| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Studies |
Department of Obstetrics and Gynecology, Division of Reproductive Biology and Endocrinology, University of Alabama at Birmingham, Birmingham, Alabama 35233
Address all correspondence and requests for reprints to: C. R. Parker, Jr., Ph.D., Department of Obstetrics and Gynecology, Division of Reproductive Biology and Endocrinology, 618 20th Street South, University of Alabama at Birmingham, Birmingham, Alabama 35233-7333.
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
417,20-lyase
(i.e. the A4/17-OHP ratio) activities also
were unaffected by E2 therapy. The responsiveness of F to
ACTH was significantly greater than that of DHEA, A4, or
17-OHP regardless of the circulating E2 levels.
Furthermore, F and A4 were significantly more sensitive to
ACTH stimulation than were 17-OHP and DHEA, and this was not altered by
E2 administration. We conclude that transdermal
E2 replacement to postmenopausal women does not
significantly alter AA sensitivity or responsiveness to ACTH. Hence, it
is unlikely that the hypoestrogenism of menopause contributes to the
decline in AAs noted with age. Furthermore, menopausal estrogen
replacement, at least in physiological amounts administered
transdermally, cannot be expected to reverse the suppressed production
of these androgens. |
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
5-adrenal steroid pathway to ACTH stimulation was
reported to be selectively attenuated compared to that of the
4-adrenal steroid pathway in the same patients and to
overall responses seen in younger women (4). Interestingly, in these
studies the 4 adrenal steroid response to ACTH appeared
to be greater in older subjects than in younger populations. It has
been suggested that the underlying mechanism(s) behind the declining
5 response with age may be a declining adrenal zona
reticularis cell mass, altered enzymatic activity (e.g. of
P450C17), or a decrease in the response of AAs to unchanged levels of
endogenous ACTH at the receptor level (5). Although DS levels begin to decline before menopause (6), several investigators have suggested that menopause-associated estrogen deficiency may further suppress AA secretion (7, 8). Urinary androgen metabolites and dehydroepiandrosterone (DHEA) are increased after oral estrogen administration to girls with gonadal dysgenesis (9, 10), suggesting diminished 3ß-hydroxysteroid dehydrogenase (3ßHSD) activity. Furthermore, in vitro studies have clearly demonstrated inhibitory effects of 17ß-estradiol (E2) on 3ßHSD activity in human adrenal cells (11, 12). Finally, the presence of estrogen receptors in the adrenal cortex of several animal species, including primates, suggests that E2 may be a physiologically important regulator of adrenal steroidogenesis (13).
Conflicting results have been reported concerning the effects of estrogen replacement therapy on AA production in either ovariectomized premenopausal or postmenopausal women. Although some investigators reported that oral E2 therapy or an acute iv infusion did not affect adrenal function (14, 15), others found that oral E2 therapy increases AA production through inhibition of 3ßHSD activity (7). In contrast, Lobo et al. found that the oral administration of 2.5 mg/day ethinyl E2 caused an increase in both 3ßHSD and 17,20-lyase activities, as measured by the steroid response to acute ACTH maximal stimulation, to levels similar to those found in premenopausal ovulatory women (8). These investigators also reported that oral conjugated estrogens, given in doses as low as 0.625 mg daily for 4 weeks, significantly increased serum DHEA levels above baseline in postmenopausal women.
Prior studies on the effects of estrogen on AA production have used supraphysiological doses and/or oral administration that induce a first pass hepatic effect and may not accurately reflect the adrenocortical response to physiologic endogenous E2 levels. Furthermore, these studies used solely the acute ACTH-(1–24) stimulation test to estimate adrenal steroidogenesis, which does not assess adrenocortical sensitivity or subtleties in responsiveness to ACTH. Thus, in the present study we have hypothesized that the hypoestrogenism of menopause decreases the adrenocortical sensitivity and/or responsiveness of AAs [specifically androstenedione (A4) and DHEA] to ACTH, which can be restored by replacement of E2, given at a physiological dose. To test this hypothesis, we determined the adrenocortical responses to a graded continuous ACTH infusion and to a 0.25 mg ACTH-(1–24) bolus before and after 3 months of 0.05 mg daily transdermal E2 replacement in 14 postmenopausal women.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Fourteen healthy postmenopausal volunteers, 54–67 yr of age (mean = 59.9 yr), with a mean body mass index and weight of 26 ± 3 kg/m2 and 68 ± 12 kg, respectively, were recruited. All subjects had undergone natural menopause at least 9 months before the study, had at least one ovary, and had not received hormone therapy in the past 6 months. Eleven subjects were white and three were black. These studies were approved by the institutional review board of the University of Alabama-Birmingham, and all subjects gave written informed consent.
