| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Department of Medicine (F.H., S.S., P.L., B.A.), Endocrine and Diabetes Unit, University of Würzburg, 97080 Würzburg, Germany; Department of Pharmacology (C.M.-G.), University of Heidelberg, 69120 Heidelberg, Germany; and Division of Medical Sciences (P.M.S., W.A.), Institute of Biomedical Research, University of Birmingham, Birmingham B15 2TT, United Kingdom
Address all correspondence and requests for reprints to: Dr. Wiebke Arlt, M.D., Division of Medical Sciences, Institute of Biomedical Research, Endocrinology, Room 233, University of Birmingham, Birmingham B15 2TT, United Kingdom. E-mail: w.arlt{at}bham.ac.uk.
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
), thereby serving its role as a principal precursor for human sex steroid biosynthesis. The direction of conversion of DHEA defines its biological action in a specific target tissue and is dependent on the tissue-specific expression of steroidogenic enzymes. Therein DHEA paradigmatically illustrates the concept of intracrinology that involves activation, action, and metabolism of steroids within one and the same peripheral target cell (3, 4).
|
The fetal adrenal produces vast amounts of DHEAS, which are desulfated by abundantly expressed placental STS activity, thereby enabling biosynthesis of estriol, a marker steroid for fetal development (11). However, with the exception of breast tissue and prostate (4, 9), regeneration of DHEA from DHEAS seems to be a path rarely taken in adults, with a recent paper describing surprisingly low levels of expression and activity for STS in adult human tissues (12).
We have previously shown that orally administered DHEA is readily converted to both DHEAS and downstream steroids (13, 14) and established this DHEA challenge test as a diagnostic research tool to explore differences in downstream conversion of DHEA toward androgens (15). Here we have analyzed the interconversion of DHEA and DHEAS in humans by frequent serum sampling for steroid hormones after administration of DHEA and DHEAS, respectively. In addition, we have analyzed DHEA-DHEAS interconversion in vitro, employing the human hepatic cell line HepG2.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Eight healthy young men (median body mass index, 25.7 kg/m2; range, 20.3–27.4 kg/m2; median age, 22 yr; range, 18–22 yr) were recruited via local advertising. All subjects had normal serum concentrations of DHEAS, androstenedione, and testosterone (T), normal hematology counts, and normal hepatic and renal function parameters. Exclusion criteria for subjects were hypo- or hyperthyroidism, diabetes mellitus, current or previous long-term glucocorticoid or sex steroid treatment, and current intake of drugs known to induce hepatic P450 enzymes. Before the initiation of the study, the protocol had been approved by the Ethics Committee of the University of Würzburg, and informed written consent was obtained from all study participants.
Study protocol
The study was performed on three occasions: at baseline, after oral administration of 100 mg DHEA (25-mg capsules, DHEA Natrol; EPI, Chatsworth, CA) at 0900 h, and after iv administration of 25 mg DHEAS (DHEAS Mylis, Organon, Tokyo, Japan) in 20 ml of 5% glucose from 0900–0910 h. Purity of the preparations was at least 99% as assessed by HPLC, according to the manufacturers and as previously assessed by us (14). On all three study days, frequent blood sampling was performed, starting after an overnight fast at 0830 h (–30 min), followed by sampling at 0 (0900 h), 15, 30, 45, 60, 90, 120, 180, 240, and 360 min. Standardized meals were served at 1030 h and 1300 h. The time interval between study days was at least 7 d.
Measurements
Serum steroid hormone concentrations were determined by established specific RIAs: DHEA (Diagnostic Systems Laboratories, Inc., Sinsheim, Germany) [cross-reactivity to DHEAS, 0.04%; to 4-androstene-3,17-dione (androstenedione), 0.46%; and to T, 0.03%]; DHEAS (DPC Biermann, Bad Nauheim, Germany) (cross-reactivity to DHEA, 0.08%; to androstenedione, 0.12%; to T, 0.10%; to estradiol, 0.03%; and to estriol, 0.03%); androstenedione (DPC Biermann) (cross-reactivity to DHEA, 0.02%; to dihydrotestosterone, 0.05%; and to estrone, 0.08%); T (DPC Biermann) (cross-reactivity to androstenedione, 0.5%; to dihydrotestosterone, 3.1%; and to estradiol, 0.02%); and 5
-androstane-3,17ß-diol-17-glucuronide (ADG) (Diagnostic Systems) (cross-reactivity to dihydrotestosterone-glucuronide, 1.2%; no cross-reactivity to 5-androstane-3ß,17ß-diol or 5-androstane-3,17ß-diol-3-glucuronide). Cross-reactivities to other steroids relevant to this study were less than 0.01% (for additional details regarding assay validity see Ref. 14).
