New comprehensive biochemistry vol 18 hormones and their actions part i

Role of lipoproteins in steroidogenesis Although cholesterol is accepted as the major precursor of steroid hormones as a result of side-chain cleavage to pregnenolone see below, researc

PART I

New Comprehensive Biochemistry

Hormones and their Actions

Editors

Department of Biochemistry, Royal Free Hospital School of Medicine, University

of London, Rowland Hill Street, London NW3 2PF, England

R.J.B KING

Hormone Biochemistry Department, Imperial Cancer Research Fund

Laboratories, P.U Box No 123, Lincoln’s Inn Fields,

London WC2A 3 P X , England

H J van der MOLEN

Postbus 93138, 2509 A C Den Haag, The Netherlands

Nederlandse Urganisatie voor Zuiver- Wetenschappelijk Onderzoek ( Z W O ) ,

1988 ELSEVIER Amsterdam New York Oxford

All rights reserved No part of this publication may be reproduced, stored in a retrieval system, or trans- mitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without the prior written permission of the Publisher, Elsevier Science Publishers B.V (Biomedical Division),

P O Box 1527 1000 BM Amsterdam, The Netherlands

No responsibility is assumed by the Publisher for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from any use or operation of any methods, prod- ucts, instructions or ideas contained in the matcrial herein Because of the rapid advances in the medical sciences the Publisher recommends that independent verification of diagnoses and drug dosages should

be made

Special regulafions for readers in the USA This publication has been registered with the Copyright Clearance Center, Inc (CCC), Salem, Massachusetts Information can be obtained from the C C C about conditions under which the photocopying of parts of this publication may be made in the USA All other copyright questions, including photocopying outside of the USA, should be referred to the Publisher ISBN 0-444-80996-1 (volume)

USA

Library of Congress Cataloging in Publication Data

Hormones and their actions / editors B A Cooke, R J B King, H.J van der Molen

p cm (New comprehensive biochemistry; v 18A-)

Includes bibliographies and index

1 Hormones Physiological effect I Cooke Brian A 11 King, R.J.B (Roger John Benjamin) 111 1V Series: New comprehensive biochemistry; v 18A, etc

[DNLM: 1 Hormones-physiology W1 NE372 v 18 / WK 102 H812781

ISBN 0-444-80996-1 (pt 1)

Molen, H J van der

QD415.NJ8 vol 18A etc

Department of Medicine, University of Colorado Health Sciences Center, 4200 East

Ninth Avenue, Denver, C O 80262, U S A

Molecular Endocrinology Laboratory, Imperial Cancer Research Fund, P 0 Box

123, Lincoln’s Inn Fields, London W C 2 A 3 P X , England

Contents

List of contributors v

Section I General aspects of hormones and hormone actions Chapter 1 The biosynthesis of steroid hormones: an update by D B Cower 3 1 Introduction

2 Role of lipoproteins in steroidogenesis

3 Mitochondria1 cholesterol

3.1, Transport of cholesterol into mitochondria

3.2 Intramitochondrial transport of cholesterol

4 Side-chain cleavage (SCC) of cholesterol

5 Biosynthesis of corticosteroids

5.1 5.2 1 I@- and 18-hydroxylases

5.3 Formation of aldosterone .

6 Biosynthesis of the androgens

6.1, 6.3 Interconversion of 4-androstenedione and testosterone

6.4 Conversion of testosterone into Sa-dihydrotestosterone (Sa-DHT) 7 Biosynthesis of oestrogens

8 Secretion of synthesized steroid hormones

9 Conclusion

Acknowledgements

References

Enzymes involved in corticosteroid biosynthesis

Action and properties of 17-hydroxylase and C-17,2 0-lyase

6.2 Conversion of S-ene-30-hydroxy- to 4-en-3-oxosteroids

3

4

4

4

6

8

11

12

13

14

15

17

18

20

20

20

24

25

25

25

Chapter 2 Overview of molecular aspects of steroid hormone actions b y R J B K i n g 29

1 Introduction 29

2 Intracellular events in steroid action 29

2.1 Intracellular location of receptors 29

2.2 Receptor structure 31

2.3 D N A binding 31

3 Specificity of steroid action 32

3.1 Ligand availability 32

3.2 Ligand specificity of receptor 34

3.4 Availability of responsive genes 35

3.5 Specificity of the steroid response element 36

References 37

3.3 Agonismiantagonism 35

Chapter 3 Gene regulation by steroid hormones by M.G Parker 39

1 Introduction 39

2 Structure and function of steroid receptors 39 3 Steroid receptor-DNA interactions 42

3.2 Specific D N A binding 43

4 Steroid receptor-chromatin interactions 46

5 Steroid hormone-activated gene networks 46

References 47

3.1 Non-specific D N A binding 43

Chapter 4 Characterization assay and purification of steroid receptors by M A Blankenstein and E Mulder 49

1 Introduction 49

2 Properties of steroid receptors 50

2.1 Binding properties 50

2.2 Physico-chemical properties 52

3 Assay of steroid receptors 53

3.1 General aspects and radioligand assays 53

3.2 Separation of bound and free ligand 54

3.3 Immunological assays 54

3.4 Other steroid receptor assays 55

4 Purification of steroid receptors 55

4.1 General protein purification 55

4.2 DNA-affinity chromatography 56

4.3 Steroid affinity chromatography 56

4.4 Immunoaffinity purification 57

References 58

5 Characterization of steroid receptors 57

Chapter 5 Mechanism of action of thyroid hormone by J Nunez 61

1 Introduction

2 Thyroid hormone production transport and uptake by the target cells

3 Thyroid hormone nuclear receptors and cellular binding proteins

3.1 Nuclear receptors

4 Induction and repression of pituitary hormones

4.2 Thyrotropin

Regulation of lipogenesis in the liver

5.1 Malic enzyme

5.2 Fatty acid synthase

4.1 Growth hormone

5

61

63

64

65

66

66

68

68

68

70

6 Effects of thyroid hormone on the receptor-adenylate cyclase system in the adipocyte

and the hepatocyte 70

7 The muscle cell: P-adrenergic responsiveness and the expression of myosin heavy chains 72 8 Thyroid hormones and brain development 73

8.1, Neuronal differentiation 74

8.2 Glial cell differentiation 75

9 Conclusions 76

References 76

Chapter 6 Metabolism of thyroid hormone by T.J Visser 81

1 Metabolic pathways of thyroid hormone

1.1 Introduction

1.2 Deiodination

2 Type I iodothyronine deiodinase of liver and kidney 2.1 Properties and distribution

2.2 Substrate specificity

2.3 Inhibitors and affinity labels

2.4 Reaction mechanism

2.5 Cofactor requirements

3.1 3.2 Type 111 iodothyronine deiodinase .

3.3 Possible other iodothyronine deiodinases 1.3 Conjugation

3 Iodothyronine deiodinases of other tissues

Type I1 iodothyronine deiodinase .

