peripheral tissues suggests that the sulfoconjugated estrogens are primarily targeted to the brain and that they might subserve primarily a neuroendocrine action.
It is not clear whether developmental changes in STS expression in the fetal brain generate the ontogenetic rise in HPA axis activity at the end of gestation. A more important variable in the process is likely to be the rate of estradiol-3-sulfate synthesis and secretion. Although the fetal brain is rich in STS, the only brain region that demonstrated an ontogentic pattern of activity was the hippocampus. It is possible that estradiol sulfate action is modulated by developmental changes in STS activity in the hippocampus, although this is not clear at the present time. It is also possible that the rate of entry of sulfoconjugated estrogens into the fetal brain might be ontogenetically regulated and that alterations in the brain uptake rate have a significant effect on estrogen bioavailability in the fetus.
In summary, we conclude that estradiol-3-sulfate circulates in high concentrations in fetal plasma, and that exogenous infusions of the steroid stimulate fetal HPA axis activity. We propose that the estradiol sulfate is deconjugated at the fetal brain via the action of STS and that the liberated estradiol has a potent effect on the pathways ultimately controlling ACTH release from the fetal pituitary. We speculate that because of the large pool of estradiol precursor circulating in fetal plasma and because of the preferential expression of STS in the fetal brain, estradiol-3-sulfate could be an important influence on fetal HPA axis activity in late gestation and might be involved in the triggering of parturition.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Abbreviations: HPA, Hypothalamus-pituitary-adrenal; STS, steroid sulfatase.
Received July 26, 2002.
Accepted for publication October 23, 2002.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
and ß. Endocrinology 138:863–870
5-steroids in fetal and maternal sheep plasma in relation to spontaneous parturition at term. Endocrinology 110:1402–1407This article has been cited by other articles:
 |
|
 |
|
  J. Gersting, C. E. Schaub, M. Keller-Wood, and C. E. Wood Inhibition of Brain Prostaglandin Endoperoxide Synthase-2 Prevents the Preparturient Increase in Fetal Adrenocorticotropin Secretion in the Sheep Fetus Endocrinology, August 1, 2008; 149(8): 4128 - 4136. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. E. Wood, R. Cousins, D. Zhang, and M. Keller-Wood Ontogeny of Expression of Organic Anion Transporters 1 and 3 in Ovine Fetal and Neonatal Kidney Experimental Biology and Medicine, October 1, 2005; 230(9): 668 - 673. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. E. Wood Estrogen/Hypothalamus-Pituitary-Adrenal Axis Interactions in the Fetus: the Interplay Between Placenta and Fetal Brain Reproductive Sciences, February 1, 2005; 12(2): 67 - 76. [Abstract] [PDF]
|
|
|
|
 |
|
 C. E Wood and D. Giroux Central nervous system prostaglandin endoperoxide synthase-1 and -2 responses to oestradiol and cerebral hypoperfusion in late-gestation fetal sheep J. Physiol., June 1, 2003; 549(2): 573 - 581. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
