In vivo activity of 11ß-hydroxysteroid dehydrogenase type 1 in man: effects of prednisolone and chenodesoxycholic acid.

The 11ß-hydroxysteroid dehydrogenases (11ß-HSDs) play a pivotal role in glucocorticoid (GC) action. 11ß-HSD1 is a predominant reductase, activating GCs from inert metabolites, whereas 11ß-HSD2 is a potent dehydrogenase inactivating GCs. Knowing the metabolic effects of GCs, a selective inhibition of 11ß-HSD1 represents a potential target for therapy of impaired glucose tolerance, insulin insensitivity and central obesity. In vitro, 11ß-HSD1 is selectively inhibited by chenodesoxycholic acid (CDCA) and upregulated under GC exposure. Therefore, we aimed to investigate the effects of CDCA and prednisolone on hepatic 11ß-HSD1 activity in vivo by measuring 11-reduction of orally given cortisone (E) acetate to cortisol (F). CDCA or placebo was given to 5 male healthy volunteers within a randomised cross-over trial before and after oral administration of 12.5 mg E acetate at 8:00 h. For measurement of in vivo effects of GCs on 11ß-HSD1 activity, hepatic reduction of 25 mg E acetate before and after treatment with prednisolone (30 mg for 6 days) was determined in 7 healthy males. Serum GC levels were determined using a fully automated liquid chromatographic system. CDCA had no effect on the activity of 11ß-HSD1 in vivo. Prednisolone therapy leads to a marked rise in serum F concentrations and an elevated F/E serum ratio. This proves GC-induced activation of hepatic 11ß-HSD1, which could not be extinguished by a parallel increase of IGF-1 under prednisolone. CDCA does not affect in vivo activity of 11ß-HSD1 when given in therapeutic dosages. During GC treatment, increased hepatic activation of E to F may aggravate metabolic side effects of GCs such as seen in the metabolic syndrome.

Administration of glycyrrhetinic acid: significant correlation between serum levels and the cortisol/cortisone-ratio in serum and urine.

In vitro, cortisol and aldosterone have a similar affinity to the mineralocorticoid receptor. The 11beta-hydroxysteroid dehydrogenase catalyzes the interconversion of cortisol to its inactive 11-oxo-metabolite cortisone. This interconversion is responsible for the in vivo specificity of the mineralocorticoid receptor. A defect of this enzyme leads to a pseudohyperaldosteronism with hypertension and hypokalemia, the so-called apparent mineralocorticoid excess syndrome. Glycyrrhetinic acid, a compound of licorice, also leads to pseudohyperaldosteronism by an inhibition of the 11beta-hydroxysteroid dehydrogenase. We studied the pharmacokinetics of glycyrrhetinic acid and its effect on the 11beta-hydroxysteroid dehydrogenase. Ten healthy students, aged 24 to 38 years, were included in the study. On the first day 500 mg glycyrrhetinic acid were given orally at 08.00 h. Blood and urine samples were taken prior to and 2, 4, 7, 10 and 24 hours after ingestion of glycyrrhetinic acid. We measured the serum level of cortisol, cortisone and glycyrrhetinic acid and the urinary excretion rates of cortisol, cortisone and their 20-dihydrometabolites. For determination of glycyrrhetinic acid and steroid levels we used a fully automated liquid chromatographic analyzer which allows the highly specific and simultaneous determination of steroid profiles even in the matrix of urine. Ratios of the 11-hydroxy- and 11-oxo-metabolites were calculated and correlated to the serum level of glycyrrhetinic acid. We found a significant correlation of the steroid-ratios to the serum levels of glycyrrhetinic acid. Coefficients of correlation were 0.9873, 0.7812, 0.7396 and 0.5844 between the serum level of glycyrrhetinic acid and the cortisol/cortisone-ratio in serum (p < 0.0001), the cortisol/cortisone-ratio in urine (p = 0.0279), the 20alpha-dihydrocortisol/20alpha-dihydrocortisone-ratio in urine (p = 0.0119) and the 20beta-dihydrocortisol/20beta-dihydrocortisone-ratio in urine (p = 0.0419), respectively. We conclude that the ratios of cortisol to cortisone and of the 20-dihydrometabolites of cortisol to the 20-dihydrometabolites of cortisone provide a simple noninvasive tool for monitoring the in-vivo activity of the 11beta-hydroxysteroid dehydrogenase.

