Gas-liquid chromatography of steroids with glass capillary
columns : a breakthrough
Citation for published version (APA):
Luijten, J. A. (1973). Gas-liquid chromatography of steroids with glass capillary columns : a breakthrough.
Technische Hogeschool Eindhoven. https://doi.org/10.6100/IR58789
DOI:
10.6100/IR58789
Document status and date:
Published: 01/01/1973
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
providing details and we will investigate your claim.
CHROMATOGRAPHY OF
STEROIDS
WITH GLASS
CAPILLARY COLUMNS:
A BREAKTHROUGH
PROEFSCHRIFT
TER VERKRIJGING VAN DE GRAAD VAN DOCTOR IN DE TECHNISCHE WETENSCHAPPEN AAN DE TECHNISCHE HOGESCHOOL EINDHOVEN, OP GEZAG VAN DE RECTOR MAGN1FICUS. PROF, DR, lR. G, VOSSERS, VOOR EEN COMMISSIE AANGEWEZEN DOOR RET COLLEGE vAN DEKANEN IN BET OPEN-lIAAR 'fE VERDEDIGEN OP VRIJDAG 11 MEl 1973
TE 16.00 UUR
DOOR
JOHANNES ADRIANUS LUYTEN
GEBOREN TE ETTEN-LEUR1973
Prof.Dr.Ir. A.I.M. Keulemans. Promotor
Dr. B.J. Degenhart, Coreferent.
CONTENTS
1. XNTRoDUCTION
2. STEROIDS
2.1 General structure and stereoisomerism 2.2 Different groups Of steroids
2.3 HOrmones 2.3.1 2.3.2 2.3.3 2.3.4 l>:strogens Androgens I?rogestagens Corticosteroids 7 13 13 15 16 16 17 18 19
2.4 NOmenolature of steroid hormones 20
2.5 Biosynthesis and metabol~sm in the human body 21 2.6 Why i~ is meaningfUl to analyse steroids 27
2.7 ReferenCes 31
3.
CURRENT FROCEDURES IN STEROID ANALYSIS 323.1 General remarks 32
3.2 Classical deteotion and quantification methoas 34
3.2.1 The Kober reaction for estrogens 34
3.2.2 The Zimmermann reaotion for 17-ketosteroids 35 3.2.3 Reactions on the side chain of c-21 steroids 36 3.2.4 Speotrophotometric and fluorornetrio
estima-tions 39
3.3 The Use of raoiOisotopes 40
3.4 ~5says oased on steroid-protein interaction 42
3.4.1 Terminology and theoretical aspects 42 3.4.2 Methodological considerations in competitive
protein binding analysis 44
3.5 The role of chromatography in steroid separations 46
3.5.1 Chromatographic systems 47
3.5.2 Column chromatography 49
3.5.4
3.5.5 3.5.6
Thin-Layer chromatography
Gel filtration and gel permeation Ion exchange chromatography 3.6 Additional identification techniques 3.7 References
4. CAPILLARY COLUMNS IN THE GLC OF STEROIDS 4.1 Historical background
4.2 Packed versus capillary columns 4.3 Injection systems
4.4 An all-glass solid-state injection device 4.5 Qualitative and quantitative evaluation 4.6 Column preparation
4.6.1 Modification of the glass surfaoe 4.6.2 The intermediate test
4.6.3 Coating procedures 4.7 Description of apparatus 4.8 Column characteristics 4.9 Referenoes 52 53 54 54 55 59 59 61
63
66 67 73 74 75 76 79 80 89S. DERIVATIZATION METHODS FOR STEROIDS; RETENTION PARAMETERS 91 5.1 Introduction
5.2 The choice of a suitable derivative
91
93
5.3 Reactions with silylating agents 94
5.4 Stabilization of the dihydroxyacetone side chain 97
5.5 Methoximes and mixed derivatives 99
5.6 Parameters for steroid identification 103
5'.6.1 5.6.2 5.6.3
Relative retention times 103
Ret@ntion indices 106
Methylene units and comparison with reten-ion indices
5.7 ~ersilylation of steroids 5.8 Other identification m~thods 5.!J References
112
H8 119
6. APPLICATION TO URINARY SAMPLES 123
6.1 Introduction 123
6.2 Pretreatment of urinary samples 125 6.3 Isolation of steroids with Amberlite XAD-2 126
6.4 Quantitative aspects of the sample cleaning
up.
127 6.5 The estimation of estriol and pregnanediol inpregnancy urine
6.5.1 The feto-placental unit 6.5.2 Results
6.6
Steroids in the perinatal period130
130
111
133
6.6.1 Biochemical aspects 133
6.6.2 GLC of steroids in the urine of newborn humans
6.7 Urinary steroid profiles
6.7. 1 E:xampl.es
6.8 References
7. GAS OIROMATOGRAPH¥-MASS SPECTROMETRy OF STEROIDS
134 140 141 145 147 7.1 Introduction 147
7.2 Connection with capillary columns 148
7.3 Experiment.al conditions 150
7.4 Performance of the cOmbination .51 7.5 :fragmentation of steroid MO-TMS derivatives 155
7.6 Application to the profiling technique 157
7.7 Auxiliary techniques with the combined instrument 165
7.8 References 167 Appendix 168 Summary 170 Samenvatting 173 ACKnowledgement 176 Levensbericht 176
Chapter 1
INTRODUCTION
Since the early successful ~n~lys~s procedures for
ste-ro~dhormones have been reported, these methods have not directly formed part of the routine procedures within the scope of the clinical chemistry laboratory. An import~nt reason for this has been the great technical difficulty of measuring the often minute cOncentrationS of steroids usually present in blood and urine. As the study of steroid endocri-nology and biochemistry has been highly dependent on the development of analytical techniques, the reason of incom~ plete knowledge about synthesis and metabolism of steroids in the past, is understandable.
In the past 10-15 years much knowledge has been gained about the precise sites and pathways of steroid metabolism, to a large extent due to the great number of methods for the detection and quantification of steroids that has been evolved. AmOng these, the methods based on gas-liquid chro-matography (GLC) have contributed to an invaluable degree. Gradually more information has been obtained concerning the functiOn of steroid-produoing glands ( adrenal, testis, ovary, and feto-placental unit ) and the interaction of the hypothalamic-hypophysal system with these organs. Estimations of steroids in urine and pl~sma underlie this information. Therefore evaluation of adrenal, gonadal and placental function by steroid analysiS is possible at present and consequently the clinical chemist is approached with requests for steroid estimations. Well-known examples are the assess-ment of the feto-placental state by means of urinary estriol excretion, the analySiS of the group of l7-ketost@roids, 17-ketogenic steroids etc.
The ~nalysis of characteristic steroid metabolites in cases of disorders in secretion and metabolism of steroids constitutes a more complex problem. Although several cl@ar-cut disorders are known, in SOme cases still a question mark has to be put.