4 References

Transport of iodothyronines into tissues

5 Regulation of thyroid hormone metabolism

81

81

82

84

85

85

86

87

89

90

93

93

95

96

97

99

100

Chapter 7 Characterization of membrane receptors: some general considerations by L.E Reichert Jr 105

1 Introduction 105

2 Preparation of receptor probe 106

3 Preparation of membrane receptors 107

3.1 General considerations 107

3.2 Membranes from cell cultures 108

3.3 Membranes from tissue homogenates 109

4 Hormone binding characteristics of the membrane receptor 111

4.1 Specificity 111

4.2 Selection of appropriate in vitro system 112

4.2.1 Effects of time, temperature, buffer 112

4.2.2 Steady-state (equilibrium) conditions 112

5 Molecular properties of the membrane receptor 113

6 Solubilization of the membrane receptor 114

7.Su mma r y 115

References 115

Chapter 8 Metabolism and intracellular processing of protein hormones

by A S Khanna and D M Waisman 117

1 Introduction

2 Biosynthesis of protein hormones

2.1, Transcription and translation

2.3 Cleavage of signal peptide

3.1, Structures of prohormones

2.2 Interaction of signal peptide with R E R membrane

3 Processing of prohormones 3.1.1 Pro-opiomelanocortin (POMC) peptide family

3.1.1 1 T h e P O M C gene

3.1.1.2 Distribution and processing of POMC gene products

3.1.1.3 Additional modifications of P O M C peptide family

3.2 Significance of 'pro' sequence

3.3 Cleavage at dibasic amino acids

3.4 Cleavage at monobasic amino acids

Processingenzymes

3.5.1, Endopeptidases

3.5.2 Exopeptidases

3.6 Post-translational modifications

4 Storage of protein hormones

5 Release of protein hormones 6 Circulation in blood

7 Degradation of protein hormones 7.1 7.2

Degradation of glycoprotein hormones

Internalization of protein hormones

8 Conclusions References

117 118 118 120 120 121 122 122 122 122 123 123 123 124 124 124 125 126 127 127 128 128 128 129 130 130 Chapter 9 Internalization of peptide hormones and hormone receptors by D.L Segaloff and M Ascoli 133

1 Introduction 133

2 General features of receptor-mediated endocytosis 134

3 Methods used to assess receptor-mediated endocytosis 137

3.1 Morphological approaches 137

3.2 Biochemical approaches 138

4.1 Microaggregation 144

4.2 Internalized and degraded hormone 145

4.3 Receptor down-regulation 146

5 Conclusion 147

References 147

4 Biological consequences of receptor-mediated endocytosis 144

Chapter 10 Physiological aspects of luteinizing hormone releasing factor

and sex steroid actions: the interrelationship of agonist and antagonist

activities by A E Wakeling 151

1 Introduction 151

2 L H R H and L H R H analogues 152

2.1 Physiology 152

2.2 Biological activity of L H R H analogues 154

3 Steroid antagonists 156

3.1 Physiology 156

3.2 Antiandrogens 160

3.3 Antioestrogens 161

References 104

Section I1 Specific actions of steroid hormones Chapter I1 The functions of testosterone and its metabolites by W.I.P Mainwaring S A Haining and B Harper 169

1 Introduction

2 T h e functions of androgens in various target organs 2.1 Testis

2.2 Urogenital tract

2.3 Haemopoietic organs

2.4 Salivary glands

2.5 Kidney

2.6 Muscle

2.7 Liver

2.8 Central nervous system

2.9 Anterior pituitary

2.10 Breast

2.11 Hair

2.12 Sebaceous glands

2.13 Skin

2.14 Bone

169

174

174

175

177

178

179

182

185

186

188

188

188

189

189

190

2.15 Lymphocytic organs 190

2.16 Accessory sexual glands 190

2.16.1 Prostate 190

2.16.2 Seminal vesicle 191

2.16.3 Epididymis 191

2.17 Exotic systems 191

3 Concluding remarks 192

Acknowledgements 194

References 194

Chapter 12 Oestrogen actions b y R L Sutherland C K W Watts and

C.L Clarke

1 Introduction

2 Oestrogen receptors

3 4 Oestrogen control of gene expression 5 Oestrogen control of cell proliferation 6 Antioestrogen actions

7 Conclusions

Acknowledgements

References

Oestrogen receptor genes

Chapter 13 Glucocorticoid receptor actions b y I/ Gehring 1 Introduction

2 3 Lymphoid cell variants with altered hormone responsiveness

4 Glucocorticoid receptor defects

5 Molecular weights of glucocorticoid receptor polypeptides

6 Partial proteolysis of glucocorticoid receptors

Glucocorticoid induced lymphocytolysis

7 Functional domains of glucocorticoid receptors

7.1 The M domain

7.2 The DNA binding domain

7.3 The hormone binding domain

Hormone independent gene activation by truncated receptors

7.5 A chimaeric receptor

Glucocorticoid response elements

7.4 8 9 Higher order structures of glucocorticoid receptors References

Chapter 14 Progesterone action and receptors b y N L Krett D P Edwards and K B Horwitz

1 Introduction

2 3 Mechanisms of action

Physiology and clinical uses

3.1 Recent technological developments

3.1.1 Receptor purification

Affinity labeling of receptors

3.1.3 Anti-receptor antibodies

Cloning of the PR cDNA

3.2 Progesterone receptor structure

3.1.2 3.1.4 3.2.1 The A- and B-receptor question

3.2.2 Native PR structure: purification studies

3.2.3 Native PR structure: immune analyses

3.2.4 Native receptor structure: phosphorylation

3.3 Intracellular localization

3.4 Receptor function: regulation of gene expression

197

197

200

203

205

207

210

212

213

213

217

217

217

218

220

221

222

222

224

226

227

229

230

230

233

235

241

241

241

243

243

243

244

244

245

245

245

249

251

254

255

257

3.4.1 Nuclear matrix

3.4.3 DNA hormone response elemcnts

4 Conclusions

References

3.4.2 Acceptor proteins

Acknowledgements

Chapter 15 The pleiotropic vitamin D hormone by L Cancela G Theofan and A W Norman

1 Introduction

2 Production and metabolism of vitamin D

3 Modes of action of 1.25(OH), D,

3.1 Introduction

3.2 Receptor-mediated genomic interactions

3.2.1 1.25(OH),D3 receptor characteristics

3.2.2 Evidence for the genomic actions of 1.25(OH), D,

3.3 Evidence for non-genomic actions of 1 25(OH)2D3

4 Vitamin D and the maintenance of mineral homeostasis

4.1 T h e kidney

4.2 T h e intestine

4.3 Bone

4.4 T h e reproductive stages

5 Non-classical vitamin D responsive systems

5.1 T h e pancreas

5.3 Neural tissues

5.4 Contractile tissues

5.4.1 Skeletal muscle

5.4.2 Cardiac muscle

6 Vitamin D and the immune system

7 Clinical disorders related to vitamin D

8 Summary

References

5.2 Reproductive organs

258 258 259 261 262 262 269 269 269 271 271 271 271 272 274 276 276 277 277 278 280 280 280 281 281 281 282 282 284 285 286 Subject Index 291

General aspects of hormones

and hormone actions

0 1988 Elsevier Science Publishers BV (Biomedicd Division) 3

CHAPTER 1

The biosynthesis of steroid hormones:

an up-date D.B GOWER

Division of Biochemistry, United Medical and Dental Schools (Guy's Hospital), London