Metabolism of synthetic corticosteroids by 11 beta-hydroxysteroid-dehydrogenases in man.

The presence of an 11 beta-hydroxyl group is essential for the anti-inflammatory and immunosuppressive effects of glucocorticoids. Interconversion of the 11 beta-hydroxyl into the corresponding 11 beta-keto group and vice versa by 11 beta-hydroxysteroid-dehydrogenase (11 beta-HSD) may thus play a pivotal role in the efficacy of these steroids. Therefore, we have compared the metabolism of the endogenous glucocorticoid cortisol (F) with that of synthetic 9 alpha-fluorinated steroids by 11 beta-HSDs in humans in vivo and in vitro. Whereas 30% of the free steroids in urine after oral administration of 5 mg of F is F itself and 70% the inactive keto-product cortisone (E), the urinary excretion of an identical dose of oral 9 alpha-fluorocortisol (9 alpha FF) is 90% 9 alpha FF and 10% inactive 9 alpha-fluorocortisone (9 alpha FE). Kidney slices similarly convert F much faster to E than 9 alpha FF to 9 alpha FE; conversely, renal 11 beta-reduction of 9 alpha FE to 9 alpha FF is much more effective than that of E to F. Kinetic analyses in human kidney cortex microsomes prove that the preference of fluorinated steroids for reduction in human kidney slices is catalyzed by 11 beta-HSD type II: the NADH-dependent conversion of 11-dehydro-dexamethasone (DH-D), another fluorinated steroid, to dexamethasone (D) is very effective (high affinity, high Vmax), whereas reduction of E to F is very slow. In human liver microsomes (11 beta-HSD type I), nonfluorinated (E) and fluorinated 11-dehydrosteroids (DH-D) are both reduced to their corresponding active 11-hydroxyderivatives but with a Michaelis-Menten constant about 20-fold higher than for kidney microsomes (11 beta-HSD-II). Our results suggest that the decreased renal 11 beta-oxidation of 9 alpha-fluorinated steroids may offer pharmacokinetic advantages for renal immunosuppression. Furthermore, administration of fluorinated 11-dehydrosteroids is a new and exciting idea in glucocorticoid therapy in that small amounts of oral DH-D may pass the liver largely unmetabolized (11 beta-HSD-I has low affinity for such steroids) and may then be activated to D by high-affinity 11 beta-HSD-II, thus allowing selective immunosuppression in organs expressing 11 beta-HSD-II (kidney and colon).

Determination of 18 beta-glycyrrhetinic acid in human serum using the fully automated ALCA-system.

We report a method for the determination of 18 beta-glycyrrhetinic acid (glycyrrhetinic acid) in human serum using the ALCA-system. The technology of the ALCA-system is based on the principles of adsorptive and desorptive processes between liquid and solid phases. The assay is run fully automated and selective. Procedural losses throughout the analysis are negligible, thereby allowing for external calibration. The calibration curve is linear up to 10 mg/l and concentrations as low as 10 micrograms/l are detectable. CV is 2.5% for within- and 7.5% for between-assay precision at a level of 50 micrograms/l and 1.2% for within- and 8.5% for between-assay precision at a level of 500 micrograms/l. Specific and expensive reagents are not necessary and time-consuming manual operations are not involved. This assay can be selected from a wide spectrum of methods at any time. Thus, the present method is well-suited for drug monitoring purposes in the routine laboratory. In a pharmacokinetic study we measured serum levels of glycyrrhetinic acid in ten healthy young volunteers after ingestion of 500 mg glycyrrhetinic acid. Maximum levels of glycyrrhetinic acid were 6.3 mg/l 2 to 4 hours after ingestion. Twenty-four (24) hours after ingestion seven probands still had glycyrrhetinic acid levels above the detection limit with a mean level of 0.33 mg/l.

Human kidney 11 beta-hydroxysteroid dehydrogenase: regulation by adrenocorticotropin?