Several factors affect the utility of methods for the tracing out of the different forms of diseases involving steroid metabolism in the routine clinical chemistry labo-ratory. Often the bOllndary between research and routine work is a diffuse one and it must be left to the clinical chemist to define those sorts of measurements which come nearest to providing really diagnostic aid to the clinician. In certain instances steroid measurements provide strong corroborative evidence for hormonal diseases. Conclusive and critical evidence can usually be obtained if. the measurements form part of a larger investigation. Also ste-roid estimations will probably lead to new discoveries in steroid metabolism.
Amonqst the numerous applications in the field of analy-tical chemistry, Chromatography represents one of the greatest technical advances in biochemical endocrinology of the past two decades. In its many ramifications j,t has been applied successfully to the isolation, purification, partial identification and quantification of many steroids. Compared to paper- or thin-layer chromatographic procedures, gas cnromatography has been rather slow in penetrating clinical chemistry. So, at present only few hospital laboratories use GLC methods for routine steroid determinations. CJ.assical colorimetric and spectrophotometric methods are considered to yield data sufficiently accurate fOr diagnostiC purposes in many instances. Recently, the rapid development of com-petetive protein binding and radioimmunoassay methods have attracted many clinical investigators, as these methods offer possibilities for the application of steroid assays
on a large scale. While several classical detection ana
quant~fication methoas only infor~ us about the total amount of a group of structurally related steroids, methods based on steroid protein interaction are dev~sed in ~rinciple
for the estimation of indiv~dual steroids. Moreover, attention must be paid to specificity and blank values while only a limited number of steroids can be analyzed with these modern teohniques.
GLC has proved to be a useful tool for steroid analysis. Since the first GLC analyses of steroids with pac~ed columns have been reported ( around 1960 ), the analysis of these substances has been benefited by progress made in the field of column technology and detector systems. Moreover, develop-ments in sample pretreatment methods, including derivative formation_ have enhanced the reliability of gas chromato-graphic steroid estimations in various samples.
Since changes in the amount of a single steroid occurring in a biological sample may be of less significance than changes in the relative amounts of several Chemically related substances, analysis procedures provid~ng qualitative and quantitative information On a large nu~ber Of compon@nts in complex mixtures is welcome. A gas chromatograph is the in-strwnent "par excell@nce" capable of producing this infor-mation. For obvious reasons the number of compounds that can be analy~ed is directly related to the resolVing power ot the chromatographic colwnn. Often people neglect the capacities of a gas Chromatograph as a separating device, On using the apparatus for the estimation of a single steroid obtained from a biologioal extract after meticulous purification and subfractionating steps. Although this approach sometimes increases the specificity of a method also losses of steroids, chemical transformations On adsorbent layers and contami-nations fro~ solvents and chemicals may occur. The necessity of a far pushed sample purification is less stringent when
disposing of a high-resolution 9as Chromatograph. This high r€solution is attainable with capillary columns.
Although the introauction of capillary columns approxi~ mat@ly coincidsd with the early successful applications of GLC in the field of steroids, up t i l l now the use of capil~ lary columns fOr steroias has only been demonstrated inci-dentally. To a large extent this fact can be imputed to ~roblems related with,
- the preparation of durable capillary columns coated with thermostable stationary phases.
- the lack of suitable sampling techniques on capillary columns for high boiling compounds like steroids,
Bere we meet the first stage of our work in which the tech-nical aspects of the work with capillary columns made of glass are emphasized. Procedures will be given for the pre-paration of capillary columns, including several ways ot glass deactivation and alternative coating techniques. Different column parameters will illustrate the performance of the obtained columns.
Additional technical aspects of the use of capillary columns recieve our attention, a special place being oQcupied by the injection technique.
The applicability of this column type for steroid analysis will be shown from chromatograms giving the sepa-ration of standard steroids and steroids in pregnancy urine, in the urine of newborns and in the urine of adults
( profiling technique ). In this context the sample pre-treatment prior to the chromatographic step is envisaqed, with emphasis on derivative formation.
Especially when dealing with complex mixtures, the iden-tification of the separated components requires sp~cial attention. In this respect parameters based upon the
chromato-graphic retention ( relative retention, retention ~ndex, MU value ) are of great value for the assignment of the
stero~d structure. However an alternative for the identifi-cation ~is availa~le and in some cases even indispensable.
parallel to GLC, mass spectrometry ( MS ) has become a valuable technique in analytical chemistry. While the former, strictly speaking is a separat~on technique, the latter provides unique structural information on organic molecules. These teChniques have two common characteristics, namely the requirement of a vaporized sample and the high senSiti-vity. Following Successful pioneering work on coupling a gas Chromatograph with a mass spectrometer, nowadays GC-MS
has become an ~mportant tool for the analytical character-ization of organic compounds. particularly in the study of natural products, where one may encounter complex mi~tures containing trace amounts difficult to isolate and ~dentify, GC-MS can effectively be used.
In the technique of GC-MS the availability of a good gas
chrornatograph~c resolution is important. If the resolution of the column is poor, there will be rni~ed fractions enter-ing the ion source, producenter-ing complex mass spectra, which are difficult to interpret. On using capillary columns, purer fractions are supplied for an~lysis to the mass spectrometer.
For studies wh~re large numbers of ~amples have to be
analyzed, manual 5cannin9 and evaluation of all maSS spectra becomes unfeasible. Since high capacity and speed in the proo@ssing of data is deSirable, the full exploitation of GC-MS as a routine method in qualitative and quantitative analysis absolutely requires computer handling of data. We w~ll corne in touch with this subject on studying total urinary steroid profile5 with the Combination cap~llary column-mass spectrometer-computer.
It has been suggested in the past that the development· of instrumental methods, often implying a rather expensive equipment, has led to an increased preSSUre upon the clini-cal chemist to employ more and more relatively difficult techniques in the laboratory. However, for several reasons classical steroid estimations will not be abandoned for the time being, furthermore the potentialities of a GC-MS
instrument surpass by far the possibilities ot most clinical laboratories. Nevertheless progress on analytical methods made in research laboratories as the result of a close cooperation with people directly connected with the clinic, will lead to new and improved techniques for the measure-ments of biological active compounds applicable on a larger scale in the hospital. The present thesis is deemed to support that view for the group of steroids, but a similar approach applies to the GLC analysis of organic acids, biological amines, pharmaceutiCs etc.
Chapter 2
STEROIDS
2.1 General structure and stereoisomerism
The collect.ive name "steroids" comprises an extensive group ot organic compounds all having a structure that Can be derived from perhydro-l,2-cyclopentano phenanthrene, also called gonane.
The oarbon atoms or the ring system are numbered as shown in fig. 2.1. H 26 12 P 2 3 HO Fig. 2.2 Cholesterol.
