Quantitative aspects of the determination of steroid profiles from urine by capillary gas chromatography

Quantitative aspects of the determination of steroid profiles

from urine by capillary gas chromatography

Citation for published version (APA):

Leunissen, W. J. J. (1979). Quantitative aspects of the determination of steroid profiles from urine by capillary

gas chromatography. Technische Hogeschool Eindhoven. https://doi.org/10.6100/IR126621

DOI:

10.6100/IR126621

Document status and date:

Published: 01/01/1979

Document Version:

Publisher’s PDF, also known as Version of Record (includes final page, issue and volume numbers)

Please check the document version of this publication:

• A submitted manuscript is the version of the article upon submission and before peer-review. There can be

important differences between the submitted version and the official published version of record. People

interested in the research are advised to contact the author for the final version of the publication, or visit the

DOI to the publisher's website.

• The final author version and the galley proof are versions of the publication after peer review.

• The final published version features the final layout of the paper including the volume, issue and page

numbers.

Link to publication

General rights

Copyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright owners and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights. • Users may download and print one copy of any publication from the public portal for the purpose of private study or research. • You may not further distribute the material or use it for any profit-making activity or commercial gain

• You may freely distribute the URL identifying the publication in the public portal.

If the publication is distributed under the terms of Article 25fa of the Dutch Copyright Act, indicated by the “Taverne” license above, please follow below link for the End User Agreement:

www.tue.nl/taverne Take down policy

If you believe that this document breaches copyright please contact us at: openaccess@tue.nl

QUANTITATIVE ASPECTS OF THE DETERMINATION OF STEROIO

PROFILES FROM URINE BY CAPILLARY GAS CHROMATOGRAPHY

QUANTITATIVE ASPECTS OF THE DETERMINATION

OF STEROIO PROFILES FROM URINE BY

CAPILLARY GAS CHROMATOGRAPHY

PROEFSCHRIFT

TER VERKRIJGING VAN DE GRAAD VAN DOCTOR IN DE TECHNISCHE WETENSCHAPPEN AAN DE TECHNISCHE HOGESCHOOL EINDHOVEN, OP GEZAG VAN DE RECTOR MAGNIFICUS, PROF. IR. J. ERKELENS, VOOR EEN COMMISSIE AANGEWEZEN DOOR HET COLLEGE VAN DEKANEN IN HET OPENBAAR TE VERDEDIGEN OP

VRIJDAG 2 NOVEMBER 1979 TE 16.00 UUR

DOOR

WILHELMUS JOHANNES JOZEF LEUNISSEN

GEBOREN TE HEERLEN

Dit proefschrift is goedgekeurd door de promotoren:

Prof.dr.ir. C.A.M.G. eraroers en

voor José

voor Monique en Ralph voor mijn ouders

CONTENTS

CONTENTS 4

CHAPTER 1 GENERAL INTRODUCTION 8

10 References

CHAPTER 2 STEROIO HORMONES 12

2.1 General structure and stereo-isomerism 2.2 Classification and biological action 2.3 2.4 2.5 2.6 2.2.1 Androgens 2.2.2 Oestrogens 2.2.3 Progestagens 2.2.4 Corticosteroids Biosynthesis and metabolism 2. 3. 1

2. 3. 2

Biosynthesis MetaboZism

Regulation of the biosynthesis Mode of action References 12 14 14

15

16 17 18 18 19 25 27 28

CHAPTER 3 QUANTITATIVE ASPECTS OF THE SAMPLE

PREPARATION PROCEDURE 30

3.1 Introduetion 30

3.2 Materials 31

3.3 Procedures for sample preparatien 32 3.3.1 Extraction of steroids and

steroid conjugates from urine

3.3.2 Hydralysis and salvolysis

32

of steroid conjugates 34 3.3.3 Purification of the sample 37 3.3.4 Selection of a suitable

3.4 Results and discussion 39 3.4.1 General remarks 39 3. 4. 2 XAD-2 extraction 41 3. 4. 3 Hydralysis with Helix pomatia

juice and salvolysis in

acidified ethyl acetate 46

3.4.4 Alkali wash 52

3.5 Determination of the accuracy and repeatability of the final

procedure 56

3.6 References 58

CHAPTER 4 FORMATION OF METHOXIME-TRIMETHYLSILYL

DERIVATIVES 63

4. 1 Introduetion 63

4.2 Materials 64

4.3 Procedures for the preparatien of

MO-TMS derivatives 65

4. 3.1 Formation of methoximes 65 4.3.2 Silylation of hydroxyl

groups with different

levels of steric hindrance 66 4.4 Determination of the optimum

conditions for MO-TMS

derivatisation 68 4. 4.1 Prepara ti on of MD-derivatives 68 4.4.2 E/Z isomerism 71 4. 4. 3 Persilylation of hydroxyl groups 76 4.4.4 Repeatability of the selected procedure 81 4.4.5 Purification of MO-TMS derivatives 82 4.5 References 86

CHAPTER 5 GAS CHROMATOGRAPHIC ANALYSIS 89

5.2 Preparation and characteristics of non-polar and polar phase-coated capillary columns 5.2.1 General remarks 5.2.2 Materials

5.2.3 Preparation of columns coated with non-polar

phases 5.2.4 Preparation of columns 90 90 92 92

coated with polar phases 93 5.2.5 Characteristics of

prepared columns 5.3 Description of apparatus 5.4 Qualitative analysis

5.4.1 Analysis of reference steraids on non-polar and polar columns 5.4.2 Optimization of

conditions

5.4.3 Determination of retention

95 95 98 98 100 indices lOl

5.4.4 Two-dimensional gas chroma

-tography 104

5.4.5 Gas chromatography-mass speetrometry

5.5 Quantitative analysis 5.5.1 Experimental

5.5.2 Accuracy and repeatability of the determination of n-alkanes

5.5.3 RepeatabiZity of the

anaZysis of MO-TMS steroid

110 112 112

112

derivatives 113

5.5.4 Determination of caZibration curves

5. 6· References

113 115

CHAPTER 6 APPLICATION TO URINARY SAMPLES 6.1 Introduetion

APPENDICES

SUMMARY

6.2 Procedure for determining steroid profiles from urine 6.2.1 Experimenta~

6.2.2 Repeatabi~ity of the se~ected methad 6.3 Stercid profiles of various

patients with an endocrine disorder

6.4 Hormonal aspects of breast cancer in post-menopausal woroen

6.4.1 Introduetion

6.4.2 Results and discussion 6.5 References SAl1ENVATTING DANKBETUIGING CURRICULUt-1 VITAE 119 119 121 121 121 124 144 144 145 150 154 156 158 160 161

CHAPTER 1

GENERAL INTRODUCTION

Progress in the field of biochemistry and related bio-sciences is largely dependent on the availability of

analytical techniques for the qualitative and quantitative determination of the compounds of interest.

Befare about 1950 there was little knowledge about the production and metabolism of steroid hormones, the excretion of metabolites and the interactions of the hypothalamus and hypophysis with the secretory organs. One of the reasans for this was the absence of suitable techniques to estimate steraids and steroid conjugates.