In ectopic adrenocorticotropin (ACTH) syndrome (EAS) with higher ACTH levels than in pituitary Cushing's syndrome and during ACTH infusion, the ratio of cortisol to cortisone in plasma and urine is increased, suggesting inhibition of renal 11 beta-hydroxysteroid dehydrogenase (11 beta-HSD) by ACTH or by ACTH-dependent steroids. Measuring the conversion of cortisol to cortisone by human kidney slices under different conditions, we tested the possibility of 11 beta-HSD regulation by ACTH and corticosteroids. Slices prepared from unaffected parts of kidneys removed because of renal cell carcinoma were incubated with unlabeled or labeled cortisol, and cortisol and cortisone were quantitated after HPLC separation by UV or radioactive detection. The 11 beta HSD activity was not influenced by incubation with increasing concentrations (10(-12)-10(-9) mol/l) of ACTH (1-24 or 1-39) for 1 h. Among 12 ACTH-dependent steroids tested (10(-9)-10(-6) mol/l), only corticosterone (IC50 = 2 x 10(-7) mol/l), 18-OH-corticosterone and 11 beta-OH-androstenedione showed a significant dose-dependent inhibition of 11 beta-HSD activity. The percentage conversion rate of cortisol to cortisone was concentration dependent over the whole range of cortisol concentrations tested (10(-8) - 10(-5) mol/l. A direct inhibitory effect of ACTH on 11 beta-HSD is, therefore, unlikely. The only steroids inhibiting the conversion of cortisol to cortisone are natural substrates for 11 beta-HSD. Kinetic studies show a saturation of the enzyme at high cortisol concentrations. Thus, the reduced percentage renal cortisol inactivation in EAS seems to be due mainly to overload of the enzyme with endogenous substrates (cortisol, corticosterone and others) rather than to direct inhibition of 11 beta-HSD by ACTH or ACTH-dependent steroids, not being substrates of 11 beta-HSD.

Pharmacokinetics of methylprednisolone and rejection episodes in kidney transplant patients.

Rejection crises after kidney transplantation could be associated with individual variability of pharmacokinetic parameters of steroids. We therefore investigated the individual pharmacokinetics of methylprednisolone on day 2 (60 mg intravenously) and day 4 (60 mg per os) in 40 patients after kidney transplantation. Methylprednisolone was determined in serum by HPLC. Within 6 months, all rejection episodes were recorded and confirmed by kidney transplant biopsy. Values are given as nonparametric medians with the 95% confidence interval (0.95 CI). The 7 patients with a rejection within the first 10 days had a methylprednisolone clearance of 437 ml/min (162-756) that was significantly higher than the 220 ml/min (121-604) in the 22 patients without a rejection episode (P = 0.04). In the complete group of 18 patients having a transplant rejection episode within 6 months, the methylprednisolone elimination half-life after oral dosage was 2.5 hr (1.6-3.9) and significantly shorter than 2.9 hr (1.7-4.0) in 22 patients without rejections (P = 0.03). No differences were seen for body weight, number of mismatches, cold ischemia time, immunosuppressive regimens, and other pharmacokinetic parameters of methylprednisolone (e.g. bioavailability, distribution volume, trough levels). We conclude that pharmacokinetic variability may contribute to the lack of immunosuppressive efficacy in patients with a short halflife of steroids. Therefore, a twice daily dose fraction might be useful for low-dose steroid regimens in kidney transplantation.

Decidua and placenta in mice after treatment with a synthetic glucocorticoid.

To investigate a possible long-term effect of glucocorticoids on decidua and placenta of mice, a single dose of 24 mg kg-1 body weight triamcinolone acetonide in crystalline suspension was given subcutaneously to NMRI mice on gestational day (GD) 2. Deciduae and placentae, as well as corticosterone and triamcinolone concentrations in maternal plasma of GDs 10 and 17 were examined. NADPH-cytochrome P450 reductase involved in drug biotransformation was detected immunocytochemically and showed co-localization with NADPH diaphorase histochemistry in the decidua and placenta. Both reactions were higher in endothelial cells of decidual sinusoids on GD 10, but were lower on GD 17 in the trophoblast, spongiotrophoblast and extraplacental visceral yolk-sac epithelial cells of treated mice than in untreated animals. Histochemistry of 11 beta-hydroxysteroid dehydrogenase, an enzyme that metabolizes biologically active adrenocortical steroids and their synthetic congeners in the placenta, showed higher activity on GD 17 in enlarged labyrinthic trophoblast I cells of treated mice than in untreated animals. As corticosterone concentrations were still decreased on GD 17, when triamcinolone concentrations were no longer detectable, a long-term suppression of adrenal gland function seems obvious.

Impaired renal 11 beta-oxidation of 9 alpha-fluorocortisol: an explanation for its mineralocorticoid potency.