Some st.eroids have a side chain attached to the c-17 carbon atom. For example cholesterol, the principal sterol of the animal organism, has an eight carbon atoms long side ohain and, as most common steroids, also two angular methyl groups at C-13 and C-lO. The carbon atoms ot these two me-thyl groups are numbered C-18 and C-19. See fig. 2.2.
As is well known, cyclahexane can exhibit two conrorma-tions, the chair and the boat form, the former being ther-modynamically more stable under normal conditions. The oar-bonhydrogen bonds of cyclohexane can be divided in equato-rial bonds, situated close to the plane of the rins, and axial bonds which lie about perpendicular to the main plane of the ring. See"fig. 2.3.
e
Fig. ~.3 Conformations of cyc1ohexane.
Stereoisomerism of ~teroids occurs because the cyclo-hexane rings can be joined in different ways. In natural steroids rings A and B can be fused both cis and trans. Rings E and C and rings C and D hOwever always are in the trans configuration for common steroids. As deptcted in fig. 2.4 the carbon-hydrogen bond at position C-5 is in-dicated by a full line in the case of cis configuration of A and B (Sa-position) and a dotted line reflects the trans configuration (5a-position).
A-B cis
~
~
H A-B transeXt
I I HRetrosteroids constitute a grou~, different from the natural steroids by a cis jOint of rings Band C.
Such a cOntrast between normal and retrosteroids may be significant for explaining the difference in biological activity.
A second form of stereoisomerism is caused by the way in which groups attached to the ring are oriented. The methyl groups at C-I0 and C-13 are on the same side of the nuclear plane in the case of all natural steroids and they are simply deSignated by solid lines, thus
im-plying ~-configuration. Substituents may be on the same side or On the opposite side of the nucleus in relation to the C-IO and C-13 methyl groups. Groups on the same side as these two methyl groups are said to possess a
~-configuration (indicated by a solid line); 5ubstituents on the opposite side have an ~-configuration (indicated by a broken line).
2.2 Ditferent groups of steroids
Although the term "steroids" often is interpreted as steroidhormones, some other important classes of compounds form part ot the steroids (1):
The sterols: A series of compounds structurally Similar to cholesterol which compound is present in almost all living organisms.
~he bile acids: To amino acids coupled compounds that play an important role in the fat metabolism.
Heartglycosides, sapogenins and steroid alkaloids: All gly-cosides exerting a specific physiological action.
of the about 10,000 known steroids, some 200 occur in the human body where some of them are inVOlved in the regu-lation of several body processes. The above mentioned types of compounds, characterized by a common base structure will not be considered in the further course of this thesis. Centrally are the so-called steroidhormones.
2.3 BormOnes
Hormones are substances, produced oy oertain cell struc-tures, transported to tissues where, sometimes after meta-bolic conversion, an interaction with maoromolecules can re-sult in an physiological response,
Many prOCesses in the body are regulated by hormOneS. We may mention here: growth, carbohydrate metabolism, sexual development, electrolyte balance and nitrogen balance. Ex-amples of hormones with a nonsteroidal skeleton are: Pituitary hormones, ACTH, LH and TSH. These compounds, pro-teins or polypeptides, mainly control the secretion of other endocrine glands ("feed-back" mechanisms).
Thyroxine: An iodine containing amino acid derivative produced by the thyrQid gland, which inter alia influences growth. Adrenaline: A catecholamine largely produced in the adrenal medulla, possessing many biological activities such as the conversion of liver glycogen to glucose, the release of hOrmones from the pituitary gland and the constriction of the arteries. These groups ot hormones, though interre-lated to the sterOid hormones, are not the subjeot of this investigation.
Important hormonally active steroids are seoreted by the adrenal gland, the testis, the ovary and the foeto-placental unit.
With r@gard to the number of oarbon atoms a subdivision of steroid hormOnes (henceforth oalled steroids) can be made.
We distinguish:
estrogens C-18 steroids
androgens C-19 steroids
progestagens and oorticosteroids: C-2l steroids
2.3.1 Estrog@ns
Estrane is the parent hydrocarbon from which the estro-gens are derived. Estroestro-gens differ in a characteristic way
from all other steroide by the presence, of an aromatic A ring and the lack of a methyl group at C-IO.
A representative of this series is estradiol, the most potent estrogen in the human body (fig. 2.5).
Fig. 2.5 5a-Estrane and ~~tradiol.
The estrogens, produced by the ovary, the placenta and to a smaller degree ~y the adrenal cortex, in combination with other hormones, are amongst other things responsible for the development and maintainance of the female sexual organs.
2.3.2 Androgens
The structural unit for androgens is androstane (fig. 2.6). Important androgens are: testosterone, androsterone and dehy-droepiandrosterone. Testosterone, with a conjugated keto func-tion in ring A, has the most important androgenic activity in the human body.
Apart from determining the sexual development of the male, androgens exhibit anabolic activity: they promote the nitro-gen retention by increasing protein synthesis and by decrea-sing the rate of amino acid catabolism. The gonads, under in-fluence of gonadotropin and the adrenals under inin-fluence of corticotropin are the sources of Qndrogens. Testosterone also occurs in females, however the concentration is much lower than in males.
2.3.3 Progestagens
Progestagens have a s~de chain of two carbon atoms (num-bered C-20 and C-21l at C-17, see fig. 2.7.
Fig. 2.7 56-Pregnane and Progesterone.
The main steroid in the series of progestagens is pro-gesterone. Progesterone, produced by the corpus luteum of the ovary, creates a favourable medium for the implantation of the fertilized ovum. OUring pregnancy progesterone, se-creted by the placenta, together with the estrogens, pro-tects the development of the fOetus. At present i t is known that also the adrenal cortex and the testis are natural sources of progesterone.
2.3.4 corticosteroids
Corticosteroids also have the pregnane nucleus. Contrary to most progestagens corticosteroids have a Keto or hydroxyl group at C~ll and a hydroxyl function at C-2.1.
The cortioosteroids are produced in the adrenals. usually a division is made between mineralocorticosteroids, secreted by the zona glomerulosa and glucOcorticosteroids that originate from the ~ona fasciculata of the adrenal.
Aldosterone (fig. 2.8), with an aldehyde function at C-18, is the most potent of the mineralocorticosteroids in man. This group of steroids COntrols the balance of salt and water: the sodium retention and the potassium excre-tion.
Fig. 2.S Aldosterone and Cortisol.
The glucocorticosteroids promote the gluconeogenesis, i.e. they stimulate the conversion of proteins to carbohy-drates and fats by the liver. Other effects of cortisol
(fig. 2.8), being the most potent glucocorticosteroid in the human body, are the anti-inflammatory activity, the role i t plays in resistance to shock and infection and its de-pressing action on the permeability of various membranes. Though less powerful than the mineralocorticosteroids cor-tisol can retain sodium, while stimulating the excretion of potaSSium.
Though many empirioal d~ta are known on the biologioal effeots of hormones, whether they have a steroid struoture or not, the real mechanism of their action is not yet com-pletely elucidated.