In addition, DHEA was measured by a specific in-house RIA established at the Steroid Laboratory of the University of Heidelberg, using tritiated steroid (Amersham Biosciences, Freiburg, Germany) and a specific antibody, raised and characterized in the steroid laboratory, as described elsewhere (16). Before RIA, a recovery-corrected extraction of serum samples by n-hexane was performed, thereby efficiently removing DHEAS as tested by chromatographic purification. The standard curve ranged from 2.5–43.8 nmol/liter (0.2–3.5 pmol per tube), and the sensitivity was 4.1 nmol/liter (0.33 pmol per tube). The recovery of a known amount of DHEA added to plasma samples and measured repeatedly was 97.8 ± 10.1% (mean ± SD). The intra- and interassay coefficients were 6.6–11.2% and 12.4–14.2%, respectively.
Statistics
All data are reported as mean ± SEM. The area under the concentration-time curve 0–6 h (AUC0–6 h) for the measured serum steroid hormones was calculated by means of trapezoidal integration. The normal distribution of results was ascertained by using the Kolmogorov-Smirnov-Liliefors test. Comparisons of results at baseline, after oral ingestion of DHEA, and after iv administration of DHEAS were performed by t test for paired samples. Significance was defined as P < 0.05.
RT-PCR
Total RNA was extracted from 2.9 x 107 HepG2 cells using a single-step extraction method (RNeasy Mini Kit, QIAGEN, Hilden, Germany) according to the manufacturer’s protocol. RNA concentration was determined by spectrophotometry at A260, and purity was assessed by measuring the RNA/DNA ratio at A260/280. For expression analysis of STS and SULT, aliquots of 1.5 µg total human liver and colon RNA (Clontech, Heidelberg, Germany) or 1.5 µg total RNA from HepG2 cells were reverse transcribed for 1 h at 44 C with random decamer primers using the RETROscript kit (Ambion, Austin, TX). For simultaneous amplification of target genes and 18S, 200 ng cDNA was amplified in a reaction volume of 50 µl containing 200 µM of each dNTP, 0.6 µM forward and reverse primer, 100 nM QuantumRNA 18S internal standards (Ambion) [optimized ratio of 18S Primer:Competimer, 1:9 (SULT2B1a and SULT2B1b) or 2:8 (STS and SULT2A1)], 1x PCR buffer, and 2.5 U HotStart Taq polymerase (QIAGEN). PCRs were subjected to an initial HotStart Taq activation for 15 min at 94 C followed by 31 cycles (STS and SULT2A1), 38 cycles (SULT2B1a), or 37 cycles (SULT2B1b), respectively, of denaturation at 94 C for 30 sec, annealing at 64 C for 30 sec, and extension at 72 C for 30 sec. PCR products were separated by agarose gel electrophoresis and visualized under UV light (360 nm). Primer sequences are shown in Table 1.
|
To study the interconversion of DHEA and DHEAS in vitro, HepG2 cells were incubated in six-well plates (1 x 106 cells per well) with Ham’s F10 (Sigma-Aldrich, Taufkirchen, Germany) containing 10% fetal calf serum (PAN Biotech, Aidenbach, Germany). Media were changed after 24 h. After 48 h and at approximately 70% confluency, cells were incubated in 2 ml serum-free Ham’s F10 media with 0.1–2 µM DHEA (Sigma-Aldrich) for SULT activity and with 0.1–2 µM DHEAS (Sigma-Aldrich) for STS activity. Each incubation also contained 22,200 cpm 4-[14C]DHEA (47.8 mCi/mmol; NEN Life Science Products, Cologne, Germany) or 222,000 cpm [3H]DHEAS (60 mCi/mmol; NEN). Assays were done in the linear time range of the enzymatic reaction as determined by preceding time course experiments. Before steroid extraction, the media were saturated with NaCl to facilitate subsequent efficient extraction of hydrophilic DHEAS; preceding pilot experiments using this procedure had confirmed a 99.8% recovery of DHEAS. Steroids were extracted from the incubation media with ethyl acetate, concentrated by evaporation under continuous nitrogen flow, and assayed by thin-layer chromatography on silica gel TLC plates (PE SIL G/UV; Whatman, Maidstone, UK) using methylene dichloride/acetone (92.5:7.5) as the solvent system. Substrates and conversion products were identified by comparison with the comigration of reference steroids (all from Sigma-Aldrich) and quantified by phosphorimager analysis (Fuji Film FLA 3000 phosphorimager; Fuji Photo Film Europe, Dusseldorf, Germany). Conversion rates were normalized to protein levels as determined by Bradford assay (Bio-Rad, Munich, Germany). All assays were performed in triplicate, and data are presented as mean ± SD.