9 alpha-Fluorocortisol (9 alpha FF) is about 200 times more potent as a mineralocorticoid than cortisol (F) in man, although it binds with the same affinity as F and aldosterone to the human mineralocorticoid receptor. The low mineralocorticoid activity of F has been shown to be due to its rapid conversion by the kidney to cortisone (E), which does not bind to the receptor. Therefore, we compared the conversion of F to E with that of 9 alpha FF to 9 alpha-fluorocortisone (9 alpha FE) by 11-hydroxysteroid dehydrogenases in man in vivo and in vitro. Single oral doses of 9 alpha FF, 9 alpha FE, and F were given to normal males, and the excretion of free 9 alpha FF, 9 alpha FE, F, and E was measured in urine. Human kidney and liver slices were incubated with unlabeled steroids, and the free 11-hydroxy- and 11-oxosteroids were quantitated after high performance liquid chromatography separation by UV absorption. Oral F (5 mg) is excreted 70% as free E and 30% as free F (percentage of free steroids only). Oral 9 alpha FF (5 mg) is excreted 90% as free 9 alpha FF and 10% as free 9 alpha FE. Free 9 alpha FF excretion is 14 times greater than that of F after ingesting an identical dose. Oral 9 alpha FE (4 mg) is also excreted 90% as 9 alpha FF and 10% as 9 alpha FE. Kidney slices convert F much faster to E than 9 alpha FF to 9 alpha FE. The conversion of 9 alpha FE to 9 alpha FF is, on the contrary, much faster than that of E to F. Thus, the equilibrium of the reaction is on the 11-oxo side for F/E and on the 11-hydroxy side for 9 alpha FF/9 alpha FE. The interconversion of both pairs of steroids is inhibited by glycyrrhetinic acid in a dose-dependent manner. Liver slices do not measurably convert 9 alpha FF to 9 alpha FE, but do rapidly convert 9 alpha FE into 9 alpha FF. Reflecting this negligible conversion of 9 alpha FF to 9 alpha FE and the low plasma-protein binding of 9 alpha FF, free urinary 9 alpha FF excretion is much higher than that of F after the same oral dose.(ABSTRACT TRUNCATED AT 400 WORDS)

Urinary oestriol-16-glucuronide determined by "on-line" liquid chromatography.

A fully automated method for the specific assessment of oestriol-16-glucuronide in urine is described. "On-line" sample preparation, including enrichment, pre-purification, focusing and injection, is combined with automated high-performance liquid chromatographic separation and fluorimetric quantification. Losses of oestriol-16-glucuronide throughout the total procedure are negligible. Thus, external calibration is feasible for quantification. Coefficients of variation are 4.44% (n = 12) for intra- and 7.85% (n = 9) for interassay variability. Assay sensitivity is 430 nmol/l. The excretion rates of oestriol-16-glucuronide relative to creatinine were estimated in 85 pregnancy urines. These oestriol-16-glucuronide estimates correlated well with those of total urinary oestriol, determined by high-performance liquid chromatography after acid hydrolysis (r = 0.957). The reference ranges of oestriol-16-glucuronide for the different weeks of gestation were evaluated. Unlike the determination of total oestriol, the present method does not need an hydrolysis step. The method is therefore well suited for the biochemical monitoring of fetal well-being under emergency conditions.

20-Dihydroisomers of cortisol and cortisone in human urine: excretion rates under different physiological conditions.

The urinary excretion rates of free cortisol and cortisone as well as of their 20-dihydroisomers have been studied in normal subjects under different physiological or pharmacological conditions. For the estimation of steroid excretion rates, a fully automated, liquid-chromatographic method was used. In normal subjects, the median steroid excretion rates of free cortisol, cortisone, 20-alpha-dihydrocortisol, 20-beta-dihydrocortisol, 20-alpha-dihydrocortisone and 20-beta-dihydrocortisone were 6.7, 8.0, 9.8, 5.2, 5.7 and 1.3 mumol/mol creatinine. The excretion rates measured at three different intervals of the day followed a circadian rhythm similar to that known for the cortisol secreting activity of the adrenal gland. After adrenal stimulation by i.v. application of 250 micrograms of tetracosactide hexaacetate, (Synacthen, corticotropin beta 1-24) excretion of urinary cortisol was significantly higher than those of the other steroids. During a 24 h infusion of corticotropin beta 1-24, the excretion rates of cortisol and its C-20 reduced isomers increased to a significantly greater extent than those of cortisone and its C-20 reduced isomers. During a four-hour infusion of hydrocortisone, the relative increase of cortisol excretion was greater than that of the other steroids. During a five-hour infusion of metyrapone at different dosages, the excretion of all steroids decreased in a dose-dependent manner. The present data indicate that the 20-dihydroisomers of cortisol and cortisone in human urine primarily originate from the peripheral metabolism of cortisol rather than from adrenal secretion.(ABSTRACT TRUNCATED AT 250 WORDS)

A fully automated method for sample clean-up prior to chromatographic and immunological analysis.