The first methods that became available for steroid analysis were spectroscopie determinations of groups of structurally related steroids. Classical determinations of this type include the Kober reaction for oestrogens1

, the Zimmerman reaction for 17-oxosteroids2 and the Porter-Silber reaction for corticbsteroids .with a 17a,2 1-di-hydroxy-20-one side-chain3

• For the routine evaluation of adrenal or gonadal functions improved versions of these methods are still widely used in laboratories for clinical chemistry, mainly because large numbers of samples can be analysed in a relatively simple way. These methods, how -ever, are not specific because of the contribution of many steraids with related structures and of the interterenee of non-steroidal chromogenie materials. Spectroscopie

deter-'

minations are therefore of limited diagnostic significance. Values within the normal range never exclude endocrine dysfunction because neither qualitative nor quantitative

changes in the excretion of individual metabolites can be detected. Spectroscopy can also be used for determining individual steroids, but such analyses require many purification and pre-separation steps4•

Modern techniques like radioimmunoassay (RIA) 5 _{are better} approaches for estimating individual steroids. With RIA methods a high specificity can be achieved, provided a specific antibody is available. If this prerequisite is met, only few purification steps, if any, will be required. An increasing number of steraids can be determined with

RIA because more antibodies become available. RIA analyses take little time and offer the opportunity to determine large numbers of samples. The sensitivity of these analyses allows the estimation of steraids at the ng/1 level. These advantages have recently led to a marked increase in the number of RIA determinations.

If more individual steraids have to be estimated in one determination (profile analysis), separation prior to detection is inevitable. For this purpose chromatographic methods are preferable. The successful separation of a limited number of steraids with liquid chromatography6_' 7 has been reported. However, the high resolving power of capillary gas chromatography8 _{is usually necessary to} separate all steraids to be estimated. The specificity of gas chromatographic determinations with capillary columns is usually equal to or better than that of RIA analyses, though its sensitivity is generally lower.

The gas chromatography of steraids is rather cumhersome owing to their low volatility and their sensitivity to thermal and catalytic decomposition. Between 1960 and 1970 this analysis was considerably improved, mainly through the preparation of suitable derivatives and columns. Gas chromatographic steroid profile analysis was first described by Gardiner and Horning9 _{in 1966. They}

urinary steraids in one single run on a packed column. The main advance in this technique since that time has been the introduetion of glass capillary columns10, which allow a better resolution and therefore a more specific

determination. The increasing number of recent

publications on this subject indicates the importance of this technique for endocrinology. Unfortunately, the significanee of gas chromatography for the analysis of steraids in serum or plasma is small because of the extremely low concentrations of most steroids. For the determination of plasma levels of individual steraids RIA offers a simpler and less time-consuming alternative.

Unambiguous identification of steroid metabolites in urine from gas chromatographic retentien data alone is generally impossible. The combination (capillary) gas chromatography-mass speetrometry (GC-MS) is at present the most powerful analytica! technique for the separation and identification of compoundsin complex mixtures11_{• The}

specificity of GC-MS exceeds that of RIA; when applied in the single or multiple detection mode, its sensitivity approaches that of RIA.

Ih studies of the gas chromatographic determination of steroid profiles from urine, reported in literature, only few investigations deal with genuinely quantitative

analyses. The significanee of steroid profi le analysis for diagnosis and biochemica! research, however, calls for a quantitative method. The investigations described in this

thesis aim at optimizing all the steps in the determination of steroid profiles by capillary gas chromatography (i .e. sample preparation, derivatisation and gas chromatographic analysis) . Applications of the developed methad to the diagnosis of endocrine disorders are discussed from profiles of various patients.

REPERENCES

2. 1'1. ZillUUerman, Z. Physiol. Chem. , 233 (1935} 257.

3.

e.c.

Porter and R.H. Silber, J. Biol. Chem., 185 (1950) 201.

4. S. Görög and G. Szasz, Analysis of Steroid Hormone Drugs, Elsevier, Amsterdam, 1978, p. 274.

5. G.E. Abraham, in E. Heftmann (Editor), Modern Methods of Steroid Analysis, Academie Press, New York, 1973,

p. 451.

6. P. Vestergaard, A. Bachman, T. Piti and M. Kohn, J. Chromatogr. , 111 (1975) 75.

7. G. Keravis, M. Lafosse and H.M. Durand, Chromatographia,

lQ

(1977) 678.

8. E. Heftmann, Chromatography of Steroids, Elsevier, Amsterdam, 1976.

9. W.L. Gardiner and E.C. Horning, Biochim. Biophys. Acta,

115 (1966) 524.

10. J.A. VÖllmin, Chromatographia,

l

(1970) 233.

CHAPTER 2

STEROIO HORMONES

2.1 GENERAL STRUCTURE AND STEREOISOMERISM

The steroids are a group of naturally occurring compounds which can be derived froro perhydro-1,2-cyclopentanophenan-trene, also called gonane1_' 2

• They include steroid hormones, sterols, bile acids, cardiac glycosides

(cardenolides and bufadienolides), spirostans and

furostans (formerly called sapogenins), steroid alkaloids and the vitamin D series.

27

Fig. 2.1 5a-Gonane and 5a-Cholestane.

The steroids considered here contain up to 27 carbon atoms, numbered as shown in fig. 2.1. The hydrogen atoms attached to carbon atoms are not written in the structure unless their configuration needs special attention. Each of the cyclohexane rings exists in either the boat or the chair conformation. The latter is thermodynamically more stable at room temperature; ~G = 6,9 Kcal/mole3 • In both confor-mations two types of C-H bonds can be distinguished:

equatorial (e) bonds, which form an angle of approximately 30° with the main plane of the ring system, and axial (a)

bonds, which lie more or less perpendicular to this plane.

Stereoisomerism is caused by different orientation in space of one ring to another. In naturally occurring steroids rings A/B can be fused cis or trans; rings B/C and C/D are always joined trans (see fig. 2.2).

.

e

cis _trans

Fig. 2.2 Cis and trans fusion of rings A and B.

~

.

If groups or atoms which are attached to the nucleus, lie above the plane of the ring system (i .e. on the sarne side as the angular methyl groups at C-10 and C-13), they are called B since the methyl group at C-13 was arbitrarily given the B-configuration. Groups lying below the plane of the ring system are called a. Conventionally the a -honds are indicated by interrupted lines and the B-bonds by continuous lines.

Fig. 2.2 shows that if rings A and B are fused trans, the

C-H bond at C-5 is orientated below the plane of the ring. Therefore this isomer is called the 5a-isomer. If rings A and B are joined cis, the C-H bond at C-5 lies above the plane of the ring; this structure is called the 58-isomer. Substituents at C-20 shown to the right of the carbon atom (using a Fisher projection) are termed a and those

shown to the left B·

Definitive rules for the nomenclature of steroids are described in extenso by the IUPAC4

• In spite of these

rules many trivi.al narnes are still widely used. Yet some of these trivial narnes are recognized by the IUPAC

commission. A cornprehensive list of trivial and systernatic narnes and abbreviations is given in appendix I.

2.2 CLASSIFICATION AND BIOLOGICAL ACTION

Horrnanes are defined as organic cornpounds produced by glands that have internal secretion (endocrine glands) . In man the most important endocrine glands are the

hypothalamus, pituitary, thyroid, pancreas, adrenals and gonads (ovaries. and testes) .