Study protocol
The subjects underwent the following studies before and after 3 months of continuous transdermal E2 replacement (0.05 mg/day; Estraderm, Ciba Co., Geneva, Switzerland). A fasting morning blood sample was obtained to characterize the basal hormonal milieu of the subjects. Next, the volunteers ingested 1 mg dexamethasone (DEX) at 2300 h the night before the ACTH infusion and then again at 0700 h with breakfast the next morning. One to 2 h later, an indwelling catheter was inserted into an antecubital vein of each arm, and a 30-min control infusion of normal saline (50 mL/h) was initiated. After the control infusion, a blood sample was obtained for determination of DEX-suppressed steroid hormone levels. Subsequently, a continuous ACTH-(1–24) (Cortrosyn, Organon Co., West Orange, NJ) infusion was initiated at a rate of 20 ng/1.5 m2/h and was doubled each 60 min until a maximum rate of 1280 ng/1.5 m2/h was obtained (i.e. 20, 40, 80, 160, 320, 640, and 1280 ng ACTH/1.5 m2/h). At the end of each 60-min infusion period, a 15-mL blood sample was withdrawn. After obtaining the last infusion blood sample, a bolus of 0.25 mg ACTH was administered, and a final blood sample was withdrawn 60 min later to determine maximal adrenal steroid output. During the infusion, the volunteers were recumbent and were allowed access to crackers and juice as desired to minimize hypoglycemia and hunger during the 7.5-h infusion period. The blood samples were centrifuged at 3000 rpm for 15 min, and the serum was stored at -70 C for later RIA analysis.
Hormonal measures
All steroids were quantified by RIA, as indicated below. Interassay variation was limited by analyzing in the same RIA samples obtained before and after E2 treatment in the same patient. DHEA was quantified by direct RIA using a highly specific double antibody method from Diagnostic Systems Laboratories (Webster, TX). A4 was quantified by solid phase RIA method from Diagnostic Systems Laboratories (Webster, TX). 17-Hydroxyprogesterone (17-OHP) was quantified using a solid phase RIA method from Diagnostic Products Corp. (Los Angeles, CA). F was quantified by use of a direct in-house assay on methanol-treated diluted serum (1:10). The assay employs a highly specific antiserum (gift from Dr. C. E. Gomez-Sanchez, Columbia, MO), [3H]F as assay tracer, and dextran-coated charcoal for separation of bound and free hormone. The assay sensitivity is 7.8 pg/tube, and the intraassay coefficients of variation for low and high values are 7.1% and 4.9%. Circulating E2 levels were assayed by a solid phase RIA method from Pantex (Pantex, Santa Monica, CA). Serum FSH levels were determined with a solid phase RIA kit (Nichols Institute Diagnostics, San Juan Capistrano, CA).
The enzyme activities of 3ßHSD and 417,20-lyase were
calculated as the product/precursor ratios of the estimated adrenal
contributions to circulating steroid levels. The adrenal contribution
for each steroid was taken as the net difference in hormone
concentration between either the morning value (basal) or that after a
bolus of 0.25 mg ACTH (max) and the concentration after overnight DEX
suppression.
Statistical analysis
Comparisons of hormone concentrations under basal conditions, after overnight DEX suppression, and 1 h after the bolus dose of 0.25 mg ACTH and of estimated enzyme activities in women before and after E2 therapy were accomplished by means of t test or nonparametric test, as appropriate.
A major aim of this study was to determine measures of the sensitivity and responsiveness of several adrenal hormones to ACTH in postmenopausal women both before and after 3 months of E2 therapy. To accomplish this, we sought to determine the minimal dose of exogenous ACTH required to activate secretion of each hormone (i.e. sensitivity), and to determine the rate of rise of each hormone (i.e. responsiveness) once production was activated. We modeled our experimental and statistical approach after the studies of Komindr et al. (16), with certain modifications that were based on their results and those of our preliminary study of the response of the adrenal to ACTH in men (17). To be as precise as experimentally feasible, we were required to use doses of ACTH in the low concentration range that might be minimally effective to stimulate steroidogenesis and to use a sufficiently large number of ACTH doses in the range where activation was likely to occur.
