CCA 4581
URINARY EXCRETION OF DEHYDROEPIANDROSTERONE IN NORMAL CHILDREN AND IN PATIENTS WITH ADRENOCORTICAL DISORDERS
S. B. PAL AND W. M. TELLER
Universität Ulm, Abteilung Endokrinologie und Stiffwechsel, Zentrum für Innere Medizin und Kin- derheilkunde, D 79 Ulm (Donau), (West Germany)
(Received May 7, 1971)
SUMMARY
A systematic analytical procedure for the determination of urinary dehydro- epiandrosterone (DHA) was developed. An aliquot of 100 ml from a 24-h urine collec- tion was subjected to hot hydrolysis at neutral pH for 6 h. Liberated steroids were extracted with diethyl ether and the ketones were separated from the non-ketonic compounds by Girard’s T reaction. Dehydroepiandrosterone was separated by paper chromatography in Bush A system and, after eluting, the concentration of DHA was measured spectrophotometrically by a modified Zimmermann reaction. The overall average recovery of the procedure was 83.3% where DHA-7 x-T(n) sulphate — potas- sium salt was added to the urine and taken through the analysis. The identity of DHA was confirmed after making various chemical derivatives and checking their chromatographic mobilities against derivatives prepared from authentic steroids. When this procedure was applied to the urines of 18 normal children (10 boys and 8 girls), aged between 5 years 1 month-15 years 8 months, the average level of DHA was 0.6 mg/24 h (range 0.22-1.9 mg/24 h). The excretion of DHA was 26% (range 12.1-47%) of total neutral 17-ketosteroids (estimated after enzymatic hydrolysis of urine followed by solvolysis). From 10 children, aged 9-15 years, with various adreno- cortical disorders, the levels of DHA were found to be above normal. Of these 10 patients, 2 had adenoma, 4 carcinoma and 4 nodular hyperplasia; the average value of DHA was 3.6 mg/24 h (range 2.5-5.1 mg/24 h). The excretion of DHA was 68.4% of total neutral 17-ketosteroids (range 55.5-83.6%). All these patients had to undergo surgery and, in each case, DHA was estimated after 3 months, when the average value was found to be 0.7 mg/24 h (range 0.38-1.1 mg/24 h). The average excretion of DHA was 30% (range 15.8-43%) of total neutral 17-ketosteroids. From these results it can be concluded that urinary DHA determination could be used for the diagnosis of adrenocortical disorders.
Part of this work was presented at the 9th Meeting of the European Society for Paediatric Endo- crinology, Lyon 16-20th July, 1970.
INTRODUCTION
It is well established that the adrenal is able to secrete steroids which are strongly androgenic in certain disease states, such as in adrenal cancer, adrenal hyper- plasia and virilization. Dehydroepiandrosterone (38-hydroxy-5-androsten-17-one) which has weak androgenic potency, is secreted by the adrenal and can also serve as an efficient precursor of urinary androsterone and etiocholanolone. It has been isolated from human adrenal tissue and, furthermore, biosynthesis of DHA from [1-14C]acetate has been reported. It has also been isolated from adrenal venous blood. Dehydro- epiandrosterone is found in the urine of normal women and in male castrates. Its urinary excretion increases after the administration of ACTH and decreases after the administration of cortisol. The level of DHA in urine is sometimes very high in patients with various diseases of the adrenal1 and there is little doubt that it derives partly from adrenal precursors. Therefore, it was hoped that an accurate determination of urinary excretoin of DHA in various adrenocortical disorders could be clinically useful.
First, a suitable procedure for the determination of urinary DHA alone, had to be set up. Urinary DHA is mainly excreted as a sulphate. As, in our experience, no reliable sulphatase preparation was commercially available that would give effi- cient hydrolysis of DHA sulphate, various hydrolytic procedures were studied in order to find an effective way of hydrolysing DHA sulphate and of determining this steroid accurately.
This communication will be confined to cases of children only, although, during the course of this work, urine samples from adult subjects and patients were used for recovery experiments.