As to their chernical structure three different types of horrnanes can be distinguished: a) horrnanes derived frorn arnino acids by chernical reactions, ether than condensation b) peptides or proteins and c) steroid horrnones. Only the last type will be dealt with in this thesis.

Stercid horrnanes (henceforth called steroids) are rnainly secreted by the adrenal glands and gonads; during

pregnancy also by the foeto-placental unit. According to their biological action steraids are divided into four classes: androgens, oestrogens, progestagens and

corticosteroids.

2.2.1 Androgens

Androgens are derived frorn androstane (fig. 2.3), which, cornpared to gonane, has two additional methyl groups at C-10 and C-13. Sa-Dihydrotestosterone is the most potent andregen in the hurnan body, though testosterone,

androstenedione, DHEA and DHEA-S* manifest also important andregenie activity.

Androgens are secreted by the adrenal cortex (DHEA and DHEA-S) and the testes (androstenedione and testosterone) or the ovaries. Peripheral conversion of androstenedione into testasterene is of biologica! irnportance only in wamen. Androgens stimulate the differentiation of the male

* Capital G or S added to a steroid indicates the glucuronide or sulphate conjugate (see 2.3).

Fig. 2.3 5a-Androstane and 5a-Dihydrotestosterone.

reproductive track in utero. They are responsible for the development and rnainterrance of the male primary and

secondary sexual characteristics. Testosterone also promotes the spermategenesis in the seminiferous tubules of the testes. Another important action of testosterone

is the retentien of nitrogen, resulting in an increase in protein synthesis (anabolic effect) .

2.2.2 Oestrogens

All natural oestrogens (derived from oestrane) lack a methyl group at C-10 and possess an aromatic A-ring. The three most important oestrogens are oestradiol, oestrone and oestriol. Oestradiol is biologically by far the most active.

HO

Fig. 2.4 5a-Oestrane and Oestradiol.

In females with regular menses oestrogens are mainly secreted by the ovaries. During pregnancy the

foeto-placental unit is the most important souree of oestrogen production. To a smaller extent oestrogens are also produced by the adrenal cortex and by peripheral conver-sion from androgens (mainly of androstenedione into oestrone) . In post-menopausal woroen and rnales oestrogens originate almost entirely from conversion of androgens5

•

In rnales direct secretion of oestradiol by the testes is of minor importance. Oestrogens are very important for the maintenance of the female reproductive track and for the development and maintenance of the mammary glands.They are also involved in the control of the menstrual cycle. Most probably they play an important role in the transport and nidation of the fertilized ovum and in the proteetion of the foetus during pregnancy.

2.2.3 Progestagens

The chemical feature of the progestagens is the short side-chain at the C-17 position. The structure of 58 -pregnane and progesterone (the main representative of this class) is shown in fig. 2.5.

Fig. 2.5 56-Pregnane and Progesterone.

In females the corpus luteum is the main souree of

progesterone; it is also produced by the foeto-placental unit during pregnancy. It appears as intermediate in the biosynthesis of androgens, oestrogens and corticosteroids. Progesterone acts synergistically with oestradiol in

i.t produces the secretory hypertrophy of the endometrium suitable for nidation of the fertilized ovum. Progestagens are pre-eminently protectors of the pregnancy.

2.2.4 Corticosteroids

Unlike progestagens the characteristic structure of

biologically active corticosteroids is the CH₂ 0H-CHO-side-chain at C-17 in combination with the 3-oxo-4-ene

conjugated system. Corticosteroids are produced by the adrenal glands. They are divided into mineralocortico-steroids produced by the zona glomerulosa and gluco-corticosteroids produced by the zona fasciculata.

Mineralocorticosteroids are involved in the regulation of the water and electrolyte balance in the human body. They stimulate the sodium retentien and the potassium depletion. Aldosterone (fig. 2.6) is the most potent mineralocortico-steroid.

0 0

Fig. 2.6 A~dosterone and Cortiso~.

Cortisol (fig. 2.6} is biologically the most active glucocorticosteroid. In general corticosteroids are responsible for the maintenance of homeostasis; they

enable the body to cope with all the internal and external demands made on it. Therefore their actions are highly various. A prominent feature is the stimulation of the gluconeogenesis, i.e. the synthesis of carbohydrates

(mainly glucose) from phosphoenolpyruvate and amino acids. In the liver gluconeogenesis stimulates the glycogen

deposition, glucose production, amino acid uptake and protein synthesis. In extrahepatic tissue the glucose utilization and protein synthesis are reduced, whereas the protein breakdown is promoted. In adipose tissue the glucose utilization is reduced, the permissive role of glucocorticosteroids in the mobilization of lipids being enhanced. In normal humans some of these actions are counterbalanced by insulin. Glucocorticosteroids also exert a suppressive influence on inflammatory reactions

(infection, allergy), stress and shock. The precise action of glucocorticosteroids in suppressing the inflammatory and allergie responses is unknown. They increase capillary resistance and oppose the vasodilatory effect of histamine. Cortisol also manifests inhibitory effects upon the resorption of calcium from bone matrix in the intestine and the secretion of growth hormone by the pituitary. Most natural glucocorticosteroids also show some mineralocorticoid activity.

More details about the production and biological effects of all steraids mentioned above can be found in the literature6

- 10•

2.3 BIOSYNTHESIS AND METABOLISM

2.3.1 Biosynthesis

The precursor of all steraids in the human body is

cholesterol. The biosynth.esis of this compound starts wi th the formation of squalene from activated acetic acid, present as acetyl coenzyme A. The details of this pathway h.ave been reviewed by Clayton11

• The cyclization of

squalene via an epoxide intermediate into lanosterol is at present not fully understood. Lanosterol is converted into cholesterol via zymosterol (8,24-Sa-cholestadiene-38-ol) and desrnasterol (5,24-cholestadiene-3B-ol)12 •

The conversion of cholesterol into pregnenolone involves the oxidation of the parent compound to an enzyme-bound dihydroxyderivative (20a,22R-dihydroxycholesterol).

The side-chain is removed by the action of 20a,22-C-27 desmolase to yield pregnenolone and isocaproic acid. The major pathways in the biosynthesis of steroid horrnanes in the human body are shown in fig. 2.7.

Androgens are formed via two principal pathways:

1. pregnenolone + progesterone + 17a-hydroxyprogesterone + androstenedione

t

testasterene

2. pregnenolone + 17a-hydroxypregnenolone + DHEA + androstenedione

t

testasterene

Oestrogens originate from the conversion of androstene-dione via 19-oxoandrosteneandrostene-dione into oestrone, which is rapidly reduced to oestradiol. In humans these two oestrogens are in equilibrium. The pathway for the biosynthesis of corticosteroids passes either via

progesterone and 11-deoxycorticosterone to corticosterone and aldosterone or via 17a-hydroxyprogesterone and 11-deoxycortisol to yield cortisol and cortisone.