The reason why certain tissues are sensitive to the action of a hormone, and other ones are not, is still un-certain. The question how the interaction between hormone and tissue results in the final physiological effect is not yet satisfactory answered.
It has been proved that tissues, sensitive to estrogens or androgens contain sp@cial cellular structures with an outstanoing affinity for One of these types of hormones. These structures function as hormone-receptors. Hormone re-ceptors appear to occur in several eel membranes and most probably they have a protein structure. There are indications that the hormone-protein complex influences the formatiOn of secondary messengers which in their turn give rise to the physiological or morphological effect of the hormone.
2.4 Nomenclature of steroid hormones
The nomenclatUre of steroids obeys several rUles which will only briefly be given here. A comprehensJ.ve account of all stereochemical features of steroids is given by Fieser and Fieser (2), by Klyne (3) and by the IUPAC (5).
The suffix "ane" indicates a fully saturated compound "ene" pOints out the J?resenoe of a double bond, "diene" two double bonds and so on. Often the symbol d was used to in-dicate a double bond. the lOcation of a double bond is given by the lower numbe~ ot the two carbon atoms. If the double bond does not connect two successively numbered carbon atoms, this is explicitly indicated in the name.
Considering the position of the hydrogen atom at C-5 one formerly distinguished between the "normal" (56) and "allo"
position (5~). So 5~-pregnane in fact should no longer be called alloprsgnane. A configurat~on different from a refe-rence steroid at any other carbon atom may be indicatsd by the prefix "epi": androsterone (3a-hydroxy) and 3-epiandro-sterone or (3S-hydroxy).
Hydroxyl groups are indicated by the suffix "01, "dial" etc. or l;Jy the prefix "hydroxy". Ketones have the suffix ·ons", "dione" etc. or the prefix 'oxo", "dioxo" and 50 on.
Other prefixes are: "dehydro", indic;;J.tin<;r the elimina-tion of two hydrogen atoms and "d1hydro", suggesting the presence of two extra hydrogen atoms. "D@soxy" and "desoxo" are used for describing the 1055 of a hydroxyl group and keto group respectively.
The pre!ix "nor" relates to the elimination of a me-thyl group or meme-thylene group from a sidechain e.g. 19-Nor-testosterone.
Examples (see the figures 2.5, 2.6, 2.7 and 2.8) Estradiol-17e l,3,5(10)-Estratriene-3, 176-diol Testosterone
Progesterone Cortisol
176-Hydroxy-4-androsten-3-one 4-Pregnene-3,20-dione
lIB, 17a,
21-Trihydroxy-4-pregnen-3,20-dione.
An extensive list of systematic and trivial nawes of steroids is given in the Appendix.
2.5 Biosynthesis and steroid metabolism in the human body The synthesis of steroids has its starting pOint with the acetylcoenzyme A that is transformed by various enzyme systems vi~ mevalonic ~cid and pyrophosphates to squalene.
Squ~lene, consisting of isoprene units, undergoes cycl1~ation and oxidation to lanosterol (fig. 2.9).
In lanostero~ the double bonds are shifted and three me-thyl groups are stepWise converted and eliminated to form Cholesterol. cholesterol can be considered as the compound
HO
Fig. 2.9 Squalene and Lano8te~ol.
from which all steroidhormones can be deriv~d. Up to now the details of the reactions that take place in the biosynthesis of cholesterol are still partly unknown.
Cholesterol is converted to pregnenolone, the latter be-ing the ~recursor of progesterone. Progesterone is a central
interm~diate metabolite of the biosynthetic pathways found in the gonads and the adrenal cortex.
Fig. 2.10 shows the major pathways for the biosynthesis of several hormonally active free steroids.
From 17~-hydro~yprogesterone C-I9 androgens and C-l$ estrogens are produced. In the adrenals progesterone, via l7~-hydroxyprogesterone or II-desoxycorticosterone, is
metabolized to several corticosteroids like cortisol (com-pound F), cortisone (compound E) or oorticosterone (com-pound B).
Corticosterone. aldosterone, cortisone and cortisol are stepwise reduced to the dihydro derivatives and then to the tetrahydro derivatives. The 20-keto group may give rise to 20a- as ~ell as to 20B-hydroxy derivatives and the reduotion of the 64-double bond may result in the production ot the Sq- or 5~-compounds. The 3~keto function yields pre-dominantly 3~-h¥droxyl groups.
The corticosteroids can also partly be degraded to 11-ox¥genated 17-ketosteroids.
HOHi!C I 0-0
II.DESC»:"'CORTICO~TmlNl!
t coc ~ () C:ORTICOST~NE" t .) 0 - 0 off
"0Apart from the produotion of characteristic hormones like cortisol ~nd aldosterone, the adrenals secrete steroids with androgeniC and estrogeniO action. Dependent on the stage of the female cycle, the ovary oan produce different amounts of estradiol and progesterone.
The latter compound also appears in the adrenal corte~ as intermediate in the synthesis of cortisol. Consequently an analogy between gonadal and adrenal secretion products can be noticed.
Besides this analogy we are oonfronted with the fact that during the metabolism steroids of different origin and/or hormonal actiOn can give rise to the same metabolites. Etiocholanolone present in the urine can come from the meta-bolism of gonadal or adrenal testosterone and dehydroepian-drosterone through the cOmmon metabolite androstenedione. Moreover several adrenocortical steroids showing a l7-hydro-xy group and 20-ketone group, give rise to the 17-ketosteroid etiooholanolone.
This analogy between gonadal and adrenal secretion pro-ducts constitutes One of the problems in endocrinology.
With regard to the biosynthesis of steroids the impor-tant generalization has been established that the synthesis o{ steroidhormones involves a large number of pathways com-mon to all steroid secreting tissues, Certain molecular mo-difications of the steroids are caused by the action ot enzymes with analogous action present in gonads and adrenals.
The basic steroid ring structure undergoes alterations by enzymes such as hydroxylases, which ~ntroduce hydroxyl groups, dehydrogenases, enzymes that remove hydrogen from a hydroxyl group and 6 dehydrogenases that eliminate hydro-gen {rom a C-H group. Each of the enzymes attacks selective positions of the steroid.
For a long time one has regarded the metabolism o{ ste-roids largely as a definitive step of deactivation carried out mainly by the liver and the kidney. An example of such
a deactivation is the conju~ation of the steroids with glu-curonic acid or sulphuric acid, to make thew soluble in water. Later, several findings have changed this id@a~ even conjugated steroids sometimes proved to be metaboli~ed at the steroid part of the molecule without hydrolysis of the conjugate. So, in the metabolism of 17-ketosteroids (alSO called l7-oxosteroids), dehydroepiandrosterone, present as the sulphate, is a precursor of a number ot adrenal 17-ke-tosteroids including androsterone and etiocholanolone.