A fully automated technique for the extraction and clean-up of low-molecular weight analytes from human serum and urine is presented. Its efficiency for sample clean-up prior to immunological assay is demonstrated for six adrenal and gonadal steroid hormones, calcidiol and the peptide hormones gastrin, insulin and glucagon. A more intensive sample clean-up, which is mandatory for liquid chromatographic analysis, is reported for serum cortisol. With the exception of calcidiol, the extraction of steroids is almost complete. Recovery of peptides is about 80% and depends on the nature of the peptide and on protein-analyte dissociating diluents. Precision of recovery is lower than 7% (CV) for all analytes studied. One hundred serum or urine samples can be cleaned up without loss of efficiency by only one solid-phase cartridge. The technique is in principle applicable to all other analytes with physico-chemical structures similar to the analytes in the present study.

Total urinary estriol determined by "on-line" liquid chromatography with ultraviolet detection.

We report a fully automated method for the specific assessment of total estriol in urine. Acid-hydrolyzed urine is concentrated on a short reversed-phase column and prepurified with methanolic Tris buffer, the pH of the buffer increasing from 7 to 11. The anionic form of the estriol molecule is eluted with methanolic NaOH. By acidification in a mixing chamber, estriol in its neutral form is focused on the top of a second short reversed-phase column, effectively chromatographed on an analytical column, and quantified by ultraviolet absorbance at 278 nm. Losses of estriol throughout the total procedure are negligible and thus external calibration is feasible for quantification. Analytical recoveries for estriol-supplemented urines ranged from 98.3 to 105%. Replicate analyses of a urine containing 68 mumol of estriol per liter gave a CV of 2.76%. As little estriol as 2 mumol/L can be detected. Results from pregnancy urines correlated well with those of radioimmunoassay. The method is especially suited for clinical emergencies in a routine laboratory.

Simultaneous enzyme immunoassay of cortisol and dehydroepiandrostone-sulphate from a single serum sample using a transferable solid phase system.

An enzyme immunological methodology for the direct and simultaneous estimation of serum cortisol and dehydroepiandrosterone-sulphate (DHEA-S) useful for the biochemical differential diagnosis of Cushing's syndrome has been developed. The combined estimation of both steroids is more economical in time, work and materials than two separate assays. Two solid phases, a microtitre plate and a covering transferable needle lid system were used in the present procedure. Both solid phases are first coated with anti-rabbit IgG and then each with a specific antiserum. Horseradish peroxidase was used as marker enzyme and tetramethylbenzidine as the chromogen for measuring enzyme activity. No extraction or deproteinization steps are involved. The turn around time for 41 samples (in duplicate) is 3 h. The detection limit of the assay is 5 pg/well for cortisol and 10 pg/well for DHEA-S. Results of the present method correlated well (cortisol, r = 0.95; DHEA-S, r = 0.98) with those of commercial radioimmunoassays using iodinated labels. Thus, this technique offers a convenient non-isotopic procedure in the routine clinical laboratory.

Automated liquid chromatographic determination of the 20-dihydro isomers of cortisol and cortisone in human urine.

A fully automated method for the simultaneous assessment of cortisol, cortisone and their 20-dihydro isomers in human urine is described. On-line sample enrichment, prepurification, focusing and injection are combined with automated high-performance liquid chromatographic separation and quantification. Losses of steroids throughout the total procedure are negligible. Thus, external calibration is feasible for quantification. Coefficients of variation range between 8.7 and 17.0% for inter-assay variability and between 1.3 and 5.2% for intra-assay variability. Assay sensitivity is 15 nmol/l. In normal students, the medians of the relative excretion rates of free 20 alpha-dihydrocortisol, 20 alpha-dihydrocortisone, 20 beta-dihydrocortisol and 20 beta-dihydrocortisone were 10.9, 6.1, 7.7 and 4.4 mumol/mol creatinine. The fully automated feature renders the present method well suited for routine diagnosis of hypercorticoidism.