2.3.2 Metabolism

Although there are a great number of steroid metabolites, only a few enzymatic conversions are involved in their production. These may be simplified as follows:

1. Reduction of the double bond at C-4 into 5a- and 5~­ dihydroderivatives and subsequent reduction of the 3-oxogroup, resulting in the formation of four isoroers

(3a-OH-5a-H, 3S-OH-5a-H, 3a-OH-5S-H and 3S-OH-5S-H). In humàn hepatic tissue the reduction of the 3-oxogroup

leads predominantly to 3a-hydroxylated steroids. 2. Reduction of the 20-ketone function to 20a- and

20S-hydroxylated metabolites. It is mainly 20a-isomers that are formed by this reaction.

3. Generation of a 17-oxogroup either through oxidation of the 178-hydroxyl group in C-19 steraids or by cleaving off the C-21,20 side~chain of C-21 steraids containing a 17a-hydroxyl group.

HO CHOLESTEROL

(1)

.

yH3

f f

HO ~ PREONENOLONE

®

.

9H3

c~3rrtiH

HO~

17a- HYDROXYPREONENOLONE • CH O

~

DEHYDROEPIANDROSTERONE

9HJ

ff

PROOESTERONE

®t

9H

3

~:

17a- HYDROXYPROOESTERONE ANDROSTENEDIONE DHEA-S OESTRONE OESTRADIOL

Fig 2.7 Outline of the major pathways in the biosynthesis of steroid hormones.

9H20H C=O 4 3 0 11· DEOXYCORTICOSTERONE

®

t

yH20H C=O OH 3 ' 0 11· DEOXYCORTISOL CORTICOSTERONE @

I

HO 9H20H • I C=O 0

f f

CORT/SOL 18-HYDROXYCORT/COSTERONE • yH20H

ff:"

t

yH20H

~

CORT/SONE ALDOSTERONE

Key to.the enzymes

<D

20a,22-C-27 desmolase

@

17a-hydroxylase

G>

38-hydroxysteroid

@

21-hydroxylase

dehydrogenase

4. Introduetion of a hydroxyl group (possible at different positions).

Free steroids are poorly soluble in water because of their lipophilic structure. They are converted by liver and kidney enzymes into metabolites that are readily soluble in water, so as to eliminate them via the urine.

The final excretion products are usually conjugated as 6-glucosiduronides (usually termed glucuronides) or sulphate esters.

COOH

~0

HOOH OH

Fig. 2.8 Steroid glucuronide and sulphate ester.

The most important roetabolie pathways of the androgens are shown in fig. 2.9. The principal urinary roetabolie produets are androsterone and aetiocholanolone, derived mainly from

ahdrostenedione but partly also from DHEA and testosterone. About 80% of androsterone and nearly all aetiocholanolone are exereted as 3-glueuronides. Cireulating DHEA-S and most DHEA are excreted as DHEA-3-sulphate. In urine only l i t t le free DHEA or DHEA-3-glueuronide can be deteeted. Another roetabolie pathway of androstenedione is 116-hydroxylation

in the adrenal glands, which yields 116-hydroxyandroste ne-dione. This compound is converted into 116-hydroxy

-androsterone, 11-oxo- and 116-hydroxyaetiocholanolone. These metabolites are also (but to a smaller extent) derived from cortisol and eortisone.

The most important oestrogen metabolite is oestriol (from oestrone via 16a-hydroxylation) , mainly excreted as 16-glueuronide. Quantitatively less important products are

I

IV w I

~=

~g

_0~

_ff

DHEA ANDROSTENED/ONE TESTOSTERONE

I

I'

I

0 !

.

ó

ctsD

~

.

.lÓ

~-

:

i

_:

~-~

_{AETIOCHOLANOLONE} HO DHEA ANDROSTERONE I .--~---- - - _ _ _ ï _________ - - - - -! ____ --- ---··

HO~

.

~~

~~

HO.-~

HO-'q:y-

HO_.q:y-11{3 -OH-ANDROSTERONE 11 {!>-OH-AET/OCHOLANOLONE 1 1-0XO-AET/OCHOLANOLONE

I N

"""

I 11-0XO-AETIOCHOLANOLONE 11-0XO-ANDROSTERONE

-H~JÓ

HO

11 {>

'cr:r-

-oH-AETIOCHf:)lANOlONE Fig. 2.10 Principa~ metabo~ic pathway of cortiso~.

H~rA

HO

'

cr:r-

ft{l-OH-ANDROSTERONE

16-epioestriol (from oestrone via 16B-hydroxylation) and oestrone-3-sulphate.

The main metabolite of progesterone is 5S-pregnane-3a, 20a-diol-3-glucuronide, although also other prcgnanediols and some pregnanolones are excreted. 17a-Hydroxy

-progesterone is converted into 5B-pregnane-3a,l7a,20a-triol which appears in the urine as 3-glucuronide. Some

pregnanediolones and small amounts of androsterone and aetiocholanolone arealso roetabolie products from 17-a hydroxyprogesterone.

Fig. 2.10 shows that, after reduction of their 3-oxo-4-ene conjugated system, cortisol and cortisone are excreted to a large extent as their tetrahydroderivatives (THE, aTHE1 THF, aTHF). Further reduction of the C-20 ketene function also yields considerable amounts of cortols and cartalones

(20a- and 20B-hydroxysteroids). Corticosterone yields a mixture of THA, THB and aTHB, while 11-deoxycortisol

(compound S) is converted into THS. Less important is the transformation of cortisol and 21-deoxycortisol into 11 S-hydroxylated C-19 steraids and of cortisone into 11-oxo-aetiocholanolone. Less than 10% of the total production of these metabolites originates from adrenal conversion of cortisol and cortisone via this pathway. Corticosteroids are usually excreted as 3-glucuronides, although also 21 -sulphates and other types of conjugates have been found.

The main roetabolie product of aldosterone is 3a -hydroxy-58-tetrahydroaldosterone, excreted as 3-glucuronide.

2.4 REGULATION OF THE BIOSYNTHESIS

The production of glucocorticosteroids by the adrenals and of sex horrnanes by the gonads is controlled by the

hypothalamus and the pituitary, the two most important endocrine glands. The secretion of glucocorticosteroids by the adrenals is stimulated by ACTH (adrenocortico

-trophic horrnanel, secreted by the anterior lobe of the pituitary (adenohypophysis). The release of ACTH from the pituitary is not constant throughout the day, but

proceeds with pulses. Measurements of the plasma ACTH concentratien point to a circadian periodicity, paralleled by the concentratien of 17a-hydroxycorticosteroids in plasma. This leads to the conclusion that the release of corticosteroids from the adrenals is also subject to a circadian rhythm.

Two other adenohypophyseal hormones, FSH (follicle stimulating horrnanel and LH (luteinizing hormone) , stimulate the production of sex horrnanes by the testes and the ovaries. In females FSH and LH_promote the follicle development and the ovulation. They regulate, tagether with oestrogens and progesterone (formed by the corpus luteum) , the menstrual cycle in a very complex way. In rnales FSH stimulates the spermategenesis in the seminiferous tubules of the testes. LH promotes the

production of androgens in the Leydig cells of the testes. The release of these adenohypophyseal horrnanes is con-trolled by hypothalamic neurohormones, also denoted "releasing factors". ACTH-release is regulated by CRF

(= corticotrophin release factor) . Recent evidence8 suggests that FSH and LH are released by the same factor

(FSH-RF/LH-RF, also called GnRH = gonadotrophin release hormone). Yet the presence of a s t i l l unknown hormone that only or chiefly releases FSH, cannot be excluded.