In urine the major part of the steroids is present as the anion or the salt of the conjugates.
Fi~. 2.11 shows the structure of pregnanediol-3~-glucuro nide and that of dehydroepiandrosterone sulphate is given in fig. :2. 12.
Fi9. 2.ll Pregnanediol-3~-91ucuronide.
Cortisol metabolites like t@trahydrocortisol (THf),
allO-~HF, t@trahydrocortisone (THE), the cortols and cor-tolones are mainly conjugated with glucuronic acid. The glucuronic acid qroup is linked with the steroids in the 3~-position. For testosterone, also present as a qlu-curonide, this linkage is at c-17. Dehydroepiandrosterone is largely excreted as a sulphate. As mentioned earlier this conjugate can be cleaved by the enzyme sUlphatas8.
In urine also free steroids do occur. In the presence of small amounts of proteins and other particles in urine i t is probable that the relatively minute quantities of free steroids are partly associated with, or absorbed on, such particles in a non-specific manner, on account of their low SOlubility in water.
In blood all steroids are to SOme extent associated with plasma proteins. Here we can distinguish between spe-cific binding proteins for certain free steroids, fOr ex-ample cortisol is coupled with transcortin, and free ste-roids more loosely bound to albumin and other non-specific plasma proteins.The conjugates of steroids tooshow linkages with plasma albumin. It is known that plasma albumin COn-tains a large number of binding sites with a high affinity fo~ anionic groups. The binding of steroid sulphates and gluouronides wit.h thl;t protein will be important with regard to the distribution, secretion and biological activity of the steroids.
R@presentative values for the concentrations of some steroids in human urine and peripheral plasma are presen-ted in Table 2.1.
The values in Table 2.1 encompass what is nowadays an important region from relatively easy to rather difficult prOblems of sensitivity with current techniques. So, at present, a variety of techniques will permtt tor example
Table 2,1 concentrations of Some Steroids in Human Urine and peripheral plasma (4).
Stero:ld Urine plasma
(jUgj 24h) (jUg/lOOml) Total corticoids 5,000-20,000 5-25 cortisol 50-100 5-20 Aldosterone 5-15 0.002-0.015 Testosterone (male) 20-200 0.50-1.50 Testosterone (female) 2-15 0.02-0.10 P;J;ogestero ne (female) 0.02-2.00 Progesterone (pregnancy) 5-25 Pregnanediol (female) 1,000-8,000 PregnanedioL (pregnancy) 5,000-75,000 Estradiol (:eemale) 0-15 0.005-0.020 Estriol (female) 0-75 Estriol (pregnancy) 100-'50,000 5-20
the measurement of testosterone in male plasma and estriol in the urine of a pregnant woman, but only few techniques enable U5 to determine estrogens in pla5ma and testosterone in the urine or pla5ma of women.
2.6 Why i t is meaningful to analyse steroids
A question the clinical chemist involved in endrocrino-logy has to answer is: is the secretion rate of a steroid su~normal, in the normal range or abnormally large? An an5wer to this question directly reflects the possibly de-viating action of an endrocrine orsan. Certain steroids can
be secreted in too large amounts (hyperfunction) or in "too lowamount.s (hypofunction). Al50 i t Can happen that 50me ste-roids are produced that normally do not occur as significant secretion products of a certain endocrine gland (dysfunction).
Here we have to say something about the expression "normal value". In the literature sometimes considerable differences are encountered in tables with normal values. In fact i t is not easy to determine a normal value. Apart from differences caused by the method of analysis, one has to do with a biOlogical spreading. For example the excre-tion of steroids in a 24-hours urine of one healthy indi-vidual may vary between certain limits from one day to the other, while the excretion of the same steroids in another healthy person may fluctuate between somewhat different limits. So, in any case, it is better to talk about "a nor-mal range" instead of Qa nOrnor-mal value",
The decision, from the result of an analysis, that one has to do with a pathological case must be based on solid reasons. Here problems arise like: which weight can be attributed to one individual analysis; is i t necessary to isolate more steroids at once and so on. This subject will be treated later.
In cases of biochemical defects one may notice that pro-ducts of the reactions prior to the deficient step in the metabolism pattern reach abnormally high concentrations and that their conversion by branches of the main pathway be-comes spectacularly prominent. If the main pathway is out off, through the absence of a certain enzyme, this is called "a block".
The most common defect in steroid biosynthesis is the lack of steroid 21-hydroxylase activity. Pregnanetriol andpregnanetriolone are secreted in abnormally high quan-tities. As can be seen in fig. 2.10 an insufficient acti-v'i ty ot this enzyme causes that too low amounts of cortisol and corticosterone are secreted by the adrenal corteX. As a consequence, the production of ACTH by the pituitary gland is increased (feed~back mechanism). The ACTH in its turn stimulates the adrenals and an overproduction of 17-keto5teroids results in excessive androgenicity.
Other forms of this "adnmogenital syndrome" are the
ll~-hydroxylase deficiency and the absence of 3~-hydro
xysteroid-dehydrogenase. The former leads to an accumula-tion of Il-desoxycortisol (compound S) and II-desoxycorti-costerone, normally found in small amountS. In the case of 3B~nydroxysteroid-denydrogenase deficiency no oxidation of the hydroxyl group in position C-3 and nO shift of the double bond from C-5 to C-4 occurs. Consequently the progesterone production is too lo~ and the same applies to the aldoste-rone and cortisol production.
The symptoms of these disorders in children are variable. !n female infants usually abnormalities in the external genitalia are seen which can lead to misidentification of the sex. At older age the excessive production of adrenal andro-gens leads to rapid growth and symptomS of virilization in patients with the 21-hydroxylase- and llB-hydroxylase defect. In patients with 3a-hydroxysteroid-dehydragenase deficiency there are only minimal symptoms of virilization due to limited androgen production by adrenal glands and testes. These patients always have symptomsof denydration and salt loss due to the limited aldosterone production.
Early diagnosis and treatment
ot
the described dis-orders in steroid metabolism is essential to prevent abnor-malities in growth and development and to substitute the cortisol and aldosterone lack.In severe cases for example when the enzymes necessary for the breakdown of the side chain of cholesterol are absent, a complete adrenal insufficiency generally leads to an early demise.
It is also worth mentioning tnat the production of spe-cific enzymes that modify cnemical groups at spespe-cific lo-cations is indirectly regulated by the genes. Therefore, i t is not so strange that abnormalities in steroid biosynthesis are in fact inherited diseases. In this context the expres-sion "inborn error of metabolism" is used.
Besides the ~aren0genit~l syndrome some other disorders are known, having their typically modified steroid pattern. Cushing's syndrome: Hyperadrenalism, mostly in WOmen, which dis~ase may be associated with a tumour of the hypophysis or adrenal hyperplasia and tumours.