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Baseline studies revealed the expected diurnal rhythm of serum DHEA with higher concentrations in the morning and lower concentrations toward the afternoon, whereas serum DHEAS levels did not exhibit diurnal variations (Fig. 2, A and B). The administration of both 100 mg DHEA orally and 25 mg DHEAS iv resulted in significant increases in serum DHEAS that were sustained over a 6-h period (Fig. 2A). Despite the obvious differences in pharmacokinetic patterns caused by the different routes of application, the chosen doses of oral DHEA and iv DHEAS resulted in similar increases in serum DHEAS, with no statistical difference between the AUC0–6 h (Table 2). This had been predicted from the results of preceding bioequivalency studies (data not shown). Oral DHEA also led to an expected significant increase in circulating DHEA (Fig. 2B). However, by contrast, DHEAS iv did not result in a significant increase in serum DHEA (Fig. 2B and Table 2).
|
|
), we additionally performed a RIA with preceding extraction step allowing for efficient removal of potentially interfering DHEAS. The results of this assay revealed identical concentrations for serum DHEA during the baseline day and after iv administration of DHEAS [AUC0–6 h, 49 ± 3 vs. 48 ± 4 ng/ml·h (170 ± 11 vs. 168 ± 15 nmol/liter·h), not significant) (Fig. 2C), thereby confirming the lack of conversion of DHEAS to DHEA. Downstream conversion of DHEA and DHEAS toward sex steroids
Analysis of conversion to downstream steroids revealed a similar pattern with significant increases in androstenedione and estrone concentrations after oral DHEA but not after DHEAS iv (Fig. 3, A and B, and Table 2). Similarly, androstanediol glucuronide, a metabolite of dihydrotestosterone and a reliable marker of androgen generation within peripheral target cells, increased only after oral DHEA but not after the administration of DHEAS iv (Fig. 3C and Table 2).
|
RT-PCR expression analysis.
In vitro measurement of DHEA and DHEAS interconversion was carried out in the human hepatoma cell line HepG2. Before enzyme activity assays, we used RT-PCR analysis to ascertain that HepG2 cells, like normal human liver, express both elements of the DHEA-DHEAS shuttle system, STS and DHEA sulfotransferase (SULT2A1). Semiquantitative RT-PCR analysis revealed abundant expression of STS, the enzyme converting DHEAS to DHEA, in both normal liver and HepG2 cells, with slightly higher expression in HepG2 (Fig. 4A). Similarly, DHEA sulfotransferase (SULT2A1) was found to be expressed with equal abundance in both human liver and HepG2 cells (Fig. 4B). Expression of SULT2B1, which is thought to contribute to generation of DHEAS from DHEA, was assessed by isoform-specific primers. SULT2B1a (isoform a) was not found to be expressed in liver or HepG2 cells (Fig. 4C), whereas SULT2B1b (isoform b) was weakly expressed in both human liver and HepG2 cells, with slightly higher expression in HepG2 cells (Fig. 4D).
|
). However, by contrast, incubation of HepG2 cells with DHEAS did not result in any measurable conversion to DHEA (Fig. 5B) or other steroids. Even in time-course experiments including incubation periods of up to 24 h we were not able to detect any conversion of DHEAS to DHEA (data not shown).