SEI 9RT, England

1 Introduction

During the past 50 years, numerous experiments have been performed in attempts

to unravel the complex pathways whereby steroid hormones, that is the corticoste- roids, androgens and oestrogens, are formed in mammalian and other tissues A

very large number of books and reviews have already been written on the subject and this present chapter will seek to (a) summarize the pathways concerned and how the evidence for these was obtained and (b) provide an update of advances over the past decade, particularly with regard to the properties of the steroid trans- forming enzymes involved and the mechanism of the reactions catalysed by such enzymes

The very early experiments which were designed to elucidate pathways of steroid hormone biosynthesis were done using large quantities of putative precursors These were either administered to the whole animal, in which case changes in urinary out- put of steroids were studied, or incubated with tissue fractions, when metabolites

of the added steroid were investigated The quantities of steroids were, of course, grossly unphysiological and it was not until labelled compounds, such as acetate and cholesterol became available commercially that greater advances were made When 'H-labelled material of high specific radioactivity became available still later, it was possible to utilize extremely small quantities of the steroid, or precursor, so that the finely balanced mechanisms of steroid hormone control were not unduly affected

Correspondence to: Professor D.B Gower Division of Biochemistry, UMDS (Guy's Hospital), Lon- don SE1 9RT, England

(D.B Gower is Professor of Steroid Biochemistry, United Medical and Dental Schools, (Guy's Hos-

pital), University of London.)

Having said this, no criticism of the early researchers in the field of steroid hormone metabolism is intended; they could only experiment with materials currently avail- able t o them

2 Role of lipoproteins in steroidogenesis

Although cholesterol is accepted as the major precursor of steroid hormones as a result of side-chain cleavage to pregnenolone (see below), research over the past decade or so has focused on the mechanisms by which steroidogenic tissues obtain cholesterol It should be borne in mind that such tissues require cholesterol, not only for steroid synthesis but also for membrane synthesis, and hence require more

of the precursor sterol than other tissues Morris and Chaikoff [l] showed that the bulk of rat adrenal cholesterol was derived from circulating cholesterol, and later work revealed a similar state of affairs in humans

Through the work of many groups [2-61, there is now no doubt that steroidogenic tissues, such as adrenal, ovary, placenta and, possibly, testis of many species derive much of their cholesterol from plasma lipoproteins These are macromolecules consisting of protein (apolipoprotein) and lipids and, depending on their hydrated densities, are classified as follows: chylomicra, very low density lipoproteins (VLDL), intermediate density lipoproteins (IDL), low density lipoproteins (LDL) and high density lipoproteins (HDL) The two last mentioned consist of a core of hydrophobic lipids, primarily cholesterol esters and triacylglycerols, surrounded by

a monolayer of hydrophilic phospholipids, cholesterol and apolipoprotein For ex- ample, plasma LDL contains apoprotein B (25%) and various lipids, of which nearly 50% is cholesterol ester Phospholipids, cholesterol and triacylglycerols constitute the remainder of the lipids [7,8]

Cholesterol appears to be taken up from plasma lipoproteins by steroidogenic tissues by two receptor-mediated pathways - the LDL pathway and the H DL path- way Not all tissues of all species can utilise both of these; thus, the LDL pathway appears to occur in all species, including man, whereas the HDL pathway occurs mainly in rodents LDL lipoproteins interact specifically with cell surface-bound re- ceptors, as shown for, e.g., adrenal [9] and ovary [lo], after which internalization occurs by endocytosis and hydrolysis of LDLs, plus their cholesterol ester comple- ment by lysosomal action

3 Mitochondria1 cholesterol

3.1 Transport of cholesterol into mitochondria

The next events in steroidogenesis must obviously include the transport of choles- terol and cholesterol ester to the required organelles, in particular, cholesterol into

biosynthesis It seems likely that the cytoskeleton, including the array of microfi- laments and microtubules, may play an important role in processing of lipoproteins and in intracellular cholesterol transport Such evidence that exists has been ob- tained in studies using colchicine, which affects microtubules and interferes with steroid production in steroidogenic tissues, thus implying the necessity for these structures [ll] A second drug, known to alter the structure of microfilaments by causing their cross-linking and polymerization, is cytochalasin B Treatment of ad- renal and ovarian cells with this caused rapid and reversible inhibition of trophic hormone-induced steroidogenesis [ 121 Further evidence for the involvement of the cytoskeleton in steroid transport has been provided by Hall and co-workers [13], who showed that transport of cholesterol into mitochondria, and steroidogenesis, were both reduced in mouse Y-1 adrenal cells in culture in response to anti-actin antibodies

The release of cholesterol from cholesterol esters occurs extra-mitochondrially

by means of a cholesterol ester hydrolase in adrenals, ovaries and testicular Leydig cells (see Ref 6 for review) This enzyme has been studied mostly in adrenal prep-

arations, and is known to be activated and de-activated by reversible phosphoryl- ation [14] and that the phosphorylation was brought about by a c-AMP-dependent protein kinase [15] Hence, ACTH stimulation of cholesterol ester activity in the adrenal occurs via the kinase and, in a similar way, trophic hormone stimulation of ovarian and testicular cholesterol ester hydrolases may occur and provide a large pool of cholesterol for steroidogenesis [ 1 ~ 8 1

Privalle and colleagues [19] have suggested a sequence of steps culminating in the transport of cholesterol into the mitochondria of steroidogenic tissues Stage 1 in- volves the binding of plasma LDL (HDL in rodents) to specific receptors, a process which is stimulated by ACTH in the adrenal due to an increase in the number of receptors [9] Receptor-mediated endocytosis of LDLs then occurs resulting in deposition of cholesterol ester in lipid droplets Stage 2 involves conversion of cho- lesterol ester to cholesterol, another process that is stimulated either by ACTH, via the c-AMP-dependent protein kinase, or by suppression of the acyl CoA-choles- terol acyltransferase (ACAT), which is needed for cholesterol ester synthesis (see Refs 2,9 for reviews) Stage 3 involves transport of the liberated cholesterol into adrenal mitochondria; this is also stimulated by ACTH and probably depends on cell architecture since, as indicated above, anti-microtubule and anti-microfilament treatments block this process [20] There is also evidence for the participation of sterol carrier protein(s) which seem to be present in many cells, including those of the adrenal cortex; such proteins may be involved in cholesterol transport from the cytosol [21,22] Stage 4 involves the intra-mitochondria1 transport of cholesterol, which occurs in high concentration in the outer mitochondria1 membrane but at much lower concentrations in the inner membrane [23] It is here that the side-chain cleavage (SCC) system resides (see below)

3.2 Intrarnitochondrial transport of cholesterol

In 1979, Simpson [24] postulated that the outer mitochondrial membrane was the site of action of a labile protein factor, necessary to facilitate the transport of cho- lesterol, and Privalle e t al [19] provided evidence to support the notion that trans- ference of cholesterol from the outer to the inner membrane required an agent that

is cycloheximide dependent When rats were ether-stressed in vivo and cholesterol SCC was deliberately inhibited, cholesterol accumulated in the adrenal mitochon- dria, most (90%) of this being associated with the inner membrane cytochrome P- 450,,, After administration of aminoglutethimide to rats to block SCC, there was

a two-fold increase in inner membrane cholesterol, while cycloheximide abolished this increase Thus, it appears that cholesterol accumulates in the inner mitochon- drial membrane as a result of stress and that transference from outer to inner mem- brane requires a protein factor