As an initial step in evaluating the responses of the subjects to the varying doses of infused ACTH, we subjected the data for each hormone to repeated measures analysis, which revealed significant ACTH dose effects on serum levels of each hormone. A dose-response curve, comprised of the mean hormone levels at each ACTH dose, was plotted for each steroid hormone to estimate the dose(s), in addition to the 0 ng ACTH/1.5 m2/h dose (i.e. basal or saline only dose), that might not have caused significant increases in hormone concentrations and thus would be treated as the baseline. Duncan’s multiple range test was performed, which identified the ACTH doses that failed to raise a given hormone above the value at the 0 ng ACTH/1.5 m2/h dose. The average hormone response for such a dose was established as the baseline for further evaluations. The baseline included responses of F and A4 to doses of ACTH ranging from 0–40 ng/1.5 m2/h and responses of DHEA and 17-OHP to doses of ACTH ranging from 0–80 ng/1.5 m2/h.
The determination of an ACTH threshold dose for each hormone
necessitated the establishment of each hormone’s threshold response,
defined operationally here as the point at which an effect was seen and
defined mathematically as a value significantly higher than baseline,
as determined by a one-tailed t test. We calculated the
experimental error (estimate of the variance, S2) by means
of the randomized complete block design for each hormone in women
before and after estrogen replacement therapy. The SEM at
each dose was equal to 
. To test for
the presence of a difference between the hypothetical average threshold
response and the average baseline described above, the equivalent of
the t test was described by the following formula:
tdf1-
/2 = [(average threshold
response - average baseline)/S
],
where df is the degrees of freedom associated with the estimate of the
variance from the randomized complete block analysis of variance, =
0.05, n is the number of observations in the average threshold
response, m is the number of observations in the average baseline, and
tdf1-/2 is the
(1 - /2)% point for the t distribution. As
baseline was to be subtracted from all responses, baseline was set as 0
hormone concentration. Thus, the least significant difference for the
average threshold response from the baseline was defined as:
threshold =
(tdf1-/2)[S
], which, in turn, gave an estimate of the threshold
response.
Upon establishing an estimate of the average threshold response, its paired coordinate, the estimate for the threshold dose of ACTH required to raise a given hormone’s concentration the equivalent of the threshold rise was determined based on the following requirement. If the incremental rise in a given hormone’s concentration could be described by a linear regression model, then the average threshold response could be taken as a point on this line without loss of generality. Consequently, the linear model would provide for determination of the paired coordinate that would define the threshold ACTH dose and also would provide the 95% confidence limits for this point estimate based on the variability of each steroid analyte.
We obtained regression lines for women before and after estrogen replacement therapy separately for each hormone. The doses chosen for inclusion in regression equations were those that were determined to not be part of the baseline, as described above. When these data were plotted, ACTH doses were truncated if they were associated with a plateau effect at the upper end of the dose-response curve. The ACTH doses that satisfied the above requirements were 160-1280 ng/1.5 m2/h for DHEA and 17-OHP and 80–640 ng/1.5 m2/h for F and A4. The slope of the dose-response curve was considered to be representative of the responsiveness of a given hormone to ACTH. The form of the linear regression model was: response = slope x ACTH dose. Upon substitution of the average threshold response for each hormone (as determined above) for the response term in the equation, we determined the ACTH dose term for each hormone evaluated. This ACTH dose amount, when added to the highest dose of ACTH infused that was part of the baseline, provided the estimate of the threshold dose of ACTH required to elicit a significant increase above baseline for each hormone. The threshold dose of ACTH is considered to be representative of the sensitivity of each hormone’s secretory apparatus to ACTH. The 95% confidence limits for the threshold dose was calculated using a formula for computing the confidence limits of inverse predictions (18). The above statistical evaluations were performed by use of SAS (Statistical Analysis Systems, Cary, NC).