MATERIALS AND METHODS
All chemicals and solvents used were of analytical grade.
Suc d’Helix pomatia. 1 ml contains 100000 units of ß-glucuronidase (Fishman), 50000 units sulphatase (Whitehead) (Industrie Biologique Française, S.A., Quai du Moulin de Cage, Genevilliers, Seine, France). These quantities of enzymes, contained in a I-ml ampoule, were dissolved into 9 ml of M sodium acetate-acetic acid buffer of pH 5.2 before use.
Glass-distilled water was used throughout the work.
Glass-stoppered test tubes 15 × 2.5 cm, approximate capacity 54 ml and 12.5 × 1.5 cm, approximate capacity 15 ml were used (Quickfit & Quartz Ltd, U.K.).
Dehydroepiandrosterone was purchased from BDH Chemicals Ltd., Poole, U.K.
Dehydroepiandrosterone sulphate-sodium salt and other reference steroids used were obtained as gifts from Steroid Reference Collection (Medical Research Council Chemistry Department, Westfield College, Hampstead, London, U.K.).
Dehydroepiandrosterone-7x-T(n) sulphate, dihydrate (potassium salt), (n indi- cates the nominal position of the isotope where is any uncertainty as to whether the labelling is confined to the normal position) specific activity 930 mC/mmole; n-hexa- decane-1,2-T were purchased from The Radiochemical Centre, Amersham, Bucking- hamshire, U.K.
2,5-Diphenyloxazole (PPO), 1,4-bis-2,4-methyl-5-phenyloxazolyl benzene (dimethyl- Clin. Chim. Acta, 38 (1972) 371-378
POPOP), naphthalene, dioxane and toluene, all scintillator grade, were used (E. Merck A.G.).
Dioxane scintillator was prepared after dissolving 60 g naphthalene, 8 g PPO and 0.2 g POPOP into a mixture of solvents containing 50 ml ethanol, 100 ml toluene and 900 ml dioxane and stored in amber glass bottles at 4°.
Toluene scintillator was prepared by adding 3.0 g PPO and 0.3 g POPOP in I litre of toluene and was stored in an amber glass bottle at 4°.
All the extracts for paper chromatography or liquid scintillation counting were evaporated to dryness throughout this work in a water bath at 40° under a stream of nitrogen. The extracts were spotted on paper chromatograms under a stream of nitrogen, later on replaced by air.
The radioactive extracts were evaporated directly in high quality glass vials (Packard Low-Potassium-I) using 15 ml of scintillation solution. The Packard Liquid Scintillation spectrometer was operated at an average efficiency of 35% for 3H with a background of 14 counts/min. Sample measurements were corrected for quenching by the internal standardization method. n-Hexadecane-1,2-T was used as an internal standard. The background and radioactive samples were counted long enough to collect 10000 counts.
A rotary-film evaporator (Rotavapor) R was used (Büchi, Flawil, Switzerland).
The paper chromatograms were prepared on a sheet of Whatman No. 2 filter paper without any previous washing and I-cm wide lanes were cut out between the strips. A sample of DHA extracted from pooled urines, purified by paper chromatog- raphy and repeatedly crystallised, was used as reference steroid at the same time as authentic DHA. Chromatograms were equilibrated for at least 3 h, if not overnight, and developed at room temperature, between 20-22°, keeping the tanks in a wooden draft-proof box, and, later on, in a thermostatically controlled hot box at 30°. Bush A system (1000 ml petroleum ether, b.p. 100°-120°: 800 ml methanol: 200 ml water) was used throughout the work.
Urine was collected over a period of 24 h under 5 ml toluene as preservative and stored at 4° until analysed. In case of adults, volumes of all the urine collections were made up to 2 litres if they were less than this volume. In the case of children, when the volume was less than 500 ml, it was made up to this volume; urine collec- tions of over 500 ml but of less than 1 litre were made up to 1 litre. Distilled water was used in each case.