Pederson and co-workers (see Ref 25) have isolated a peptide of M , 2200, from ACTH-stimulated rat adrenals, which contained 15% of basic aminoacid residues The polar side-chain groups were thought t o alter membrane structure so that transference of cholesterol towards the cyt P-450,,, on the inner membrane would

be favoured

Phospholipids are also thought to be involved in cholesterol transference In- creases in both the degree of unsaturation of fatty acyl groups and length of fatty acyl chains of mitochondrial phospholipids are known to increase the rate of cho- lesterol transfer [25] Further, the concentrations of some phospholipids in the in- ner mitochondrial membrane of rat adrenals were shown to increase after ACTH stimulation and to be related to cholesterol SCC activity [25]

Kimura [25] has recently discussed the possibility that several factors, including phospholipids and Ca2+ ions, are involved in the hexagonal phase-mediated trans-

P - 4 5 0 CHOL - CHOL

inner

m e m b r a n e

CHOL

B i l a y e r Reversed ( L a ) Hexagonal phase (HII)

port of cholesterol into the inner mitochondria1 membrane, hexagonal phases being more favourable structures that lipid bilayers for effective solute permeability, without the assistance of translocating proteins [26] Consistent with this notion are the findings that, firstly, phospholipids with highly unsaturated fatty acyl chains prefer the hexagonal phase In this situation, the non-polar groups will be oriented towards the outside of the membrane As cholesterol approaches this non-polar do- main, free from the membrane surface, the rotatory action of the hexagonal phase clusters may transfer cholesterol from outside to inside Secondly, if the fusion of inner and outer membranes occurs in the hexagonal phase (Fig l ) , this would re- sult in transference of cholesterol in the outer membrane to the matrix side ACTH stimulation may also result in the production of factors which may cause alteration

of the bilipid to the hexagonal state In response to ACTH stimulation, increased activity of cholesterol ester hydrolase will result in more free cholesterol (see above); this will be bound by sterol-binding protein and delivered to the mitochondria Thus, the membrane-bound cholesterol content will increase and so also will the hexag- onal phase and, hence, cholesterol penetration Fig 2 summarizes the possible se- quence of events [25], some of which have been discussed here

4 Side-chain cleavage (SCC) of cholesterol

Once cholesterol is transferred to the inner mitochondria1 membrane of steroido- genic tissues such as adrenals, ovaries and testes, it encounters the enzyme system known as the cholesterol SCC system This probably comprises 20- and 22-hydrox- ylases and a C-20,22-lyase, all tightly bound to the inner face of the membrane and associated with a specific cytochrome P-450,,, In addition, molecular 0, is neces- sary together with NADPH reductase and non-haem iron sulphur protein, which are called adrenodoxin reductase and adrenodoxin, respectively, in the adrenal [24] (Fig 3 )

The mechanisms whereby cholesterol is converted to pregnenolone and a C, fragment, 4-methylpentanal, have been given detailed attention by researchers during the past three decades Earlier work in this field has been reviewed by Su- limovici and Boyd [27] and more recent developments by Mitani [28] and Gower [29] Reaction mechanisms have been proposed which involve, as intermediates, hybrids of ionic and free-radical species [3O], hydroperoxides [31] or epoxides [32-341 Other evidence is consistent with a ‘sequential hydroxylation’ pathway, by which cholesterol is converted first by 22-hydroxylation to 22R-hydroxycholesterol, then to 20R,22R-dihydroxycholesterol by 20-hydroxylation and, finally, to preg- nenolone by means of the C-20,22-1yase reaction (Fig 4) In common with other

‘mixed function’ oxidases, the cholesterol side-chain cleavage (SCC) system re- quires NADPH and O,, and Shikita and Hall [35] determined stoichiometric rela- tionships between the oxidation of NADPH, 0, consumption and pregnenolone formation For cholesterol, 20s-hydroxycholesterol and 20R,22R-dihydroxycholes- terol, these ratios were 3:3:1, 2:2:1 and l:l:l, respectively Using a purified cyto- chrome P-45OsCc, Orme-Johnson et al [36] measured the dissociation

Cholesterol 22-hydroxycholestero1 20, 22-6 I hydroxycholesterol

Two studies have provided evidence for this scheme (Fig 4) Teicher et al [37] used a purified bovine cytochrome P -450,,, to confirm that hydroxylated cholester- 01s are intermediates, whereas Hume and Boyd [38] utilized an adrenal P-450,,,, with labelled cholesterol as substrate The oxidised P-450,,-cholesterol complex was reduced chemically under anaerobic conditions and then re-oxidised with the non- haem iron sulphur compound, adrenodoxin [24] (see Fig 3) Further reduction-ox- idation cycles were accomplished as shown in Fig 5 The first cycle resulted in the formation of 22-hydroxycholesterol as the major product but further oxygenation resulted in 20,22-dihydroxycholesterol and, finally, in pregnenolone Three com- plete oxygenationireduction cycles were therefore suggested as being necessary for cholesterol SCC to pregnenolone

The nature of the C, side-chain fragment produced depends, to some extent, on the tissue utilised and the conditions under which the experiments are carried out, but it is generally accepted that 4-methyl pentanal is formed first and that this may

be oxidised to the corresponding acid or, alternatively, reduced to 4-methyl pen- tanol [29]

Despite suggestions that the side-chain of cholesterol may be cleaved completely

in adrenal preparations [39,40], further work indicated that little or none of the C, ,

No reduction/oxygenation cycles

Fig 5 Pattern of product formation during single turnover cycles of anaerobic reduction/oxygenation

of the cytochrorne P-450,,, - cholesterol - adrenodoxin complex T h e results are expressed as a per- centage conversion of the total [“C]cholesterol added to the incubation Cholesterol, 0 ; 22-hydroxy- cholesterol, 0 ; 20,22-dihydroxycholesterol A ; pregnenolone, (from Ref 38, with permission)

steroid, dehydroepiandrosterone (DHA) plus the C, fragment, 2-methyl-6-heptan- one, is formed [41] However, in testis and ovarian preparations, some 30% of the total SCC fragments was 2-methyl-6-heptanone, indicating that an alternative path- way from cholesterol to D H A may occur in these tissues (see Ref 29 for a critical assessment) The ‘sequential hydroxylation’ proposals explained above have been criticized by Lieberman and his colleagues [42,43] in a ‘heuristic’ proposal for ster- oidogenic processes In a series of papers, Lieberman’s group has presented evi- dence for the true intermediates in the cholesterol SCC reaction being transient, enzyme-bound complexes Further, there may be at least two different enzyme sys- tems that catalyse the SCC of cholesterol, cholesterol sulphate and cholesterol ace- tate in bovine adrenal mitochondria1 preparations [44] Greenfield et al [45] have purified P-450,,, (which was free of adrenodoxin and adrenodoxin reductases) and have shown monophasic binding with cholesterol, cholesterol sulphate and choles- terol acetate, with dissociation constants of 1.1, 2.6 and 1.4 pmol/l, respectively Finally, in this section, mention must be made of the specificity of the SCC re-

action C,, to C,, sterols, all with saturated side-chains, underwent cleavage in rat and bovine adrenals and porcine testis, at much the same rate as cholesterol itself; however, sterols with polar side-chains, e.g 24-, 25- or 26-hydroxycholesterol, were cleaved at higher rates [46] The 5-cholesten-3P-ol structure seems to be a neces- sary requirement for the substrate for SCC [47], but the more polar sterols may en- ter mitochondria more readily than cholesterol itself and bind to P-450,,, which, in

the bovine adrenal, at least, is synthesized as a larger precursor and cleaved pro- teolytically either before, or on, insertion into the mitochondria [48]