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Figure 1
demonstrates the basal,
suppressed, and maximally stimulated values [after the 0.25-mg
ACTH-(1–24) bolus] for DHEA, 17-OHP, A4, and F before and
after E2 therapy. Transdermal E2 therapy had no
significant effect on these measures. The degree of steroid suppression
after overnight DEX administration ranged from 68–90%, with F being
the most responsive to suppression and 17-OHP the least inhibited (Fig. 2). There were no significant differences
noted in the degree of suppression of any of the four steroids in the
subjects as a result of E2 therapy.
|
|
. For example, the ACTH infusion rate
required to increase F secretion before E2 therapy was only
about one third of that required to raise the concentration of DHEA
above its baseline level. The ACTH infusion rates required to produce a
significant elevation in the measured steroids over baseline fall into
two categories (Table 1). The sensitivities of F and A4 to
ACTH were similar, but differed significantly (P <
0.05) from those of 17-OHP and DHEA. Likewise, the sensitivities of
17-OHP and DHEA to ACTH were similar. Nonetheless, neither the
sensitivity of each steroid to ACTH stimulation nor the relative rank
among steroids was altered by E2 therapy.
|
. Evaluation of responsiveness,
i.e. the slope of the dose-response curves as described in
Materials and Methods, revealed that F had by far the
highest responsiveness (>10-fold greater) compared to the other
steroids (data not shown). Next we evaluated the responsiveness of DHEA
and A4, which were similar and approximately twice that of
17-OHP. Comparisons of the slopes before and after E2
therapy revealed no significant change in hormone response to ACTH.
|
417,20-lyase (i.e. A4/17-OHP) and
3ßHSD (i.e. A4/DHEA) were calculated.
Comparison of these estimates of adrenal enzymatic activity across
treatment groups demonstrated no change with E2 therapy
(Table 2).
|
Power analysis at = 0.05 suggests that our data are sufficient
to detect 31%, 31%, 34%, and 28% differences in the sensitivities
of F, DHEA, A4, and 17-OHP to ACTH, respectively, in
response to E2 treatment. Furthermore, at = 0.05, our
data would be able to detect 14%, 54%, 45%, and 41% differences in
maximally stimulated [i.e. to 0.25 ACTH-(1–24)] F,
DHEA, A4, and 17-OHP values, respectively, in women before
as compared with those after E2 treatment.
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Prior investigations of the impact of E2 therapy in
postmenopausal and premenopausal oophorectomized women have only
determined the adrenal response to acute maximal ACTH stimulation,
which cannot detect subtle changes in adrenocortical sensitivity.
However, although not altering circulating levels of AAs,
E2 may alter the sensitivity of the adrenal cortex to ACTH
stimulation, either selectively (e.g. for AAs) or as a
whole. Nonetheless, using an 8-h graded dose infusion of ACTH-(1–24)
to incrementally stimulate the adrenal cortex, we were unable the
detect a difference in adrenocortical sensitivity after E2
therapy. Finally, in the present study we estimated adrenal 3ßHSD and
417,20-lyase activity by calculating the observed
product/precursor ratios; these activities were not altered by
E2 therapy.
Although our results are in agreement with some reports (14, 15), our data contrasts with those of others. For example, Lobo et al. found that the administration of either 0.625 mg/day oral conjugated estrogens or 2.5 mg/day ethinyl E2 enhanced adrenocortical 3ßHSD and 17,20-lyase activities in a dose-dependent fashion (8). There may be various reasons for the discrepancy between our results and those of other investigators. Orally administered estrogens, in contrast to transdermal E2, result in a significant first pass hepatic effect (19). Hence, in addition to changes in both coagulation and lipoproteins, oral estrogens yield significant increases in sex hormone and F-binding globulin levels and in renin substrate, which may be sufficient to impact adrenocortical behavior. It may be argued that longer treatment may be required for estrogens to alter adrenal function. However, Rose et al. found no differences in serum DS or A4 levels between age-matched controls and 61 postmenopausal women exposed to various doses and preparations of oral E2 for periods varying from 6 months to 22 yr (15). Administration of supraphysiological E2 doses may be required for an effect to be observed. Indeed, Lobo et al. reported significant increases in serum A4, DS, DHEA, and testosterone levels in five postmenopausal women after 4 weeks of treatment with 2.5 mg conjugated estrogens (8). Finally, it is possible that our study was of insufficient power to detect minimal differences in adrenal function after E2 therapy. Nonetheless, power analysis using our data suggests that our study design was sufficient to detect at least a 30% difference in sensitivity to ACTH stimulation and a 50% difference in the maximum response. It is unclear whether changes of less magnitude than these are also clinically relevant.