The glucocorticosteroids and sex horrnanes have an inhibitory effect on the release of ACTH and FSH/LH respectively (negative feedback mechanism). Although the current view is that the main side of this feedback mechanism is in the hypothalamus, there may also be a direct feedback action on the anterior pituitary itself and perhaps other brain areas (extra-hypothalamic areas). The most important aspects of the regulation of the

GLUCOCORTI· COSTEROIDS EXTRA HYPO THALAMIC AREAS NEURAL INFLUENCES

!

HYPOTHALAMUS CRF lFSH·RF LH·RF PITUITARY ANDROGENS OESTROGENS PROGESTAGENS

I

GONADS

Fig. 2.11 Regu~ation of the steroid biosynthesis.

Details about the processes that are involved in the regulation of the biosynthesis of steraids at a molecular level, are given by Schulster et a~. 6_•

The production of aldosterone is controlled by a totally different mechanism: the renin-angiotensin-aldosterone system. The release of renin from the juxta glomerular cells of the kidney is a very complex phenomenon, to which several systems contribute13• Renin reacts with angiotensinogen from the liver to farm angiotensin I. This is transformed by a "converting enzyme" into

angiotensin II, which stimulates the adrenals to produce aldosterone. Other important stimuli for aldosterone

production areACTHand excess K+. Increased levels of Na+ and decreased levels of K+ inhibit the release of renin

(negative feedback).

2.5 MODE OF ACTION

i t is necessary to determine the biochemical processes that occur at a molecular level and that ul timately result in the observed physiological effects of the steroids. In vivo administration of a hormone results in the

accumulation of this hormone in target organs. 3 H-Oestra-diol injected in rats is found in high concentrations in the uterus and the vagina. This phenomenon can be

explained from the presence of molecules with a highly specific binding capacity (receptors) for this special hormone. The general scheme for the interaction of steraids with cells at a molecular level has been elucidated. Many details are s t i l l to be solved. The extensive work in this field has been reviewed by King and Mainwaring14•

Unconjugated steraids freely permeate into cells. Only in target organs they combine with highly specific receptor molecules in the cytoplasm of the cells to form a steroid -receptor complex. Probably each class of steraids (e.g. androgens, oestrogens) has its own specific receptor. All receptars have a protein structure. The steroid-receptor complex enters the nucleus where i t affects the specific transcription processes at the level of the gene15

•

Nowadays i t is generally accepted that this complex combines with the nuclear chromatin (DNA or the AP

3 fraction of the non-histones proteins) . How this inter-action results in the observed physiological effects of the steraids is unknown, except for the production of specific proteins. The combination of the steroid-receptor complex and the nuclear chromatin enables a molecule of RNA polymerase to occupy an initiation site of DNA. A segment of DNA is transcribed, producing a strand of a specific protein inside the cytoplasm. The production of this protein allows the growth and maintenance of the target organ.

2.6 REPERENCES

1. J.R. Hanson, Introduetion to Steroid Chemistry, Pergamon Press, Oxford, 1968, p. 1-19.

2. W. Klyne, The Chemistry of the Steroids, Methuen, London, 1965, p. 1-45.

3. J.B. Hendrickson, J. Am. Chem. Soa.,~ (1961) 4537. 4. IUPAC-IUB, Definitive RuZes for NamenaZature of

Steroids, Butterworth, London, 1972.

5. J. Poortman, GesZachtshormonen en Mamma-carcinoom,

Thesis, Utrecht, 1974.

6. M. Tausk, PharmaooZogie van de Hormonen, Agon Elsevier, Amsterdam, 1973.

7. W.R. Butt, Hormone Chemistry, Ellis Horwood, Chichester, 1976.

8. D. Schulster, S. Burstein and B.A. Cooke, Moleaular EndoorinoZogy of the Steroid Hormones, J. Wiley & Sans, London, 1976.

9. C.L. Cape, AdrenaZ Steraids and Disease, Pitman, London, 1972.

10. N.P. Christy (Editor), The Human AdrenaZ Cortex,

Harper and Row, New York, 1971.

11. R.B. Clayton, Q. Revs. Chem. Soa. , ~ (1965) 168. 12. H.R. Mahler and E.H. Cordes, BioZogieaZ Chemistry,

Harper and Row, New York, 1971, p. 738.

13. H.L. Vader, AnaZytioaZ Aspeots of the Plasma Renin

Aotivity and a Clinical AppZioation, Thesis, Utrecht, 1978, p. 5.

14. R.J.B. King and W.I.P. Mainwaring, Steroid-CeZZ Interactions, Butterworth, London, 1974.

15. B.W. O'Malley and W.T. Schrader, Sci. Am., 234 (1976) 32.

CHAPTER 3

QUANTITATIVE ASPECTS OF THE

SAMPLE PREPARATION PROCEDURE

3.1 INTRODUCTION

Direct determination of steroid profiles from urine with gas chromatography is impossible for several reasons: - insufficient volatility of steroid conjugates

- sensitivity of some steraids to thermal decomposition - the presence of large quantities of many non-steroidal

components, which interfere in the analysis despite the high resolving power of capillary gas chromatography.

The preparation of urine samples for steroid analysis depends on the methad selected to analyse these compounds. For gas chromatographic analysis the pretreatment of these

~amples consists of the following steps:

1. extraction of steroids and steroid conjugates from urine

2. hydrolysis, possibly followed by solvolysis, of the conjugates

3. further purification of the sample 4. formation of suitable derivatives

5. removal of excess derivatisation reagents.

If more detailed information is required, steroid conju -gates or steroids (liberated by hydrolysis/solvolysis) can be fractionated prior to analysis2- 6• For steroid profi le analysis these preseparations are generally omitted.

The derivatisation and the remaval of excess derivatisa-tion reagents will be discussed in chapter 4. In this chapter methods for the first three steps are selected after a survey of literature. Each of these steps is optimized separately with regard to accuracy ctnd repeatability. After this the overall accuracy and repeatability of the finally selected procedure are determined.

A common way to test methods for sample preparatien is the use of labelled steroids. Labelled steroid conjugates, however, must be preferred, but unfortunately only few of these are available.

For steroid profile analysis the accuracy of the sample preparatien procedure must be known for as many components as possible. Therefore we used endogenously labelled urine samples, obtained after administration of (a) labelled steroid(s) to healthy persons. As these labelled steraids take part in normal metabolism, all roetabelites of these compounds will appear in the urine as labelled conjugates. For this study urine samples were used covering the whole polarity range of steraids and all types of conjugates present in normal urine samples.

3.2 MATERIALS

Non-radioactive steraids were purchased from Steraloids (Pawling, N.Y., U.S.A.) and Ikapharm (Ramat Gan, Israel). Radioactive steroids, [4-1~C]-androstenedione (60 mCi/ mmolel, [4-14C]-dehydroepiandrosterone sulphate, ammonium salt (54 mCi/mmole), [1,2-3H]-cortisol ( 0,36 C/mmole),

[6,7-3_{H]-oestrone (44 Ci/mmole) and [7-}3

H]-dehydroepian-drosterone (16,6 Ci/mmole), ·were all purchased from the Radiochemical Centre (AmersHam, Great Britain) .