As indicated in Fig 6, pregnenolone and progesterone are precursors of the cor- ticosteroids In some species, such as the rat, progesterone appears to be sequen- tially hydroxylated at C-17, 21 and l l p , whereas in the human and rabbit adrenal cortex, pregnenolone gives rise largely (but not exclusively) to the 17-oxygenated corticosteroids, e.g., cortisol, while progesterone gives rise largely (but again not exclusively) to the 17-deoxy-corticoids such as corticosterone (for reviews, see Refs 29,49) Since the 17- and 21-hydroxylases are microsomal enzymes, pregnenolone must pass from its site of synthesis in the mitochondria to the endoplasmic reticu- lum (ER), but the mitochondrial membranes do not present a significant barrier to pregnenolone efflux [50] The conversion of pregnenolone to progesterone and of 17-hydroxypregnenolone to 17-hydroxyprogesterone also occur in the smooth ER

by means of the 5-ene-3~-hydroxysteroid/3-oxosteroid-4,5-isomerase (5-ene-3p- HSD/4,5-isomerase) system Thereafter, 21-hydroxylation in the smooth E R of 17- hydroxyprogesterone leads to 11-deoxycortisol, while that of progesterone leads to DOC Conversion to cortisol or corticosterone, however, can only take place in the adrenal mitochondria because the required 1 I@-hydroxylase resides there and thus necessitates the transference of the precursors back through the mitochondrial membranes Although these various hydroxylases, including the 18-hydroxylase for

corticosterone (Fig 6), occur in all three histologically-defined zones of the cortex,

the enzyme needed for aldosterone synthesis, a presumed 18-HSD, occurs in the

Cholesterol - - L Pregnenolone a 17-hydroxypregnenolone

11-deoxycortisol Aldosterone

Cortisol

Fig 6 Biosynthetic pathways for corticosteroids 11 indicates position of hydroxylation; HSD, hydroxy-

zona glomerulosa (ZG) Further, both 18-hydroxylase and 18-HSD have been found associated with the inner mitochondria1 membrane [51]

5.1 Enzymes involved in corticosteroid biosynthesis

A great deal of information is now available about the properties, constitution and clinical manifestations of deficiencies of the hydroxylases involved in corticosteroid synthesis [42,52,53], and only a fraction of that information can be mentioned here All the hydroxylases are ‘mixed-function’ oxidases, requiring NADPH and 02, and some seem to be associated with cyts P-450, viz cyt P-450,,,, cyt P-45021, cyt P- 450,, The following equation represents the reaction catalysed:

R-H + NADPH + H + + 0 2 + R-OH + NADP’ + H 2 0

Lieberman et al [42] have criticized the kind of classical pathway diagram in Fig

6 as giving a simplistic view of events In this regard, evidence obtained during the

past 15 years has indicated that the situation is indeed more complicated than was thought earlier [42,52] Some of this evidence has been reviewed and suggests, for example, that there are at least two 21-hydroxylases Kominani’s group [54-561 have purified a cyt P-450 from bovine adrenal microsomes which catalysed 21-hydroxy- lation of 17-hydroxyprogesterone, 21 -deoxycortisol and 11 Phydroxyprogesterone, the products being, respectively, 11-deoxycortisol, cortisol and corticosterone On further purification the cyt P-450 was shown to be immunologically distinct from cyt P-450,,,, cyt P-45011, and cyt P-450,, When a cyt P-450 with 21-hydroxylating ac- tivity for progesterone and 17-hydroxyprogesterone was mixed with cyt P-450 re- ductase, the 21-hydroxylating activity for pregnenolone and 17-hydroxypregneno- lone was lost, although 17-hydroxylase and C-17,20-lyase activities could be reconstituted Other data showed that progressive purification of bovine adrenal 21- hydroxylase caused it to lose 21-hydroxylating ability for pregnenolone and pro- gesterone, while retaining that for 17-hydroxyprogesterone [57] This evidence, to- gether with other biochemical data of Kahnt and Neher [%], is consistent with the notion that there are at least two 21-hydroxylases In rat adrenals, experiments uti- lizing 21-hydroxylase inhibitors (see Ref 52 for details) indicated that, even though cortisol was formed normally, corticosterone synthesis was inhibited The sugges- tion made was that there might be one 21-hydroxylase for 17-hydroxy- and another for 17-deoxy-corticosteroids

A similar conclusion has been drawn by New and Levine [53] and may help to

explain the known clinical features of the 21-hydroxylase defect of congenital ad- renal hyperplasia, i.e., the existence of the simple virilizing form and the salt-losing type It has been suggested that the 21-hydroxylase activity is impaired in the ZF for both 17-hydroxy- and 17-deoxycorticosteroid pathways, so that ll-deoxycorti- sol levels (and also cortisol levels) are decreased (Fig 7), and the build-up of excess

1 8 0 H B Aldosterone

17-hydroxyprogesterone results in excessive androgen production In the ‘salt-los-

ing’ type, however, the 21-hydroxylase defect is in the ZG so that the conversion

of progesterone into DOC is blocked, resulting in aldosterone deficiency (hence the name of this defect, Fig 7) Further, Kuhule et al [59] have suggested that the 21- hydroxylases of Z F and ZG are different, the former acting on both 17-deoxy and 17-hydroxy substrates, the latter perhaps being specific for 17-deoxy substrates, such

as progesterone,

5.2 l l p and 18-hydroxylases

There are also distinct possibilities that at least two forms of the adrenal mitochon- drial llphydroxylase may exist, one in the Z G , involved in the conversion of D O C into aldosterone (Fig 6) and another in the ZF/ZR, concerned with the conversion

of D O C t o cortisol and 4-androstenedione to llphydroxy-4-androstenedione An alternative possibility is that several cyts P-450,,, may exist, which catalyse the llp hydroxylation of DOC, 11-deoxycortisol and 4-androstenedione [60,61] To make

Fig 8 Suggested mechanism for aldosterone biosynthesis M-Enz represents a postulated metallo-en-

zyme (from Ref 66, with permission)

the situation more complicated, it is not at all clear if there is a single cyt P-450 which possesses both 11P- and 18-hydroxylating abilities Bjorkem and Kalmer [62] reconstituted enzyme systems from rat and bovine adrenals and studied their 18- hydroxylating ability with D O C as substrate It was found that the cyt P-450 in- volved was indistinguishable from that required for 11P-hydroxylation