The total neutral 17-ketosteroids were estimated after hydrolysing 100 ml urine from 24-h collections with Suc d’Helix pomatia, followed by solvolysis2. The ketonic steroids were separated from the non-ketones by Girard’s T reaction and the concentration of 17-ketosteroids was measured on a 1/5th portion of the steroid residue by a modified Zimmermann procedure.
The total neutral 17-ketosteroids and dehydroepiandrosterone were determined in the patients before surgery and three months after surgery.
Different procedures were carried out to hydrolyse the urine and to find an efficient condition under which to hydrolyse DHA sulphate and achieve an accurate estimation of this steroid. Urine samples from a normal man, a normal woman and from various patients with endocrine disorders, were subjected to:
I. Hot acid hydrolysis (10 ml conc. HCl/100 ml urine) :
(a) extracted with diethyl ether
(b) overlay with benzene during hydrolysis3
(c) overlay with toluene during hydrolysis.
2. Hot hydrolysis under neutral pH (ref. 4)
(a) overlay with benzene during hydrolysis
(b) overlay with toluene during hydrolysis
(c) extracted with diethyl ether after hydrolysis.
3. Enzyme hydrolysis with Suc d’Helix pomatia, extracted with ether, followed by solvolysis of the aqueous phase, and washings. In each case, 100 ml urine was taken from 24-h collections, after hydrolysis, the extracted steroids were separated for ketones from non-ketones by Girard’s T reaction and further purified by paper chromatography. The concentration of DHA was finally measured spectrophoto- metrically by a modified Zimmermann procedure.
As mentioned before, comparison of different hydrolytic procedures was per- formed and DHA was recovered in each case by adding 100 µg of DHA sulphate- sodium salt and 15000 counts/min of [3H]DHA sulphate-potassium salt to 100 ml of water and to 100 ml of urine. Ten experiments were performed in each case.
Chemical derivatives such as acetate, epoxide and chromic acid oxidation product of DHA isolated from urine, were prepared in parallel with pure DHA. The respective acetate, chromic acid oxidation product and epoxide of the test and pure steroid were chromatographed at the same time. When these paper chromatograms were dipped into Zimmermann reagent to locate the position of steroids, in each case a single spot was found on the chromatogram with the same mobility as the derivative prepared from the reference steroid. The melting point of isolated DHA was checked and agreed with the published figures. A sulphuric acid absorption curve was also prepared from DHA isolated from urine, wich agreed with the curve prepared from pure steroid.
RESULTS
It was found, by comparing different hydrolytic procedures for urinary DHA, that the maximum amount of DHA was estimated after hot hydrolysis of the urine under neutral conditions overlaid with benzene, toluene or extracting with ether. The pattern was the same for urines from normal human individuals (2 males and 2 females) and patients with different endocrine disorders (2 adrenal hyperplasia, 2 simple hirsutism, I hirsutism with polycystic ovaries and I hirsutism with ovarian tumour).
When the recovery experiments (n = 10) for DHA were performed, applying different hydrolytic procedures by the addition of DHA-sulphate-sodium salt and [3H]DHA-sulphate-potassium salt to water and to urine, in each case, the maximum chemical and radioactive recovery of DHA was found after hot hydrolysis under neutral conditions, overlaid with benzene, toluene or extracting with ether.
Overlay with benzene:
DHA
[3H]DHA
Water 74.5% (SD ___ 1.23) range (72.8-76.2%) 81.7% (SD ± 0.67) range (80.8-82.6%)
Urine 74.1% (SD +0.91) range (72.7-75-3%) 81.1% (SD±0.9) range (79.4-82.7%)
Overlay with toluene: DHA
Water 74.2% (SD ± I.I) range (72.3-75.6%) Urine 74.2% (SD ± 0.82) range (73.2-75.6%)
[3H]DHA
83.8% (SD :± 0.57) range (83.2-84.6%) 80.6% (SD ± 1.28) range (78.6-82.5%)
Extraction with ether:
DHA
[3H]DHA
Water 73.8% (SD ± 1.26) range (72.3-75-7%) 83.3% (SD ± 0.94) range (82.2-84.7%)
Urine 74.1% (SD ± 0.68) range (73.2-75.1%) 81.4% (SD ± 0.73) range (80.2-82.6%)
Table I shows details of the method which was finally adopted for the deter- mination of urinary DHA used throughout the work. The results presented are not corrected for recoveries.