|
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
In our study, DHEA was administered orally and DHEAS iv. DHEAS can only be administered parenterally because oral DHEAS administration would invariably result in quick and efficient hydrolysis to DHEA in the stomach before absorption would take place. It seems unlikely that iv infusion of DHEA instead of oral administration would have altered the observed outcome of our study. It has been recently shown in the rhesus monkey that both iv infusion of DHEA and oral administration of DHEA resulted in a similar pattern of conversion to DHEAS and androgenic steroids, with a slightly lower DHEAS-to-DHEA ratio after iv DHEA because of avoidance of hepatic first pass (17). As readily illustrated by the serum DHEAS curves in our volunteers, iv administered DHEAS led to a significantly more rapid increase in serum DHEAS and therefore was available for potential conversion even earlier than DHEA after oral administration. A minor caveat of our study is that the period of frequent serum sampling was restricted to 6 h. However, DHEA concentrations after DHEAS iv were identical to those observed on the placebo day throughout the study period, as convincingly illustrated by the results of the DHEA RIA with preceding extraction step. This outcome is clearly not what to expect when thinking along the lines of the concept of free and continuous interconversion of DHEA and DHEAS.
That concept derives from earlier studies determining metabolic clearance rates for DHEA and DHEAS, mostly using mathematical approximations based on results generated after iv infusion of unlabeled or radiolabeled DHEA in healthy volunteers (18). A few of the older studies infused radiolabeled DHEAS concomitantly with radiolabeled DHEA, however admitting that calculated results regarding the DHEA and DHEAS dynamics would appear to be a simplification of the in vivo events (19). In a recent review, Tait and Tait (20) have pointed out that all previous metabolic clearance rate studies suggesting continuous conversion of DHEAS to DHEA failed to provide direct evidence for that. Tait and Tait (20) state that to the best of their knowledge, DHEAS has not been infused in any of the studies published to achieve equilibrium conditions, which would allow reliable quantification of the conversion of DHEAS to DHEA.
The striking lack of a hepatic contribution to conversion of DHEAS to DHEA observed in our in vivo study is mirrored by our in vitro findings in hepatic HepG2 cells. RT-PCR analysis confirmed a similar expression pattern of STS and SULT2A1 in normal human liver and HepG2 cells. SULT2B1 (21) is thought to also contribute to sulfonation of DHEA. However, we did not find any expression of isoform a and only weak expression of isoform b in both human liver and HepG2 cells, indicating that hepatic DHEA sulfonation is mainly catalyzed by DHEA sulfotransferase (SULT2A1).
DHEAS can be converted to DHEA in pregnant women (22), which is largely a result of abundant placental expression of STS (9). This is further supported by data provided by the manufacturer of our iv DHEAS preparation (Organon), showing a significant and dose-dependent increase in estrone and estradiol after injection of DHEAS in pregnant women. STS clearly is an important contributor to tissue-specific activation of DHEAS and estrone sulfate, respectively, to DHEA and estrone in breast tissue and prostate (4, 9). However, based on our findings, there is obviously no major contribution of hepatic STS to this. Conversion of DHEA to androgenic steroids within peripheral target cells may not necessarily affect circulating androgen levels, but reliably results in a significant increase in the androgen metabolite ADG (14, 23). Although we have observed ample generation of ADG after DHEA administration, none such was observed after DHEAS. However, we cannot exclude that STS may convert DHEAS to DHEA in some peripheral tissues, e.g. within blood cells. In the light of our findings, it would be very tempting to have a more detailed look at DHEA, DHEAS, and androgenic steroids in individuals with STS deficiency, in whom no abnormalities in steroid synthesis and metabolism have been reported yet.
Our findings raise the question why DHEAS circulates in micromolar amounts in our body, if it does not serve as a circulating regeneration pool for DHEA. It has been suggested that DHEA and DHEAS may have differential effects on neuronal outgrowth in the central nervous system (24). However, biological action of DHEAS in humans, independent of its conversion to DHEA, remains to be demonstrated.
Taken together, our findings challenge the current concept of free and continuous interconversion of DHEA and DHEAS and suggest that circulating DHEAS may not serve as a pool for DHEA regeneration. DHEA sulfotransferase rather than STS activity appears to represent the crucial rate-limiting step regulating DHEA bioavailability. Decreased DHEA sulfotransferase activity would increase DHEA, thereby driving downstream conversion and action of DHEA, whereas increasing DHEA sulfotransferase activity would result in the reverse. Therefore, it is highly likely that serum DHEAS does not appropriately reflect corresponding levels of desulfated, biologically active DHEA. Serum DHEA and DHEAS may be concordant in the physiological situation, but will be discordant in pathological conditions, in particular if DHEA sulfotransferase activity is pathologically altered.