In contrast, however, other results [63] are consistent with specific cyts P-450 from bovine mitochondria - one P-450,,,, the other P-450,, Cheng et al [64] found that 18-hydroxylation of corticosterone was inhibited by canrenone and other drugs to

a greater extent than that of DOC, and suggested that two cyt P-4501, species might

be involved Various other pieces of evidence for the possible existence of isozymes

of cyt P-450,, or for a single P-450 having multiple functions have been reviewed

a pathway reIOlAquires oxidation of the -CH,OH group at C-18 by a presumed 18- HSD, but there is no real evidence for this enzyme [65] As an answer to this prob-

lem, a mechanism has been suggested [66,67] in which two successive hydroxyla- tions at C-18 of corticosterone occur followed by spontaneous dehydration (Fig 8)

On the basis of other data, several other pathways for aldosterone biosynthesis

18- hydroxyprogesterone Progesterone 11 /?- hydroxyprogesterone

6 Biosynthesis of the androgens

As a result of a wealth of experiments on androgen-producing tissues in numerous species [68,69], two pathways for testosterone synthesis are recognized (Fig 10) After the formation of pregnenolone from cholesterol in the mitochondria, testos- terone synthesis occurs in the endoplasmic reticulum from pregnenolone and pro- gesterone According to the classical view, 17-hydroxylation of these (& precursors occurs, after which SCC by a C-17,20-lyase results in dehydroepiandrosterone (DHA) from 17-hydroxypregnenolone or 4-adrostenedione from 17-hydroxypro- gesterone The action of 17P-HSD on D H A provides 5-androstene-3pJ7pdio1, which is known to be a good precursor of testosterone, through the action of 5-ene-

Fig 10 Pathways of androgen biosynthesis in rat testis A + B + C and a + b + c are the A5 and A‘

pathways, respectively for testosterone biosynthesis Enzymes A.a 17-hydroxylase; B.b C-17,ZO-lyase; C.c, 17PHSD Reaction c is reversible

3p-HSD/4,5-isomerase activity [70]; the action of 17P-HSD on 4-androstenedione also provides testosterone Inspection of Fig 10 shows the two pathways, one in-

volving 5-ene-3P-hydroxysteroids (sometimes called A’), the other involving 4-en-

3-oxosteroids (sometimes called the A4 pathway) It will be noted further that there

are transitions from As to A4 at different ‘levels’ via the 5-ene-3P-HSDiisomerase

enzyme system, but the reverse reactions seem to occur to only a very limited ex- tent [52]

There is considerable species variation with respect to the predominance of one pathway or the other For example, in the rat and mouse testis the 4-en-3-0x0 path- way seems to predominate, whereas in human testis the 5-ene-3P-pathway is quan- titatively more important One reason for the latter situation is probably the fact that 16a-hydroxyprogesterone (also produced in human testis) inhibits the SCC of 17-hydroxyprogesterone more effectively than that of 17-hydroxypregnenolone [71]

In further experiments (see Ref 52), each substrate inhibited competitively the lyase

for the other, with inhibition constant ( K , ) of 19 pmolil for 17-hydroxyprogesterone

and 60 pmolil for 17-hydroxypregnenolone Testosterone inhibited the SCC of 17- hydroxyprogesterone competitively but uncompetitively for that of 17-hydroxy- pregnenolone These data could be taken to indicate that there are two C-17,20- lyases for the two 17-hydroxysteroids or alternatively, that a single lyase possesses

different active sites, but with very similar properties, for the two substrates Whether 17-hydroxylation of the CZl steroid precursors is a pre-requisite for C- 17,20-lyase action has been a point of discussion for many years Some results sug- gested that 17-hydroxyprogesterone might not be an obligatory intermediate in the conversion of progesterone into testosterone and several studies in the author's lab- oratory have indicated that 16-unsaturated C,, steroids can be formed in boar testis from pregnenolone or progesterone without their prior 17-hydroxylation [68] These

results have recently been confirmed [72-741, although in immature porcine testis, 17-hydroxypregnenolone appears to be necessary as an intermediate in 16-andros- tene formation [74]

Lieberman et al [42] favour the view that the intermediates in testosterone bio- synthesis may be enzyme-bound and not, therefore, readily isolable, and these workers have reviewed the evidence that is consistent with such a notion [42] Chasalow [75] followed up earlier work and has confirmed that 4-androstenedione

is formed preferentially in rat testis from progesterone rather than from 17-hy- droxyprogesterone Other data [76] were consistent with the latter not being an in- termediate Earlier work [77], using incubations of boar testis with 17-hydroxy[I4C] progesterone plus ['Hlprogesterone, showed that the 17-hydroxy-derivative was the preferred substrate for labelled testosterone and 4-androstenedione Progesterone, itself, gave rise to some testosterone but very largely to 'H-labelled 16-androstenes

6.1 Action and properties of 17-hydroxylase and C-l7,20-lyase

Both these enzymes are associated largely with the smooth endoplasmic reticulum, the distribution approximating that of cytochrome P-450 In addition to the P-450 which probably catalyses both enzymic functions (see below), NADPH and O2 are needed; a review has been published [52]

The mechanism of 17-hydroxylation was studied by synthesizing pregnenolone and progesterone with a tritium atom specifically at C-17, and incubating with a bo- vine adrenal system The 17-hydroxypregnenolone or 17-hydroxyprogesterone formed did not contain significant quantities of 'H and indicated a direct and ster- eospecific substitution of the proton at C-17 by the hydroxyl group [78]:

17-['H]progesterone + H' + NADPH + 17-OH-progesterone + ['HIOH + NADP-

The 17-hydroxylase is found in the testis, adrenal and ovaries of many species but not in adrenal of mature rats or mice This latter finding is consistent with the fact that the 17-deoxycorticosteroid, corticosterone, predominates, little or no cortisol being formed

At one time it was thought that the hydroxylase and lyase were separate entities and, in keeping with this, rat testis microsomes were shown to contain cyt P-450 species that were distinct for lyase and 17-hydroxylase activities [79,80] In contrast

to these results, there is now evidence that both enzymic activities are linked, and that a single cyt P-450 is involved Such a P-450 has been isolated from neonatal porcine testis [81,82]; the purified enzyme had a M , of 59000 * 1000 and was shown

to be homogeneous, as judged by SDS-PAGE and immunochemical techniques The purified enzyme was shown to be a glycoprotein containing haem and phospholipid, the latter being necessary for activity [82] Both the 17-hydroxylation of progester- one and the SCC of 17-hydroxyprogesterone were catalysed when the cyt P-450 was

reconstituted with an appropriate cyt P-450 reductase However, the K,,, values dif- fered with respect to the two substrates, progesterone (1.5 pmol/l) and 17-hydroxy- progesterone (2.4 pmol/l) Further work [82] indicated that detergent treatment of the cyt P-450 increased the 17-hydroxylase activity in relation to the C-17,20-lyase activity If partial denaturation of the enzyme occurs through detergent treatment, then it is conceivable that the 17-hydroxylated C,, steroid intermediate may dis- sociate from the enzyme surface, whereas in the normal situation it would remain bound and not be readily isolable