To prevent radiation damage as much as possible these sterotds were stared at 2°C, dissolved in benzene: ethanol

=

9 : 1 (v/v). These solutions were used only after a èhromatog~aphic purity check. If more than 1% impurities could be detected, they were purified with TLC or gel

filtration. XAD-2 was obtained from different suppliers;

only the material obtained from Serva (Servachrom XAD-2,

300-1000 ~m; Serva, Heidelberg, G.F.R.) could be

sufficiently purified6• HeZix pomatia juice was obtained

from Industrie Biologique Francaise (Gennevilliers,

France). The quality of methanol and ethanol was p.a.

grade (Merck, Darmstadt, G.F.R.); ethyl acetate was

naoograde (Byk-Mallinckrodt, Wesel, G.F.R.). All solvents were used without further purification.

3.3 PROCEDURES FOR SAMPLE PREPARATION

3.3.1 Extraction of steraids and steroid conjugates from

urine

For the extraction of free steraids from aqueous solutions

organic solvents of medium polarity (e.g. ether, methylene

chloride or ethyl acetate) are often used7• To extract the

more hydrophilic steroid conjugates more polar solvents

(butanol) or mixtures of these are needed. However,

solvent extraction procedures appear not very attractive for the extraction of steroid conjugates from urine

because of the following disadvantages:

~ moderate recovery of steraids and steroid conjugates

- large volumes of solvents are required

- cumbersome manipulative procedures

- occurrence of emulsions.

An entirely different methad to extract steroid conjugates

from urine, column adsorption on XAD-2, was introduced by

Bradlow8 in 1968. The XAD-2 resins are neutral copolymers

of styrene and divinylbenzene (crosslinking agent). By

changing the degree of crosslinking different types can be

prepared. The size of steroid (conjugate) molecules is most compatible with the average pore diameter of XAD-2;

a partiele size range of 300-1000 ~m is preferred to

extract steraids and their conjugates from urine. XAD-2

consists of small microspheres fused into macroreticular

microspheres is continuously porous9• Some properties of XAD-2 are given in table 3.110

•

Table 3.1 Physical properties of XAD-2.

Porosity (ml/ml) 2 Surface area (m /g)

Average pore diameter (nm) Solvent uptake (g/g) H 2

o

Me OH Et OH 0,42 300 9 0,072 0,699 0,719

The transport mechanism of molecules from the surrounding salution to the microsphere surfaces is described by Gustafson et al. 11

• The binding of organic molecules on XAD resin is an adsorption process. This was concluded from recent studies of adsorption isotherms, correlations with surface area and porosity, and thermadynamie

properties of the interactions9• The adsorption forces are mainly van der Waals forces i.e. the hydrophobic part of the molecule is adsorbed on the resin, while the

hydrophilic part is orientated to the aqueous phase12 • On the addition of strong polar solvents adsorbed compounds are desorbed.

After its initial description the XAD-2 method is

preferred by most investigators to the classical solvent extraction procedures. So far no systematic study has been undertaken to determine the recoveries of the

quantitatively most important steraids and steroid conjugates present in normal urine samples. To test

extraction by XAD-2 addition of only few labelled steraids and steroid conjugates to urine samples is commonly

applied8

, 13- 15• Only Bradlow8 has used endogenously labelled urine samples obtained after the administration of cortisol or testosterone. Reported recoveries generally exceeded 90%. These studies suggest the applicability of

this methad to most steraids and steroid conjugates present in urine samples from normal persons. Adsorption on XAD-2 is nat quantitative only for very polar steroids16 or steroid conjugates17

• Partly irreversible adsorption

has been reported for oestriol-16a-glucuronide18 •

3.3.2 Hydralysis and salvolysis of steroid conjugates

As indicated in sectien 3.1 steroid conjugates are nat suitable for gas chromatographic analysis. Cleavage of conjugates must therefore be considered an essential step in the sample preparatien procedure. Steraids are excreted in urine mainly as S-glucuronides and sulphate esters

(2.3.2). Other types of conjugates found in human urine samples, though in smaller quantities, are double

conjugates19_{, phosphates}20_{, and carboxylic acids}21 _{• To}

cleave S-glucuronides and sulphates two types of hydrolytic procedures are commonly applied: hot acid hydralysis or enzymatic hydrolysis, possibly followed by solvolysis.

Owing to the historical significanee hot acid hydralysis still retains its popularity. The advantages of this procedure are high speed and low cast, but a considerable drawback is the production of artefacts of nearly all steraids (especially corticosteroids). Studies by Goldzieher and Axelrod22 have shown that in many cases results can be incorrect by as much as 300%. Obviously hot acid hydralysis is completely unacceptable for the gas chromatographic determination of steroid profiles. Far more suitable are enzymatic hydrolytic procedures under mild conditions. Stercid 8-glucuronides are hydro-lysed by 8-glucuronidase (8-D-glucuronide glucuronoso-hydrolase E.C. 3.2.1.31). Sulphate esters are split by sulphatases or by chemical solvolytic methods23•

8-Glucuronidase can be obtained from different bacterial, mammalian and molluscan sources. In practice preparations

from the digestive gland of Helix pomatia (Helix pomatia

juice), from bovine liver (trade name: Ketodase) or E. Coli are frequently used to hydrolyse steroid glucuronides. These preparations differ in activity towards steroid. conjugates. This is most probably caused by contaminations in these only partly purified solutions. The respective roerits of the different enzyme preparations have been studied by many workers24

- 32• In his excellent review on

the hydralysis of steroid conjugates Bradlow19

_ states

that from the often conflicting data certain reasonably firm conclusions can be drawn:

1. There is no best souree of S-glucuronidase for the hydralysis of steroid glucuronides. If the proper enzyme concentration and time of incubation are chosen, all of the common enzyme preparations are capable of adequately hydrolysing steroid glucuronides.

2. There are marked differences in the rate of hydralysis by different preparations.

3. The rates of hydralysis of steroid glucuronides vary with the structure. Pregnane conjugates are most rapidly hydrolysed, followed by corticosteroid, 17-oxosteroid and oestrogen conjugates, in this order. Within each group the rate depends on the contiguration of the hydroxyl group and the adjacent ring junctions. 3-Hydroxyl-56-compounds are hydrolysed more rapidly than 3-hydroxyl-5a-compounds, and 3S-glucuronides faster than 3a-glucuronides.

The best souree for sulphatases capable to hydrolyse steroid sulphate esters is Helix pomatia juice, which contains arylsulphatase type II33 _{(arylsulphate}

sulphohydrolase E.C. 3.1.6.1). Type II indicates inhibition by phosphate and sulphate ions. Also 36-steroid sulphatase (sterolsulphate sulphohydrolase E.C. 3.1.6.2) and C-21-steroid sulphatase have

been positively identified in Helix pomatia juice. These enzymes hydrolyse all steraids sulphoconjugated at the C-3 or C-21 position except

3a-sulphate-5a-H-conjugates34

• C-17 and C-20 sulphate esters35, 36 are not hydrolysed either. Helix pomatia juice contains 100.000 Fishman units* S-glucuronidase and 800.000 Roy units** sulphatases per ml34

• The pH optimum for the enzymes of

Helix pomatia juice has been found to be very flat37•

Variations between pH= 4.5 and 5.5 had no influence on either the final amount of steraids released or the rate of hydrolysis. For good results to be obtained the buffer concentration should not exceed 0.2 M. Usually the enzyme concentration is 1-3% (v/v), whereas the temperature is maintained at 37-39°C for 18-48 hours32' 38• With concen-trations smaller than 1% (1000 Fishman units

S-glucuronidase/ml urine) a lower rate of release of all steraids has been observed37138

• In an extensive study

by Manson et al. 39

no artefacts of neutral steraids could be detected after incubation with Helix pomatia juice at 37°C for 24 hours (pH= 4.7).