It has been recently shown in a murine model that lipopolysaccharide-induced cytokine release significantly down-regulates DHEA sulfotransferase expression (25). In humans, the currently accepted concept is that in severe states of stress, an intraadrenal shift from DHEA toward glucocorticoid synthesis occurs. However, this is based on the finding of decreased DHEAS levels in severe stress such as sepsis. Our results and the evidence for down-regulation of DHEA sulfotransferase during the acute-phase response (25) would lead us to predict that although DHEAS may be low, biologically active DHEA will be increased in severe stress. This would have important implications for our understanding of the endocrine response to stress.
Similarly, adrenal hyperandrogenism is usually excluded by measurement of DHEAS levels. Typically, a woman with normal DHEAS and increased androstenedione levels is considered to suffer from hyperandrogenemia of primarily ovarian origin. Therefore research into the origin of polycystic ovary syndrome-associated hyperandrogenism generally focuses on mechanisms underlying ovarian androgen hypersecretion. However, it may well be possible that although serum DHEAS is normal, biologically active DHEA is pathologically increased, hence the finding of increased androstenedione as a consequence of efficient downstream conversion of DHEA.
The vast majority of published studies analyzing DHEA secretion measured only DHEAS but not DHEA. Therefore, our findings can be predicted to have major implications for the understanding of the physiology (adrenarche and aging) and the pathophysiology (stress, sepsis, hyperandrogenism, and polycystic ovary syndrome) of DHEA and DHEAS secretion. We currently do not suggest that measurement of DHEA should exclusively replace measurement of DHEAS. However, it may be very useful to measure both steroids and to assess DHEA/DHEAS ratios. For the latter, we certainly need to take into account the diurnal secretion pattern of DHEA but not DHEAS. Another obstacle for widespread measurement of desulfated DHEA is that this currently requires a more demanding RIA approach, whereas reliable automated ELISA for measurement of DHEAS are readily available. However, we have no doubt that future studies of the DHEA-DHEAS shuttle and its regulation in health and disease will help to gain important and highly clinically relevant insights.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
First Published Online March 8, 2005
Abbreviations: ADG, 5-Androstane-3,17ß-diol-17-glucuronide; AUC0–6 h, area under the concentration-time curve 0–6 h; DHEA, dehydroepiandrosterone; DHEAS, DHEA sulfate; STS, steroid sulfatase; SULT, sulfotransferase; T, testosterone.
Received December 6, 2004.
Accepted February 25, 2005.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
-reductase activity in women with polycystic ovary syndrome. J Clin Endocrinol Metab 88:2760–2766This article has been cited by other articles:
 |
|
 |
|
  M. O. Goodarzi, N. Xu, and R. Azziz Association of CYP3A7*1C and Serum Dehydroepiandrosterone Sulfate Levels in Women with Polycystic Ovary Syndrome J. Clin. Endocrinol. Metab., July 1, 2008; 93(7): 2909 - 2912. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
W. Arlt, F. Hammer, P. Sanning, S. K. Butcher, J. M. Lord, B. Allolio, D. Annane, and P. M. Stewart Dissociation of Serum Dehydroepiandrosterone and Dehydroepiandrosterone Sulfate in Septic Shock J. Clin. Endocrinol. Metab., July 1, 2006; 91(7): 2548 - 2554. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. S. Willenberg, M. Bahlo, W. A. Scherbaum, F. de Zegher, D. B. Dunger, L. Ibanez, and R. L. Rosenfield Hirsutism. N. Engl. J. Med., April 6, 2006; 354(14): 1533 - 1535. [Full Text] [PDF]
|
|
|
|
 |
|
 W. Arlt Androgen therapy in women Eur. J. Endocrinol., January 1, 2006; 154(1): 1 - 11. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. K. Siiteri The Continuing Saga of Dehydroepiandrosterone (DHEA) J. Clin. Endocrinol. Metab., June 1, 2005; 90(6): 3795 - 3796. [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 |
), after oral administration of 100 mg DHEA (
), and after iv infusion of 25 mg DHEAS (
). C, Serum DHEA concentrations at baseline and after iv DHEAS as determined by RIA with preceding extraction step (see Subjects and Methods). To convert the values for DHEA to ng/ml, divide nM by 3.467; to convert the values for DHEAS to µg/dl, divide µM by 0.02714.