Similar results to those of Nakajim et al [82] have been obtained using guinea- pig adrenals [83], from which a cyt P-450 can catalyse both the 17-hydroxylation of progesterone and the SCC of 17-hydroxyprogesterone

6.2 Conversion of 5-ene-3P-hydroxy- to 4-en-3-oxosteroids

The transitions at different levels of the ‘A” to ‘A4’ pathways have been alluded to above and are illustrated in Fig 10 Similar reactions must also occur in the for- mation of corticosteroids (Fig 6) Examples of such reactions are pregnenolone -+ progesterone (and their 17-hydroxylated derivatives), D H A + 4-androstenedione and 5-androstenediol + testosterone, and these transformations are catalysed by two enzymes which probably form part of a complex associated largely with the smooth endoplasmic reticulum, namely: a 5-ene-3P-hydroxysteroid dehydrogen- ase/3-oxosteroid-4,5-isomerase (3pHSDlisomerase) This enzyme system occurs in the testis (leydig cells, with lesser activity in the tubules), ovary (corpus luteum), adrenal and placenta but there is species variation with regard to the adrenal, e.g., human adrenal has rather low activity [29] The 3 p H S D requires N AD+ as cofac- tor (50% of activity is achieved with NADP’) and catalyses the oxidation of the 5-

ene-3phydroxysteroids to the corresponding 4-en-3-oxosteroids There is marked substrate specificity [52], with DH A being oxidized with the greatest ease and cho- lesterol hardly at all:

DHA > pregnenolone = 5-androstenediol > 17-hydroxypregnenolone >>>> cholesterol The intermediate 5-en-3-oxosteroids, such as 5-androstenedione (from DHA ) are further converted by isomerization to the 4-en-3-oxosteroids by means of the 4 3 - isomerase This enzyme, which requires no cofactor, is associated with the smooth

E R of adrenal, testis, ovary, liver and placenta It is relatively unstable, being in-

activated by freezing, even when pure A phospholipid environment appears to be

an important requirement since, when bovine adrenal microsomal preparations were treated with phospholipase A, 80-85% of phospholipids were hydrolysed with a

concomitant loss of 80-90% of enzymic activity [84] Restoration of activity was achieved by adding back to the lipid-depleted membranes aqueous dispersions of microsomal total lipid mixtures [84]

A great deal of research has been undertaken to determine if the dehydrogena- tiodisomerization reactions are properties of one system or of two separate en- zymes (see Ref 52); most of the evidence suggests that the former is true Probably three, or even four, substrate-specific isomerases may occur in bovine adrenal cor- tex which can act on C,,, C,, and C2, steroids Likewise, separate 5-ene-3pHSDs

may exist in the adrenal cortex for C,, and C,, steroids, because the latter did not compete with C,9 steroids for active sites of the enzymes studied

As for the mechanism of the isomerase reaction, Talalay (see Ref 52) used Pseu- domonas testosteronii as enzyme source and suggested that intramolecular transfer

of the 4 p p r o t o n occurs to the Gpposition of the steroid molecule The imidazole residues of histidine were suggested as playing an important role in the reaction, acting as alternate acceptors of the 4 P H and subsequent donors of the 6 P H Smith and Brooks [85] confirmed this intramolecular 4P to 6P-H transfer (Fig 11) Wein- traub et al [86] showed that a polar group at C-3 of the steroid substrates was nec- essary, binding possibly occurring through hydrogen-bonding with amino acid res- idues of the binding site In contrast, ring D was thought to be involved with hydrophobic binding to the enzyme A very precise fit of the steroid corresponding

t o C-11 was indicated because C-11 substituted steroids were not accepted as sub- strates whereas the region where the binding of ring A occurred was relatively open

6.3 Interconversion of 4-androstenedione and testosterone

This interconversion is catalaysed by 17phydroxysteroid dehydrogenase (17pHSD),

an enzyme generally found in the E R of numerous tissues such as adrenal, liver, testis, ovary and kidney Like many of the enzymes described above, there appear

to be different forms [52,87] For example, rat adrenal cytosol and E R contain sep- arate 17PHSDs, with NADH as the preferred cofactor The rat testicular enzyme, however, prefers NADPH Guinea-pig liver also contains two 17PHSDs, one sol- ubilized from cytosol, the other associated with the ER [88] These enzymes exhibit different activities towards C19 steroids, the cytosolic one preferring 5Preduced 17- oxosteroids and the microsomal counterpart being involved with 5a-reduced ste- roids, such as 5a-DHT In this case, the product of the reaction would be 5a-an- drostane-3,Il-l-dione

The porcine testicular 1 7 pHS D has been studied [89,90], and shown to be equally distributed between rough and smooth E R The apparent K , for testosterone was

122 pmol/l (and 40 pmolil for the purified enzyme)

6.4 Conversion of testosterone into 5a-dihydrotestosterone (Sa-DHT)

This conversion is catalysed by the 4-ene-5a-reductase, which has been studied in numerous tissues, including liver, testis, skin and pituitary In androgen-target tis- sues, such as prostate and seminal vesicles, the reductase is associated very largely with the nucleus, but microsomal counterparts also exist, usually in androgen-sen- sitive tissues [52,87,91] The enzyme has been solubilized and partially character- ized from the human and rat prostate [92,93], rat epididymis [94], rat liver [95] and porcine testis [96] The porcine testicular enzyme prefers NADPH as cofactor, only 40% activity being exhibited in the presence of NADH; the apparent K,,, was 0.6 pnolil [96] The single most effective solubilizing agent was sodium citrate [96], as shown also for the rat epididymal enzyme [94] Further metabolism of C,, steroids

involves conversion to androstanediols, reactions which are catalysed by 3 a@)-

HSDs These enzymes occur in numerous tissues and exhibit considerable hetero- geneity As reviews are available [52,91], this topic will not be discussed further here

7 Biosynthesis of oestrogens

Although it has long been known that C,, steroids, such as 4-androstenedione, give rise to oestrogens, the mechanism of this conversion has been the focus of intense study [52,97] In pre-menopausal women the major source of oestrogen are the ovaries but, in many species, the testes make a significant contribution The adre- nals seem only to produce small quantities However, it has been known for some years that, in post-menopausal women, most of the oestrogen formed is derived

mainly from plasma 4-androstenedione as a result of extra-glandular activity Adi- pose tissue and muscle are important in this respect as well as liver, kidney and hy- pothalamus

The first step in the conversion of 4-androstenedione to oestrone is the hydrox- ylation at C-19, a reaction associated with the ER and which requires NADPH and

02 It was thought earlier that the 19-hydroxy derivative was then converted to the 19-aldehyde, which gave rise to oestrone or oestradiol-l7p (from 4-androstenedi- one or testosterone, respectively) as rupture of the bond between C-10 and the an- gular methyl group at C-19 occurred through C-10,19-lyase action (Fig 12) More recent studies [42,52] have resulted in the proposal of at least three mechanisms, which all involve a second stereospecific hydroxylation at C-19 (requiring a second

DHA 17a -hyd roxypregnenolone 170-hydroxyprogesterone

mole each of NADPH and 0,) to produce a gem diol Mechanism (i) invokes the formation of an epoxide intermediate (19-dihydroxy-4p,5-o~ido-androstane-3,17-

dione) which can be aromatized subsequently [98] Mechanisms (ii) was suggested

on the basis of a 2p-hydroxylation and is consistent with 3 mol each of NADPH and