Via ACTH, tne overproduction of adrenocortical hormones leads to soaium retention, potassium loss, al~alosi5, hyperglycemia and even diabetes. In most cases a high urinary excretion of cortioosteroids and 17-ketosteroids is notioed.
Conn's syndrome: In this disease the adrenal glands generally reveal benign adenoma, rarely a malignant tumour. The aldosterone exoretion level will be high and hypokale-mia, alkalOSiS, hypernatrehypokale-mia, polyuria and hypertension are characteristic.
Addison's disease: Here we deal witn a hypofunction of the adrenal as the result of a serious pathological process affecting this gland. The urinary excretion of corticoste~ roid metabolites, including the 17-ketosteroids, is low. No response is obtained on ACTH administration. Apart from a fall in sodium and rise in potassium such patients show a low resistance to infection and readily become dehydrated. Stein-Leventhal syndrome: A syndrome in ~omen usually with an increased androgen production. The ovarian function is abnor-mal and there are indications for abnorabnor-mal androgen production
by the adrenal gland.
The exact nature of some of these disorders is still
partly unknown. Isotopic studies of steroid matabolism and the study of the labeled exoretion products, have thrown some light on unknown aspects concerning steroid production, distribution, secretion and @xoretion in these abnormal cases. The analysis of steroids in cas@s of stimUlation or suppresion of the adrenal cortex and the gonads Can contri-bute to the solution of these problems as well.
roid analysis can facilitate a fast diagnosis, can show the effect of the therapy and so develop into a routine method. An example of a routine method, having already proved its
importance, is the analysis of pregnanediol and estriol in the urine ot pregnant women. In this way hormonal distur-bances in the foetoplacental unit can be traced.
2.7 References
l . E. Heftmann, "Steroid Biochemistry", New York, Academic Press, 1970.
2. L.F. r;leser and M. Fieser, "Steroids", New York;, Reinhold, 1959.
3. W. Klyne, "The Chemistry of the SterOids", New YOrk, Wiley, 1961.
4. H.J. van d@r Molen, F.H, de Jong and B.A. Cooke, Clin. chim. Acta 34 l69 (1971). 5. Int. un. of Pure and Appl. Chern. and Int. Un. of
Biooh., Definitive Rules for Nomen-clature of Steroids, (Butterworths, London) 1971.
Chapter 3
CURRENT PROCEDURES
INSTEROID ANALYSIS
3.1 G~neral remarks
Th@ direct estimation ot a hormone in the venous efflu-~nt from a gland is not possible wi~hout interfering with th@ integrity of the body by surgical intervention in order to obtain a blood sample. Therefore indirect methods must be applied.
The level of a hormone in the peripheral blood is an interesting parameter, but studies on blood are sometimes hampered by the low or exceedingly low concentration at which most steroids are present in this medium.
By the excretion of steroid metabolites in the urine the activity of steroid prodUcing glands can also be studied. However,as said in the previous chapter, we must realize that the urinary steroid composition often reflects imper-fectly the functional state of the steroid secreting glands.
During the past thirthy years a great number of procedures for steroid estimation has been published. Al-though the principles of the described methods can differ widely, sometimes only slight differenceS in the way of execution of a procedure are noticed between several publi-cations. An explanation for this is rather obviouS. The iso-lation oe steroids from biological samples mostly consists of. sample pretreatment steps like hydrolYSiS, extraction and extract purification,followed by detection of the steroids. Many investigators have studied the effect of variations of pa,rameters in a certain method of analysis and have reported SU(;Ce<;sful changes. 'l'he devel.opment of efficient separation techniques has contributed to a higher specificity and sen-sitivity of the classical methods, often based on group
analysis. In the literature many modifications of existing classical methods are presented. In addition, in the oourse of time, very sensitive, detection methods have evolved which in combination with separation systems of high resolution resulted in methods that meet the require-ments of the usual criteria of reliability and prac-ticability.
In order to assess the validity of methods and procedures for steroid analysis the following criteria may be used:
accuracy
precision
sensitivity
specificity
the degree of agreement between the measured a~unt and the actual amount of a compound in a sample the degree of agreement of repli-cate estimations
the least quantity of substance which can be measured
the extent to which a procedure measures only the substance in questiOn.
While the above mentioned criteria cover the reliability of a method, one has to consider simplicity, speed and costs. Although often procedures of adequate sensitivity are available, the simplicity of a method is usually in-versely proportional to its specificity. For that reaSOn a compromise has to be reached, a method being chosen which is as specific as possible, but at the same time not unduly elaborate.
From that pOint of view a distinction can be made be-tween relatiyely simple routine procedures and more elabo-rate methods deSigned to establish without reasonable doubt the identity and quantity of various steroids in a small number of b~olo9ical samples d~rectly for research purposes.
This chapter presents some of the current techniques in s'teroid analysis, including both class:i..cal and modern ones, In the first sections detection methods will be treated and in later sections separation methods.
The usefuL detection ranges belonging to some methods that will be described afterwards, are given in Table 3.) for some important steroids. These values don't represent the rnl\ximurn obtainab1e sensitivity.
Table 3.1 Detection Ranges for Different Methods (1).
Steroid 17-ketosteroids Cortisol Aldosterone Estrogens Pregnanediol Testosterone Progesterone Method Zimmermann Fluorescence Isotope labeling Kober reaction Fluorescence H2SO4 reaction UV-absorption VV-absorpt.ion Range (ng) 100- 500 5- 10 1- 10 200 5 1000-2000 100- 500 100- 500
With protein binding techniques and radioimmunoassay even subnanogram amounts of cortisoL, testosterone, p~ogesterone etc. can be measured.
3.2 Classical detection and quantification methods 3.2.1 The Kober reaction tor estro~ens
The Kober reaction is the best known colorimetric estro-gen estimation method (2).Prior to the reaction of the estrogens with th~ hydroquinone solution in sulphuric acid, the method involv~s acid hydrolysis, ether extraction, a
phase-change purification procedure for th~ phenolic fraction depending on methylation of the phenol group and separation of th~ estrogen methyl ethers by alumina chromatography.
This method has proved to be of insuffioient specificity to justify its,us~ for the estimation of estrogens in the
~resenoe of significant interfering mat~rial without ~fficient
preliminary separation from such material. Even after purifi-cation a considerable spectrophotometric correction has to be appliect, indicating that much nonspecific interfering colour may be produc~d during the Kober reaction.
Th~ Kober m~asurem~nt of th~ low conc~ntrations of estrogens existing in plasma can only be performed with difficulty.
In a modification of th~ Kober reaction the colour is extracted into chloroform containing p-nitrophenol and etha~
nol. A higher degree of specificity is claim~d and evidence for its superiority over other Kober moctifications is given by Jttrich (3).
3.2.2 ~he Zimmermann reaction for 17-ketosteroids
The most widely used classical method for the determi-nation of 17-ketosteroids in urin~ and blood is based on the publication of Zimmermann (4).