Although F and A4 were more sensitive to ACTH stimulation than either 17-OHP or DHEA, this differential sensitivity was not altered by E2 replacement. Although we are currently unaware of other data pertaining to the sensitivity and responsiveness of adrenal steroids to ACTH in postmenopausal women, it is interesting to compare the relative and absolute adrenocortical sensitivities and responsiveness of DHEA, F, and A4 to ACTH as reported by Komindr in younger premenopausal women (16) to those in our older population. The mean sensitivity of F to ACTH stimulation in our older women was nearly identical to that reported for the nonobese young women in Komindr’s study. Alternatively, in our postmenopausal women, A4 appeared to be much more sensitive than DHEA to ACTH stimulation, whereas the reverse was observed in young women by Komindr et al. These results suggest an enhanced 3ßHSD and a diminished 17,20-lyase activity in older women, consistent with some findings of others in untreated postmenopausal women (4, 20). Nonetheless, a strict comparison of adrenocortical sensitivity and responsiveness to ACTH between young and old individuals remains to be performed.
In conclusion, the results of our study demonstrate that E2 at physiological doses does not alter AA sensitivity and/or responsiveness to ACTH in postmenopausal women. Thus, it is reasonable to suggest that factors other than estrogen deficiency are responsible for the natural decline in AAs seen with aging. These may include a selective alteration in P450C17 activity and/or a decrease in the mass of the reticularis, possibly regulated by ACTH-independent factors. Studies to corroborate these latter hypotheses are currently under way.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Fellow in Reproductive Endocrinology. Current address: Department
of Obstetrics and Gynecology, Medical College of Georgia, 1120
Fifteenth Street, Augusta, Georgia 30912-3360.
Received July 9, 1997.
Revised October 10, 1997.
Accepted October 30, 1997.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
-hydroxyprogesterone and
dehydroepiandrosterone in the human adrenal gland in vitro. J Endocrinol. 60:27–35.[Medline]
-adrenocorticotropin-(1–24)
in normal and obese, nonhirsute, eumenorrheic women. J Clin
Endocrinol Metab. 63:860–864.[Abstract]
This article has been cited by other articles:
 |
|
 |
|
  M. O. Goodarzi, H. J. Antoine, and R. Azziz Genes for Enzymes Regulating Dehydroepiandrosterone Sulfonation Are Associated with Levels of Dehydroepiandrosterone Sulfate in Polycystic Ovary Syndrome J. Clin. Endocrinol. Metab., July 1, 2007; 92(7): 2659 - 2664. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Reisch, M. Slawik, O. Zwermann, F. Beuschlein, and M. Reincke Genetic influence of an ACTH receptor promoter polymorphism on adrenal androgen secretion Eur. J. Endocrinol., November 1, 2005; 153(5): 711 - 715. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Slawik, N. Reisch, O. Zwermann, C. Maser-Gluth, M. Stahl, A. Klink, M. Reincke, and F. Beuschlein Characterization of an Adrenocorticotropin (ACTH) Receptor Promoter Polymorphism Leading to Decreased Adrenal Responsiveness to ACTH J. Clin. Endocrinol. Metab., July 1, 2004; 89(7): 3131 - 3137. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. E. Wood, J. M. Cline, M. S. Anthony, T. C. Register, and J. R. Kaplan Adrenocortical Effects of Oral Estrogens and Soy Isoflavones in Female Monkeys J. Clin. Endocrinol. Metab., May 1, 2004; 89(5): 2319 - 2325. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Piltonen, R. Koivunen, L. Morin-Papunen, A. Ruokonen, I.T. Huhtaniemi, and J.S. Tapanainen Ovarian and adrenal steroid production: regulatory role of LH/HCG Hum. Reprod., March 1, 2002; 17(3): 620 - 624. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. A. Laughlin and E. Barrett-Connor Sexual Dimorphism in the Influence of Advanced Aging on Adrenal Hormone Levels: The Rancho Bernardo Study J. Clin. Endocrinol. Metab., October 1, 2000; 85(10): 3561 - 3568. [Abstract] [Full Text]
|
|
|
|
|
|
C. R. Parker Jr., S. M. Slayden, R. Azziz, S. L. Crabbe, G. A. Hines, L. R. Boots, and S. Bae Effects of Aging on Adrenal Function in the Human: Responsiveness and Sensitivity of Adrenal Androgens and Cortisol to Adrenocorticotropin in Premenopausal and Postmenopausal Women J. Clin. Endocrinol. Metab., January 1, 2000; 85(1): 48 - 54. [Abstract] [Full Text]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