As indicated above the sulphatases present in Helix

pomatia juice or other enzyme preparations are not able to

hydrolyse all sulphate conjugates. Therefore enzymatic hydralysis must be followed by solvolysis6

, 32, 36 ,40- 43 in the sample preparation procedure for gas chromatographic Steroid profile analysis. Since the metabolites that

cannot be hydrolysed with Helix pomatia juice are excreted in varying amounts13

, the omission of the salvolysis

cannot be corrected.

A generally accepted methad for salvolysis of steroid sulphate esters has been described by Burstein and Lieberman44145

• They showed that all sulphoconjugated 17-oxosteroids in urine were quantitatively solvolysed following this procedure. Sulphates of corticosteroids of

* A Fishman unit is the quantity of enzyme which releases

1 ~g of phenolphtalein in 1 hourat pH = 4.5 at 37°C.

** A R?y unit is the quantity of enzyme which releases

medium polarity, as present in the urine of normal adults,

can also be split by applying this method. The mechanism

of this salvolysis which proceeds in acidified ethyl acetate, is different from common solvolytic procedures. Detailed descriptions of this mechanism are given in

literature44' 45• When sulphates of very polar

cortico-steroids are excreted (human newborns) , other methods for

salvolysis are to be preferred16146_•

Only few papers deal with the determination of the

recovery of steraids after hydralysis with Helix pomatia juice and salvolysis in acidified ethyl acetate. In some studies a small number of steroid conjugates has been

added to urine priortothese steps13115138147_{• Only}

Bradlow8 _{and Setchell et aZ.} 6 _{have used endogenously}

label led urine samples. The recoveries obtained after hydralysis and salvolysis executed separately in or combination, generally exceeded 90%.

3.3.3 Purification of the sample

Because the organic phase (ethyl acetate) still contains

many interfering compounds from urine, but especially

from Helix pomatia juice, an additional purification

after salvolysis is inevitable. This purification is usually carried out by wash~ng ethyl acetate with aqueous

sodium bicarbonate (8% w/v) or sodium hydroxide (0.1-1.0 N). Aqueous sodium bicarbonate removes only organic acids,

whereas by sodium hydroxide also phenols are washed out.

After this "alkali wash".the sample is neutralised by washing with distilled water. A disadvantage of this

methad is that some h~gh.ly polar steraids are partly lost

in the aqueous phase36• In alternative procedures using

the anion exchanger Amberlyst A-26 in the bicarbonate form, th.is effect is even more serious 3 6

' 3 8 • The use of

diethylaminohydroxypropyl Seph.adex LH-20 (DEAP-LH-20)

seems to have eliminated this problem6_' 49_{, but} DEAP-LH-20 is not yet commercial ly available.

Procedures involving the use of aqueous sodium bicarbonate 6 ,ls,•J,so,sl or sodium hydroxide38148152 - 55 have often been described. Horning et al. 40_- 42 _{and Maume et al.} 56 wash with a solution containing 5% (w/v) sodium bi-carbonate and 10% (w/v) sodium chloride.

To our knowledge no investigations have been carried out to determine losses for a large number of steroids

present in normal urine samples.

3.3.4 Selection of a suitable procedure

Based on the arguments given in the precedin~ sections the following provisional procedure has been selected for the sample preparation:

1. extraction of steroids and steroid conjugates with XAD-2

2. hydrolysis of steroid glucuronides and sulphates with

Helix pomatia juice

3. solvolysis of non-hydrolysed sulphate conjugates in acidified ethyl acetate

4. alkali wash with aqueous sodium bicarbonate or sodium hydroxide.

The sequence:extraction by XAD-2 foliowed by hydrolysis needs some explanation, since these two steps are

frequently carried out in reverse order38140_- 42_, 5_1,57 _• Urine often contains compounds that inhibit the enzymes used for hydrolysis. 6-Glucuronidase is inhibited by

compounds conjugated with D-glucuronate or D-galacturonate and by D-saccharo-1,4-lacton23_{• The latter compound is} formed from D-glucose-saccharic acid owing to the low pH at which hydrolysis takes place58_- 61

• Sulphatases from Helix pomatia juice are inhibited by sulphate and pbos-pbate ions62- 64 • Essentially smaller quantities of steroids are released by enzymatic hydrolysis, if these inhibitors have not been removed previously. Sulphate and phosphate ions can be precipitated at pH = 11 with barium salts47• Saccharo-1,4-lacton and other high molecular

weight inhibitors can be removed successfully by gel filtration on Sephadex G-2560_{• All enzyme inhibitors,}

including sulphate and phosphate ions, are easily removed by XAD-265

'66• Graef et al. 65 have shown in a comparative

study that extraction by XAD-2 prior to enzymatic

hydralysis with Helix pomatia juice increases the amounts of steraids released in the hydralysis by 10-60%. High enzyme concentrations cannot be recommended to prevent inhibition, since the concentratien of inhibit.ors can be very high. Moreover samples are heavily contaminated with

components from Helix pomatia juice, which are only

partly removed by the alkali wash. For these reasans XAD-2 extraction prior to enzymatic hydralysis must be preferred.

Rademaker et al. 13 _{have determined the recovery of the} glucuronides and sulphates of androsterone, aetiocholano-lone and DHEA after a sample preparatien procedure

consisting of XAD-2 extraction, enzymatic hydralysis with Helix pomatia juice and salvolysis in acidified ethyl

_acetate. They obtained recoveries in the order of 80%.

Killpmann and Breuer43_{,~pplying the sameprocedure}

including an alkali wash_ with sodium bicarbonate, have recorded 95% recovery of DHEA-S and DHEA-G.

3.4 RESULTS AND DISCUSSION

3.4.1 General remarks

Urine samples

Urine samples (24 hours) were collected in polyethylene bottles. After their volumes had been recorded, they were stared at -20°C befare processing. Samples containing labelled steroid conjugates* were obtained from normal subjects, to whom had been administered: 3_{H-cortisol (1 ~C~}

* All endogenou~ly labelled urine samples were generously

supplied by Dr. J.H.H. Thijssen; Dept. of Endocrinology, Academie Hospital, State University of Utrecht.

orally), 3_H-oestrone/ 1 4_{C-androstenedione,} 3_H-DHEA/ 1 4 C-DHEA-S or 3H-DHEA/14C-androstenedione (30 ~Ci 3H and 15 ~Ci 14C, all by continuous infusion) .