After reaction with m-dinitrobenzene in alkaline solution the -CH
2-C=O structure at C16-C17 gives rise to a reddish-purple colour, giving a maximum absorption at about 520 nm.
~he ketonic groups located at oth~r positions in the steroid molecule in general give less extinction on a molar base and there is a shift in the absorption maxima. By washing of the steroid extract with alkali a neutral I7-ketosteroid fraction is obtained, thus exclucting those 17-ketosteroids in which the A-ring is phenolic (~.9. estrone). The method recorded by PeterQon and rierce (5) can be taken as r~presentative for the isolation of total urinary I7-ketosteroids followed by Zimmermann detection.
A tr.~ctionat1on of the individual 17-ketosteroids can also be carried out, then the extracts can for example be subjected to paper chromatography or gradient elution On alumina adsorp-tion columns, while each tracadsorp-tion is assayed for 17-ketoste-roids using the Zimmermann reaction. For the GLC analysis of individual 17-ketosteroids is referred to chapter 6.
A number of 17-ketosteroids possess a hydroxyl group at c-3. In some compounds the OH-group occupies the ~-position in relatiOn to the methyl group at C-IO,but other steroids have the OH-group in s-posit1onl the B-17-keto5teroid frac-tion. Digitonin, a steroid sapogenin, under certain conditi-ons precipitates those steroids having a 3S-nydroxy configu-ration (6), In the urine of normal adults the ~-fr~ction of 17-ketosteroids constitutes about 15 percent of the total
17~ketosteroid fraction.
3.2.3 Reactions on the side chain of C-21 steroids
In chemical assays~he side chain and possibly also other groups of adrenocorticosteroids can react. Generally these re-actions give rise to the fOrmation of coloured complexes, the liberation of volatile products and the formation, by trans-formation, of new steroids.
The reducing properties of the ~-ketolio side ohain (20-keto, 21-hydroxy), will give rise to ooloured complexes after the addition of certain reagents. A reagent commonly used in this type of reaction is blue tetrazolium (3,3'-dianisole bis [4,4'- (3,5-diphenyl) tetrazolium chloride] ) which forms a deep-blue-coloured complex. Despite the fact that traces of impurities remaining in a final extract reduce the speCifi-city and lower the sensitivity, the blue tetrazolium TIlethods have shown their usefulness as routine reactions for clini-cal purposes (7).
A popular method for the estimation of C-21 steroids applicable to blood and urine is the Porter-Silber reaction
(8). 17u,21-Dihydroxy-20-ketost@roids form coloured hydra-zones witn sulphuric acid solutions of phenylhydrazine. These derivatives have an absorption peak close to 410nm. The ori-ginally.described procedure g~ve rise to serious steroid losses and also several components present in urine gave significant absorption values at 410 nm. So, later on some changes from the original method nave been introduced including more elaborate purification procedures, to obtain
leSS chromogenic Porter-Silber reacting material (9). One must realize that the Porter-Silber reaction cannot b@ used for corticosterone, and/or its metabolites, l7 ~ hydroxy-20~- and 20B-reduced steroids, and other 17-d@oxy-corticosteroids.
Steroids having a glycerol- or 17~, 20-diol-21-methyl type side chain can be oxidized with periodic acid to give 17-keto-steroids which can be determined with the Zimmermann reagent
(l0) ,
Thus an analysis of the 17-ketosteroid content of ali-guots of the extract prior to and after periodic acid oxida~ tion provides a measure of the adrenal steroids having the type of side chain as described above. Since these normally represent only a small percentage of the total amount of urinary corticosteroids, this method is limited in its
applicability. Moreover i t has been shown that some nonste-roidal substanoes of the reagents also react to give coloured products in the Zimmermann reaction.
Also steroids containing a 20, 21-diol, 20-one~21~ol and 17 ~, 21-oiol-20-one sioe chain can react with periooic acid. The formaldehyde which is liberated as a result of cleavage of the side chain can be measured by the formation of a coloured product upon the addLtion of chromo tropic acid. However, mainly due to pro)dems of formaldehyde reten-tion by nonsteroidal chromogens, only less reliable results have been obtained with this m$thod (11).
Steroids with a 21-methylgroup and at the same time a 20-hydroxy group or a 17 a, 20-dihydroxy group in the side chain lead to acetaldehyde on oxidation with periodio acid. With a modification in the way of acetaldehyde isolation, no interference from chromogenic substanoes present in u~ine extracts is found. ?,'his method appears to have the necessary attributes of sensitivity and specificity. The quantifica-tion, after isolaquantifica-tion, of pregnanetriol and pregnanetriolone
(12) in this manner has found important application. when added to urine, sodium- or potassium bismuthate also causes the oxidative elimination of the side chain of some 17-hydroxylat@d C-21 steroids, yielding 17~ketosteroids
(13). This principle is the basis of the method in which the difference between 17~ketosteroid values prior to and after oxidation affords a measure of the group of 17-ketogenic steroids; CH 20B
I
C=o
:P
H 3 --OH I H CH 20H15
I
OH CH 3 --OH I Hj
H 3~"
jjOH
[,Another version of this method includes the sodiurn-borohydride reduction prior to the bismuthate oxidation, thus comprising also the 17 hydroxy-2Q-keto-21-desoxy-steroids. In that case 20-ketogroups present in the 17 a-hydroxysteroids are converted to hydroxyl groups, followed by an elimination of the side chain by bisrnuthate oxidation. The 17-ketosteroids present in the original extract are reduced to 17-hydroxysteroids 50 that in the final Zimmer-mann reaction only response of 17-ketosteroids derived from 17-hydroxysteroids is found.
Methods based on the oxidative removal of the side chain, involving reduction, oxidation and fractionation steps, per-mit the total estimation of a group of structurally related steroids or determination of the individual components after chromatography. An account of existing reaction schemes is given by Norymberski (14).
3.2.4 spectrophotometric and fluorometric estimations Steroids exhibiting a, ~-unsaturated ketones, with the A4-3-keto as the most common structure, absorb UV light at ca. 240 nm. Methods using this UV absorption have been applied for the estimation of progesterone (15) and testo-sterone (16).
The presence of a ketone group in a steroid molecule has often been used to prepare derivatives that might be measured by spectrometry: 2,~-ain~trophenylhydrazones, 1so-nicotinic acid hydrazones and thiosemicarbazones have found application.
According to the experience of several authors one may encounter difficulties with these procedures if the stero~d under investigation is contaminated by trace amounts of ke-tonic impurities of the extracts or solvents.
The ohromogen formed by the action of concentrated sul-phuric acid upon 64-3-ketosteroids gives also rise
to
charac~
teristic absorption maxima.A m~thod for pregnanediol ~nd pregnanetriol with spectro-photometric detection uses the yellow oolour given in reac-tion with concentrated sulphuric acid.