TABLE I
FLOW-SHEET OF THE METHOD FOR DETERMININATION OF URINARY DHA
A 100-ml aliquot of urine from a 24-h collection +[3H]DHA sulphate was taken. 4
Hydrolysed for 6 h (100°) at pH 7, cooled. Liberated steroids were extracted with diethyl ether, washed with o.I N NaOH and water; evaporated to dryness.
¥
Ketones were separated from non-ketones by Girard’s T reaction. Steroid residue was chromato- graphed for DHA in Bush A system, eluted and evaporated.
¥
The concentration of DHA was determined spectrophotometrically by Zimmermann reaction.
Normal values
From 18 children (10 boys and 8 girls) aged between 5 years 1 month and 15 years 8 months, all of whom hospitalised, but without endocrine disorders, the excre- tion of total neutral 17-KS was between 1.0-6.0 mg/24 h (mean 2.3) and that of DHA was between 0.22-1.9 mg/24 h (mean 0.6); the percentage (%) DHA of total 17-KS being between 12.1% to 47% (mean 26%).
Pathological values
In 10 children with adrenocortical disorders, DHA was estimated. Of these 10 patients, 2 had adenoma, 4 carcinoma and 4 nodular hyperplasia; the average value of DHA was 3.6 mg/24 h (range 2.5-5.1 mg/24 h); the total neutral 17-KS was 5.2 mg/24 h (range 4.5-6.3 mg/24 h). The excretion of DHA was 68.4% of total neutral 17-KS (range 55.5-83.6%). In each case DHA was estimated 3 months after surgery and the average value of DHA was found to be 0.7 mg/24 h (range 0.38-1.1 mg/24 h). The average excretion of DHA was 30% (range 15.8-43%) of total neutral 17-KS.
DISCUSSION
The comparison of different hydrolytic procedures for dehydroepiandrosterone sulphate shows that hot acid hydrolysis was found to be disappointing. It is well established that hot acid hydrolysis forms artifacts of the steroid molecule due to dehydration, halogenation and other molecular rearrangements5; also, it is not certain
whether hydrochloric acid really completes the hydrolysis of DHA sulphate as poor chemical and radioactive recoveries were found after the addition of DHA sulphate- sodium salt and [3H]DHA sulphate-potassium salt. Hot acid hydrolysis after overlay with an organic solvent, such as benzene or toluene, was found to be more satisfactory as it is a simultaneous hydrolysis and extraction process; although recoveries of added steroid conjugates were better but the procedure was not adopted as benzene is a highly toxic solvent and toluene was found to be inconvenient to handle due to its high boiling point (109-1I2º).
Dehydroepiandrosterone is mainly excreted as sulphate conjugates in human urine. Although its excretion as glucuronides in normal children of both sexes (aged between 6 months and 16 years) had been reported6, no such information is available from the authors’ laboratory in this respect. This finding appeared, in general, to be very uncertain. It has been reported? that the glucuronide fraction may be considered as an artifact from DHA sulphate shown to be transformed to 68-hydroxy-3,5-cyclo- 5«-androstan-17-one, even under conditions as mild as glucuronide hydrolysis. This rearrangement occurred particularly to a large extent when urine was directly sub- jected to enzyme hydrolysis for 3 days. Under these conditions, the glucuronide fraction contained up to 46.5% of iso-androstenolone8,9. It was much less evident when the conjugates were first extracted from the urine and were then hydrolysed. Enzyme hydrolysis followed by solvolysis? was also rejected. In the first place, ß-glucu- ronidase does not hydrolyse sulphate conjugates; it has been reported that, unless interfering substances have been eliminated from the urine, total hydrolysis of steroid glucuronosides cannot be achieved with ß-glucuronidase. Secondly, due to the very slow activity of sulphatase, its efficiency in hydrolysing a sulphate conjugate is un- certain, but solvolysis of the aqueous layer (after the extraction of liberated steroids by enzyme hydrolysis) at pH I with sulphuric acid, sodium chloride and ethyl acetate, was found to be an ideal condition for hydrolysing all steroid sulphate con- jugates completely. This procedure was not strictly necessary for this study, especially when DHA alone was to be estimated; besides this, epiandrosterone migrates very closely to DHA on a paper chromatogram. If epiandrosterone is not separated by an additional chromatographic step, DHA could be overestimated.