Administration of steroids mainly results in the excretion of conjugated metabolites in the urine. About 90% of the radioactivity of administered 3_{H-cortisol is excreted in} the urine within 24 hours. The most important metabolites of cortisol67 are: THE (24%), THF (18%), aTHF (9%), a- and

B-cortolones (20%), a- and B-cortols (10%) and

11-oxygenated androgens (7%). After infusion with androstene-dione 65-85% oj the administered radioactivity is

recovered in the urine within 2 days68

• Androsterone, aetiocholanolone (both 15-30%) and testosterone (small amounts) are the principal metabolites. Oestrone (70% recovery within 3 or 4 days) is partly converted into oestradiol, oestriol and other oestrogens. Infusion with DHEA or DHEA-S (50-70% recovery of the radioactivity within 3 days) results in the excretion of DHEA, androsterone and aetiocholanolone.

Liquid scinti~~ation counting

Radioactivity was measured by counting 1 ml of the

samples in 11 ml of scintillation cocktail NE 262 (Nuclear E~terprises, Edinburgh, Gr~at Britain). Glass counting vials (from different sources) were used. 3_H-labelled samples have been counted in a Packard Tricarb 3320

(Packard, Brussel, Belgium) using the external standard channel ratio method for quench correction. The quench curve was recorded at least every two weeks. This curve was constructed by least square methods (Philips P 9200 time sharing system) . The maximum efficiency for 3_{H was}

36%. Dual-labelled (3_{H and} 14_{C) samples were counted in a}

Packard Tricarb 3380 or a Mark III (Searle Analytic, Des Plaines, I l l . , U.S.A.), both equipped with automatic quench, correction devices. Maximum counting efficiencies

for 3

H and 14C were 28% and 60% respectively. The

contribution of 3H to the 14C-channel was less than 0.1%; the 14C-activity in th.e 3H-channel was automatically

corrected. All samples were counted for 10 minutes or until 10.000 counts had been recorded.

Preparation of XAD-2 ao~umns

XAD-2 (Servachrom XAD-2 p.A., partiele size 300-1000 ~m)

was purified according to Setchell et a~. 6 _{prior to use.} This purification requires washings with 2 N sodium hydroxide, water, 2 N hydrochloric acid, water, acetone, ethanol and water. Fines were removed by decantation. XAD-2 was poured into columns fitted with a 25 ml

reservoir and a PTFE stopcock. These columns were washed with large quantities of distilled water before use. The purity of the resin was checked by eluting the columns with 40 ml of methanol, which was evaporated under

nitrogen. The dry residue was derivatised according to the procedure described in chapter 4. No interfering compounds could be detected by gas chromatography applying the same conditions as for steroid profile analysis.

3.4.2 XAD-2 extraction

èxperimenta~

To extract steraids and steroid conjugates from urine two types of columns of different dimensions were used. At first experiments were performed with columns of 10 cm lengthand 1.5 cm internal diameter (I.O.) containing 16 g of XAD-2. In later experiments the I.O. was reduced to

1.0 cm; these columns contained only 8 g of resi~. The extraction was carried out by applying urine (20 or 10 ml respectively) to the top of the column. Large molecules are not adsorbed because they cannot penetrate into the pores. After this the column was washed wi th 30 (20) ml of distilled water to remave inorganic salts and organic compounds of low molecular weight. In this.way the sample is thoroughly purified. After the column had drained completely, steraids and steroid conjugates were eluted with a polar organic sólvent. The flow-rate was kept constant (0.5-1.0 ml/min) throughout the procedure.

Selection of the solvent

Befare the experiments were carried out, i t was tested whether statistically different recoveries are obtained with methanol or ethanol as eluent. For this purpose 10 ml portions of a urine obtained from a normal person to whom

3_{H-cortisol had been administered, were extracted on the}

1.0 cm I.D. columns. Four columns were eluted with 40 ml of each eluent. The eluate was collected in 5 ml fractions

in which the radioactivity was determined separately. In the aqueous effluent less than 2% of the radioactivity

could be detected. The curves shown in fig. 3.1 represent

the average recovery obtained on these four columns for

each eluent. Methanol yielded a higher recovery (lOl vs 93%) . As can be concluded from the ethanol-curve continued elution will probably also yield a quantitative recovery. This observation is supported by Setchell et al. 6

, who

elute with 70 ml of ethanol (10 ml of urine as starting volume). In a second experiment, in which the recovery

was determined without collecting fractions, recoveries for methanol and ethanol amounted to lOl

+

4 and 95

+

4%

MElHANOL 100 RECOVERY

%

0 10 20 30 40

Fig. 3.1 Recovery of metabolites of cortisol from XAD-2 columns with methanol and ethanol as eluent.

respectively. Methanol was chosen as the eluent for further extractions because smaller amounts can be used. An additional advantage of methanol is its higher

volatility, reducing the time of evaporation after XAD-2 extraction. Fig. 3.1 also shows that 98% of the total reeavered activity is obtained in the first 20 ml of

methanol. However, to ensure that all steraids and steroid conjugates are eluted from the column, even if large

amounts of these components are present in the sample, i t was decided to use 40 ml of eluent.

]etermination of the accuracy and reproducibility

For the determination of the accuracy and reproducibility of the XAD-2 extraction most experiments were carried out with urine samples of normal subjects to whom 3_H-cortisol had been administered. Samples of eight different subjects were used; 14 series were processed in two different

laboratories by three technicians. No differences could be detected between the results obtained from either

different urine samples or in different laboratories. Table 3.2* shows that the results obtained for the different types of columns are identical. Only a few per cent of the labelled metabolites could be detected in the aqueous effluent. Since Bradlow8 _{has reported small} losses for different steraids added to urine, these losses are probably nat caused by specific non-adsorption of some individual metabolites. The tötal radioactivity

reeavered from the columns (losses in the aqueous phase plus recovery in methanol) generally equalled 100%. Consequently no evidence for irreversible adsorption was obtained. The XAD-2 extraction of these conjugated

corticosteroid metabolites resulted in an average recovery of 95 ~ 5% (n = 81). Identical values were obtained for the metabolites of DHEA, DHEA-S, androstenedione and

* In all the tables presented in this chapter the recovery or the accuracy is given as a percentage; the

repeatability (reproducibility) is expressed as the standard deviation. n

=

number of experiments.

Table 3.2 Accuracy and reproducibility of the XAD-2 extraction for urine samples of subjects to whom 3H-cortisol had been administered.

urine sample 1 2 3 4 5 6 7 8 series 1 2 3 4 5 6 1-6* 7 8 9 10 11 12 13 14 7-14** 1-14 x 98 94 96 89 98 92 94 96 87 89 96 95 99 96 98 95 95

s.o.

3 3 7 4 2 5 5 3 2 6 6 3 2 0 4 5 5

* Co~umn: 10 * 1.5cm; 16 g of XAD-2; 20 mZ of u~ine;

30 m~ of distilled wate~ and 50 ml of methanol. Expe~iments we~e ca~ried out in labo~atory A. ** Column: 10 * 1 cm; 8 g of XAD-2; 10 ml of urine;

30 ml of distilled water and 40 ml of methanol. Expe~iments were perfo~med in laboratory B.

n 5 4 4 5 5 5 28 5 5 8 8 8 8 3 8 53 81