The group of
~5-3t-hYdroxy
steroids oan be determined with the Allen test (17). Steroids with this structure pro-duce a reddish-purple colour with a mixture of sulphurio acid and ethanol.Fluorescence, a process in which molecules emit light when irradiatedJcan be used for the estimation of several steroids. Mineral acids as sulphuric acid and phosphoric acid are usually added to the biological extracts in methods involving fluorescence. In order to prevent the reaction of these acids with nonsteroidal substances often meticulous prepurifications are necessary.
The l~rge number of publications on fluorometric methods for corticosteroids does not imply a widely clinical appli-oation. Rather, owing to the difficulties involved in puri-fication stevs, the desoribed procedures are, more used for researoh than in routine.
An illustrative sulphuric acid fluoresoenoe method for cortisol is reported by Clark and Rubin (18) while Korenman et ~l. (19) have desoribed a fluorornetric estimation of testosterone in urine. Further i t is by phosphoric ~oid fluoresoence that pregnanetriolone was identified for the first time as an urinary met~oolite.
For estrogens sulphuric acid fluoreSCence has been ex-tensively used. Optimal conditions for estimating urlnary estrogens in this way are reported by Preedy and Aitken (20).
3.3 The use of radiOisotopes
Isotopic techniques have proved to be of great value in the stUdy of steroid metabolism. In most cases 14C_ and 311 _ labeled steroids are used.
the addition of a known number of counts of an authentic radioactive stero:ld to a sample containing- the cold steroid
(to be estimated). After establishing the specific activity of a purified and weighed amount of steroid extracted from the sample the amount of inert steroid in the original sample can be calculated by proportion (21).
Reverse isotope dilution is "designed to measure radio-active metabolites in biological fluids. Here, similar principles as in isotope dilution <l,re involved (22)'.
Likew:lse based on the fact that radioactiv@ steroids are assumed to behave in the same way as the naturally occurr:lng inert oneS, the former can be added to a biological sample to determine losses during sample treatments. The number of counts found after sam~le manipulations divided by the num-ber of counts originally added will give the proportion of the steroid lost during processing.
In the present investigation use has been made of this teohnique (see chapter 6).
Labeled reagents can be used as tools for quantifying steroids with whiCh they combine in deriVatives. For example 35S-thiosemicarbazones can be made (23). To correct for losses of the fo~med derivative during purif:lcation steps a known amount of 14c-labeled derivative can be included, while trou-bles with variable yield of derivative formation can be eli-minated by the use of
3E-!a~~led
steroids as indicators (24). This teohnique is called complete indicator technique.Radioactive steroids are also of essential imvortance in the different forms of saturation analysis. Here radioac-tive steroios fulfil the function of a traoer that reveals the oistribution of ' the steroid under invest:lgation follo-wing the ~eaction with a protein or an antibody. Moreover the labeled steroid can be used to measure the extraction recovery.
3.4 Assays based on steroid-protein interaotions
Following the initial development, roughly a deoade ago, oompetitive protein binding and radioimmunoassay have made major advance in steroid analysis.
Especially in toose cases woere minute amounts of steroids have to be quantified these teChniques are very suitable on showing sensitivities of an order commensurate with the con-centration at which many steroidS exert their effects in bio-logical systems.
Although protein binding methods were first applied to the measurement of the concentration ot steroid hormones in plasma, later on i t was pointed out that theSe techniques are applicable to urine and other biological tluids as well.
3.4.1 Terminology and theoretical aspects
The term saturation analysis refers to an assay depen-dent on the competition of two forms of toe same molecule fOr reactive sites On a molecule of lesser concentration
(25, 26).
Cornpetive protein binding (Cpa) analysis may be defined as that form of saturation analysis in which one reacting molecule is a protein with high affinity and preferentially high specifioity for the other which is thus the ligand. Vroteins, possessing these specialized properties have been shown to maKe assays of the reqUisite specificity and sen-sitivity practical for hormones in the prepurified and preseparated complex mixtures found in biological materials. A further subdivision can be made by distinguishing the na~ ture ot the protein. Assays in whiCh the protein is an anti-oody are known as radioimmunoassays. rhose in which the pro-tein is a natural binding propro-tein as competitive propro-tein bin-ding proper.
CPB analysis for steroids began as the result of the ob-servation that the binding-curve obtained on adding inert cortisol to diluted plasma containing labeled cortisol looked
as though i t could be used as a standard ourve for cortisol. Hereafter CPB methods for the quantitation of hOrmon~s were report~d by Murphy et al. (27).
The interaction between a prote~n (P) and a steroid IS) can be represented by the equation: S+P E ~ SP. At equili-brium an association constant K can be introduced:
K '" where
[5J,
[pJ
and [SP] are the molarconcentrations respectively of the free (unbound) steroid (S), the free protein (P) and the protein bound steroid (SP). Whereas
[sJ
+[sP]
corresponds to the total steroidconcen~
tration at equilibrium,[pJ
+[s~
corresponds to the to-tal protein concentration. At a constant concentration of P the fraction of bound steroid will decrease with increasing st~roid concentration.This means that at equilibrium the fraction of bound steroid is a function of the total hormOne (HT) present:
~rJ
+[sJ
Hence when
[sJi}
is knownHT can be determined.
A tracer amount of the radioactive steroid is used as an indicator to determine the fraction of the steroid which is protein bound. When~ver a suitable binding protein and a radioactive tracer are available a CPB method can be set up for a steroid estimation.
Fig. 3.1 shows the different steps involved in a satu-ration analysis. After addition of radioactive steroid (SX) to the biological medium and equiliorat~on S may, if neces-sary, be extracted from ~ts mil~eu and purified, recovery being monitored by the radioactivity in the fihal extract.
Subsequently the extracted compound is mixed with a specific protein P, in such relative concentration that part
extraction free SXradioactivity SXp radioactivity f(3). P Sp
s
separates
~ound freeFig. 3.1 ~rincipl~ of Saturation Analysis.
of S reacts with P (bound S) and part remains in the unre-act@d form (free S). Now, the ratio of reacted and unreacted S, as indicated by the distribution of the radioactive com~ pound, is sensitively dependent on the total concentration of S present. Clearly, radioactive distributions yielded by un-known amounts of S can be compared with those yielded by a set of standards, and the unknowns thereby estimated.
3.4.2 Methodological considerations in CPB analysis
Next to the extraction of the 5teroid under test, reaction with a chosen protein takes place. However, i t is impossible that an aliquot of the sample to be tested is simply added to the aSSay system. Interfering binding proteins in the steroid sample have to be removed or destroyed, for instance by pre-cipitation with organic solvents. Further, the separation of competing analogues is a troublesome aspects of CPB analysis.
If satisfactory separation systems were available speci-ficity would cease to be a probl@m. H@re the critical point J.s that separation systems as thin-layer chromatography and paper chromatography, give rise to a high blank value. Al-though a variety of ways to decrease the ~lank have been