As this study was mainly concerned with the estimation of DHA only, the hot hydrolysis procedure4 was adopted, which consists of boiling the urine at neutral pH for 6 h. In this procedure, 36-sulphates of 38-hydroxy-45 steroids were selectively hydrolysed, androsterone sulphate and epiandrosterone sulphate were not hydrolysed during the boiling of urines. The chemical and radioactive recoveries of DHA by the addition of DHA sulphate conjugates were good. Hot hydrolysis after overlay with benzene or toluene did not improve the recovery significantly. Thus the hot hydrolysis procedure was adopted; after hydrolysis of the urines at neutral pH for 6 h, they were cooled, DHA was extracted with ether and the residue obtained from the ether extract was subjected to Girard’s T reaction to separate the ketones from non-ketones and, after further purification by chromatography, DHA was estimated by the Zim- mermann reaction. The specificity of the Zimmermann reaction in human 17-keto- steroids is 100%. This reaction is easier and more rapid in a laboratory where a large number of samples are analysed.
We have not, so far, come across any values published in the literature for urinary DHA in children obtained by hot hydrolysis at neutral pH. The range of
URINARY EXCRETION OF DHA
values of DHA from normal children given in this communication agreed in general with other published values (Table II) in spite of some differences in methodology. In a few cases, they were lower than the values given here. It has also been reported that DHA was not detected in urines from children, particularly in the pre-pubertal stage. No DHA was found in the urine of children under the age of 9 years, except in sexually premature children10. Similarly, no DHA was detected in 5 boys aged 5 to 7 years and in 7 girls aged 5 to 7 years11.
| Author | Age, years | Sex | Values (range) mg /24 h |
|---|---|---|---|
| Kádár et al.12 | 5 -I3 | 18 boys | 0.1 -0.5 |
| Kádár et al.12 | 5 -13 | 20 girls | 0.1 -1.I |
| Vestergaard13 | 3 -17 | 18 boys | 0 -2.54 |
| Vestergaard13 | 3 -16 | 18 girls | 0 -0.60 |
| Paulsen et al.14 | 10 -16 | 34 (mixed) | 0.03 -0.48 |
| Teller15 | 106/12-124/12 | II (mixed) | 0.004-0.07 |
| Teller15 | 129/12-160/12 | 8 (mixed) | 0.014-0.8 |
| Steeno et al.18 | II -15 | 60 boys | 0.048-0.34 |
| Blunck17 | 7 -IO | 12 (mixed) | 0.008-0.114 |
| Blunck17 | 14 -15 | II (mixed) | 0.075-0.48 |
| Gupta18 | 3 -14 | 42 (mixed) | 0 -0.42 |
| Berger et al.19 | 5 -18 | 174 boys | 0 -0.8 |
| Berger et al.19 | 5 -16 | 127 girls | 0 -1.55 |
| Present authors | 51/12-156/19 | Io boys | 0.22 -1.9 |
| Present authors | 76/12-141/12 | 8 girls | 0.30 -I.2 |
The values of DHA from 10 children with adrenocortical disorders were found to be higher than those obtained from 18 normal children. When DHA was again estimated in these children with adrenocortical disorders 3 months after surgery, the level of DHA was within the normal range.