Enzyme immunoassay of serum cortisol using a new transferable needle lid technique.

A simple enzyme immunoassay of serum cortisol using for the first time a transferable needle lid as solid phase has been developed. The needles coated with second antibody and dipped into the wells of a microtitre plate bind the specific antibody of a competitive enzyme immunoassay mixture. Bound enzyme activity is estimated in the wells of another microtitre plate. This technique provides further advantages on the frequently used microtitre plate version. Washes between the immunological and the enzymatic reaction take very short time and are less laborious. Due to the facility of simultaneous starting and stopping of all reactions, a better precision and sensitivity is achieved. In the present cortisol assay, horseradish peroxidase covalently coupled to cortisol-21-hemisuccinate was used as enzyme label and tetramethylbenzidine as the chromogen for measuring enzyme activity. No extraction or deproteinization steps are involved. The turn around time for 41 samples (in duplicate) is 2.5 h. The detection limit of the assay is 5 pg of cortisol per well. Results of the present method correlated well (r = 0.92) with those of a commercial radioimmunoassay using iodinated cortisol.

Measurement of urinary free 20 alpha-dihydrocortisol in biochemical diagnosis of chronic hypercorticoidism.

Using liquid chromatography, we estimated the urinary excretion of 20 alpha-dihydrocortisol (20-DH) and urinary free cortisol (UFC) in normal subjects and in 40 patients with Cushing's syndrome of different etiologies. The median normal excretion rate (nmol/24 h) was 174 for 20-DH and 68 for UFC, the 20-DH/UFC ratio thus being 2.55. For patients with Cushing's syndrome, the excretion rate was 1798 for 20-DH and 298 for UFC, the ratio 6.03. We evaluated the effect of acute stimulation of adrenal secretion on 20-DH and UFC by administering corticotropin to six normal subjects. After such stimulation, the excretion rate was 566 for 20-DH and 1238 for UFC (ratio 0.45). Whereas 20-DH excretion rate exceeded the normal range in all patients, six patients had normal or even below-normal values for UFC excretion. Evidently, measurement of urinary 20-DH is a better test for chronic hypercorticoidism than is measurement of urinary UFC, and chronic hypercorticoidism can be differentiated from the acute state by the 20-DH/UFC ratio.

Urinary excretion rates of 15 free steroids: potential utility in differential diagnosis of Cushing's syndrome.

To evaluate their potential usefulness in the differential diagnosis of Cushing's syndrome, we estimated the urinary excretion rates of the following non-metabolized, unbound steroid hormones: pregnenolone, progesterone, 17-OH-pregnenolone, 17-OH-progesterone, dehydroepiandrosterone (DHEA), androstenedione, testosterone, dihydrotestosterone, 11-deoxycorticosterone, 11-deoxycortisol, corticosterone, cortisol, 18-OH-11-deoxycorticosterone, 18-OH-corticosterone, and aldosterone. These were measured in normal subjects and in patients with Cushing's disease, adrenal adenoma, or ectopic corticotropin syndrome. We used "high-performance" liquid chromatography and subsequent radioimmunoassay. Our results indicate that simultaneous estimation of urinary free cortisol and DHEA may be useful in differential diagnosis of hypercorticoid states due to adrenal adenoma and Cushing's disease.

Determination of dehydroepiandrosterone sulfate in plasma by a one-step enzyme immunoassay with a microtitre plate.

We have developed a rapid and cost-effective enzyme immunoassay for dehydroepiandrosterone sulfate (DHEA-S) in plasma, performed with samples on a microtitre plate within 2.5 h. No extraction or centrifugation steps are involved. The 3-hemisuccinate of dehydroepiandrosterone is labeled with horseradish peroxidase, then mixed with hydrogen peroxide substrate in the presence of the chromogen, tetramethylbenzidine. The detection limit of the assay is 12.5 pg of DHEA-S per well. Intra- and interassay CVs at three steroid concentrations (12.8, 1.28, and 0.16 mumol/L) ranged from 2.3 to 5.4% and 6.1 to 8.4%, respectively. Results correlated well (r = 0.95) with those of a radioimmunoassay with iodinated DHEA-S. The turnaround time for 41 samples (in duplicate) is 2.5 h, which includes 2 h of incubation time. The sensitivity of this one-step version and the linearity of its standard curve are equivalent to those of a less practicable two-step version. This technique may replace coated-tube enzyme immunoassays for routine use.