0, being required in the aromatization process [99,100] Once the 2P-hydroxy-19- 0x0-derivative of 4-androstenedione is formed (Fig 13), it decomposes with loss of hydrogen at C - l p [101,102] However, it should be noted that Caspi et al [lo31 were unable to show that the oxygen of the 2p-hydroxyl group was transferred t o for- mate, as required if the derivative was an obligatory intermediate Finally, Akhtar

et al [ 1041 proposed two possible mechanisms, each worthy of further study, which assume the dihydroxylation at C-19 of 4-androstenedione as step (i), followed by oxidation to the 19-aldehyde Thereafter, the first mechanistic proposal invokes 2p- (or lp-) hydroxylation followed by reaction with the 19-carbonyl group to yield hemi- acetals (Fig 14), which then result in oestrogen formation with loss of formate and rearrangement One oxygen atom from O2 molecule number 3 was shown to be in- corporated into the formate released Akhtar et al [lo41 suggest that formation of

a four-membered ring in XIa is unlikely, so leaving path l b of Fig 14 as one pos- sibility The alternative mechanisms invoke the formation of an intermediate en-

4 - arid r o s t en e d I one 19- h y d r o x y - 4 -androstenedlone

19,19'-d I hydroxy-4 -androstened lone ( h y d r a t e d f o r m of thel9-aldehyde derlvatlve)

Fig 13 Proposed mechanism for oestrogen biosynthesis involving double hydroxylation at C-19 and hy- droxylation at C-2 (see Ref 52)

OH

I OTCH

In the latter the arrows in structure XIV denote path b (from Ref 104, with permission)

zyme-bound ‘peroxide’ species (Fig 14) Aromatization is then envisaged, either through a Baeyer-Villiger type process (pathway a) or directly through a cyclic mechanism (pathway b) The former requires the intermediacy of the 10P-formyl derivative but this cannot be aromatized in human placental microsomes and thus effectively excludes this pathway as a viable mechanistic alternative

Further work by Stevenson et al [ 1051 has shown that 16~-hydroxytestosterone can be aromatized to oestriol via the 19-dihydroxy and 19-0x0 derivatives, these changes being identical to those indicated above [ 1041, in which an enzyme-per- oxide intermediate was postulated (Fig 14) Since the ‘aromatase’ system is known

to be catalysed by cytochrome P-450 [106], it is feasible that involvement of a P- 450-peroxide species could be envisaged, not only in the C-10,19 cleavage but also

in the preceding hydroxylations (Fig 15)

Fig 1.5 Suggested dual role of a cyt P-450 - peroxide species (10) in the hydroxylation and C-10.19 bond cleavagc steps in oestrogen biosynthesis (from Ref 105, with permission)

There is evidence [S2] that there are at least two forms of cyt P-450 involved with aromatization Likewise, there is evidence for different aromatases in human pla- centa which catalyse the production of oestrone and oestriol from 4-androstenedi- one and 16a-hydroxytestosterone, respectively Each enzyme system has been sub- fractionated into its own cyt P-450 and cyt P-450 reductase [107] This has been supported recently by Purohit and Oakey [ 1081, who measured aromatase activity for 16a-hydroxy-4-androstenedione and 4-androstenedione in the presence or ab- sence of the other substrate 4-Androstenedione competitively inhibited aromati- zation of the 16a-hydroxy derivative, with apparent K , essentially the same as its apparent K,, suggesting that both substrates bind and are aromatized independ- ently of each other The Iba-hydroxy derivative competitively inhibited the aro- matization of 4-androstenedione, thus presumably lowering the affinity of the aro- matase for the latter

8 Secretion of synthesized steroid hormones

Once the various steroids have been formed in paticular subcellular compartments, they must be released into the peripheral blood circulation There is evidence that some steroids are released by passive diffusion, as in the case of corticosterone, but for 18-hydroxylated corticosteroids, N a f / K t -ATPase activity is necessary [6,109] The situation is more complicated, however, because the presence of proteins in the adrenal cortex, which act as 'non-classical' receptors, may bind Czl steroids to different extents, thus reducing rates of steroid release (see Ref 6) So far as preg- nenolone is concerned, there is no barrier to its efflux from the mitochondria where

it is formed from cholesterol [SO] During incubation of rat testis [110], pregneno- lone was found to travel from the mitochondria, through the E R and cytosol and then out into the medium The release with time could be resolved into two com- ponents, one rapid and the second, much slower More than 25% of the pregnen- olone remained in the tissue after 150 min incubation This two-phase release may reflect the presence of two pools of steroid, the initial loss representing passive dif-

fusion and the slower phase being caused by pregnenolone binding to intracellular proteins [111,112]

Numerous other mechanisms, based on ultrastructural evidence, have been pro- posed [6] by which steroids may be secreted from their site(s) of synthesis The ste- roids may be contained in secretory organelles or in lysosomes, these acting as ve- hicles of transport to the plasma cell membrane, where secretion occurs by exocytosis

The foregoing discussion has attempted to trace the ways in which cholesterol, de- rived from plasma lipoproteins, is converted into the various steroid hormones and how these are secreted back into the blood Of necessity, many details have had to

be omitted but it is hoped that this ‘up-date’ has shown the complexities of steroid biosynthetic pathways and that earlier ‘classical’ ideas have had to be modified as greater knowledge of intermediates, isoenzymes and multiple forms of cyt P-450s has become available Perspectives for future studies are indeed exciting

Acknowledgements

Work performed in the author’s laboratory was supported by AFRC (grant nos

AG 35135 and 35144) to whom grateful thanks are expressed Mrs D.M Gower kindly prepared the manuscript for publication

Morris, M.D and Chaikoff, 1.L (1959) J Biol Chem 234, 1095

Brown, M.S., Kovanen, P.T and Goldstein J L (1979) Rec Progr Horm Res 35 215 Havcl, R J , Goldstein, J.L and Brown, M.S (1979) In: Metabolic Control and Disease (Bondy P.K and Roscnherg, L.E., eds.) p 393 W.B Saunders Philadelphia

Brown M S , Kovanen, P T and Goldstein J.L (1981) Science 212, 628

Gwynne, J T and Strauss, J.F.111 (1982) Endocr Rev 3, 299-329

Gower D.B (1984a) In: Biochemistry of Steroid Hormones 2nd Edn Ch 8 (Makin , H.L.J., e d )

pp 293-348 Blackwell Scientific Publications, Oxford

Jackson, R.L., Morisett, J D and Gotto A M (1976) Physiol Rev 56 259

Kane, J P (1977) In: Lipid Metabolism i n Mammals, Vol.1 (Snyder, F ed.) p 209 Plenum Press, New York

Faust, J R , Goldstein, J L and Brown M.S (1977) J Riol Chem 252, 4861

Schuler, I A , Scavo, L Kirsch, T M Flinkinger G.L and Strauss, J.F.111 (1979) J Biol Chem

254 8662

Ostlund, J E Jr., Pfleger, B and Schonleld, G (1979) J Clin Invest 63 75