It is essential to use great care in interpreting the results obtained from the urinary excretion of steroids in children, especially of androgens, before the onset of puberty. All methods developed for steroid investigation in adults must be reassessed before being applied to children, particularly in early infancy20. An adult human individual in good physical condition has a fairly stable maximal level of activity, especially in regard to androgen metabolism, whilst the child is undergoing constant physiological changes and has sometimes not attained normal adult levels of androgen production even at the age of 15 and 16 years. Dehydroepiandrosterone has not invariably been detected as a constituent of urine in infants and in children10,11,14, as it is in normal adults. However, from the present investigation, it is possible to con- clude that the determination of urinary DHA in children could be used as an index of abnormal adrenocortical function and was found to be clinically useful.
ACKNOWLEDGEMENTS
We wish to thank Professor Dr. E. F. Pfeiffer and other physicians and surgeons for their interest shown during the course of this work; Professor W. Klyne, Westfield College, London, for generously providing reference compounds from the Medical
Research Council Steroid Reference Section and Fräulein Bärbel Schäfer for per- forming Zimmermann estimations for dehydroepiandrosterone in normal children. Our thanks also go to Frau Martha Rupp and to Mrs. M. R. Pal for their help in typing and preparing the manuscript, respectively.
REFERENCES
I H. L. MASON AND W. W. ENGSTROM, Physiol. Rev., 30 (1950) 213.
2 S. BURSTEIN AND S. LIEBERMAN, J. Biol. Chem., 233 (1958) 331.
3 P. VESTERGAARD AND B. CLAUSSEN, Acta Endocrinol., 39 Suppl. 64 (1962).
4 K. FOTHERBY, Biochem. J., 69 (1958) 596.
5 R. I. DORFMAN AND F. UNGAR, in Metabolism of Steroid Hormones, Academic Press, New York, 1965, p. 28.
6 B. LORAS, H. ROUX, B. CAUTENET, C. OLLAGNEN, M. FOREST, E. DE PERETTI AND J. BER- TRAND, in A. VERMEULEN AND D. EXLEY (Eds.), Androgens in Normal and Pathological Con- ditions, Proc. Second Symp. Steroid Hormones, Ghent, Excerpta Medica Foundation, Amster- dam.
7 S. BURSTEIN AND R. I. DORFMAN, Acta Endocrinol., 40 (1962) 188.
8 P. VESTERGAARD, Acta Endocrinol., 39, Suppl. 64 (1962) 50.
9 J. HAMMERSTEIN, in C. CASSANO, A. KLOPPER AND C. CONTI (Eds.), Research on Steroids, Vol. III, North-Holland Publishing Company, Amsterdam, 1968, P. 323.
10 E. P. PAULSEN AND E. H. SOBEL, Amer. J. Diseases Children, 100 (1960) 546.
II F. BEAS, R. P. ZURBRÜGG, J. CARA AND L. I. GARDNER, J. Clin. Endocrinol., 22 (1962) 1090.
12 A. KÁDÁR, T. FEHÉR AND O. KOREF, Arch. Diseases Childhood, 39 (1964) 257.
13 P. VESTERGAARD, Acta Endocrinol., 49 (1965) 436.
14 E. P. PAULSEN, E. H. SOBEL AND M. S. SHAFRAN, J. Clin. Endocrinol., 26 (1966) 329.
15 W. M. TELLER, Z. Ges. Exp. Med. 142 (1967) 222.
16 C. STEENO, W. HEYNS, H. VAN BAELEN, A. VAN HERLE AND P. DE MOOR, Eur. J. Steroids, 2 (1967) 273.
17 W. BLUNCK, Acta Endocrinol., 59 Suppl. 134 (1968) 9.
18 D. GUPTA, Clin. Chim. Acta, 26 (1969) 256.
19 H. BERGER, M. FINK, H. J. FRITZ, H. GLEISPACH, P. HEIDEMANN AND J. WOLF, Z. Klin Chem. Klin. Biochem., 8 (1970) 354.
20 F. L. MITCHELL AND H. L. SHACKELTON, in O, BODANSKY AND C. P. STEWART (Eds.), Advances in Clinical Chemistry, Academic Press, New York, London, 1969, p. 141.
Clin. Chim. Acta, 38 (1972) 371-378