Control of the Formation and the Action of Androgens in Peripherai, Tissues

Cailin Chen Docteur en Médecine Hunan Medical University

CONTROL OF THE FORMATION AND THE ACTION OF ANDROGENS IN PERIPHERAI, TISSUES

These présentée B la Faculté des études supérieures de 11Universit4Laval pour l'obtention du grade de Philosophiae Doctor (PhD.)

Départment de Physiologie (Endocrinologie moléculaire) FACULTÉ DE MÉDECINE UNIVERSITÉ LAVAL QUÉBEC

MAI 1996

8 Cailin Chen, 1996 Biblioth ue naaionale du Cana"ei a Acquisitions and Acquisitions et Bibliographie Services services bibliographiques 395 Wellington Street 395. me Wellington MEawaON KlA ON4 OrtewaOkl KtAON4 Caneda Canada

The author has granted a non- L'auteur a accordé une licence non exclusive licence ailowing the exclusive permettant a la National Library of Canada to Bibhotheque nationale du Canada de reproduce, loan, distn'bute or seii reproduire, prêter, distribuer ou copies of this thesis in microfonn, vendre des copies de cette thèse sous paper or electronic formats. la forme de microfiche/fdm, de reproduction sur papier ou sur format électronique.

The author retains ownership of the L'auteur conserve la propriété du copyright in this thesis. Neither the droit d'auteur qui protège cette thèse. thesis nor substantial extracts fiom it Ni la thèse ni des extraits substantiels may be printed or otherwise de celle-ci ne doivent être imprimés reproduced without the author's ou autrement reproduits sans son permission. autorisation. To: Xun, Chenchen, and My parents L'andtoghe actif 5a-dihydrotestost6rone (Dm joue un r61e prédominant tan physiologique important dans la peau et la prostate. Afin de mieux comprendrd les mécanismes impliqués dans la formation et l'action de la Dïïï, nous avon 4tu&4 la distribution tissulaire de certaines enzymes de la stéroïdogénèse chez 1r hamster et nous avons démontré la psesence des enzymes nécessaires h li biosynthbe de la DHT dans tous les tissus stdroidogéniques et une series dr tissus périphériques. De plus, nous avons dtudié les effets des précurseur: surrénaliens dehydroépiandrostérone (DHEA)et androsténédione (~4-dione administrés de faqon percutanée et systémique sur les param&tresandrogéno sensibles chez le hamster et nous avons démontré l'effet androgéniquc systémique de ces précurseurs surrénaliens. Enfin, dans le but d'améliore] l'efficacité de traitement des maladies androgéno-sensibles, nous avons vérifié li capacite de certains antiandrogénes et inhibiteurs de la Soc -réductase B bloquer 1; formation et l'action des androghes, et ce, chez le hamster ainsi que chez li souris porteuse de carcinomes mammaires Shionogi androgéno-sensibles.

Fernand Labrie, Director LONG SUMMARY

It is well recognized that in the human and other primates, androgens are synthesized from both the testicular and adrenal origins. The intracellula! formation of the active androgen Sa-dihyotestosterone (Dm) from the inactive adrem1 steroid precursor dehydroepiandrosterone sulfate (DHEAS) involves four enzymes; namely steroid sulfatase, 3 p-hydrox ys ter oid dehydrogenase/b4-ASisomerase (3$-HSD), 17&hydroxysteroid dehydrogenase (17p-HSD) and Sa-reductase. In order to better understand the rnechanisms involved in the steroidogenic process, we have studied the tissue distribution of the enzymatic activity of the abovementioned steroidogenic enzymes required for the synthesis of active androgens in both male and female hamsters. The activities of steroid sulfatase, 3$-HSD, 17p-HSD, and Sa-reductase were measured in al1 of the 14 tissues examined. We have also investigated the effects the adrenal precursors DHEA and d-dione administered systemically oz percutaneously by comparing their effects with those of the active androgens testosterone (T) and DHT in the orchiectomized hamster. We have demonstrated that al1 the compounds stimulated the size of flank organs, the size of the underlying sebaceous glands, and the weight of the prostate. However, their stimulatory potency is in deaeasing order starting with DHT, T, h4-diane, and DHEA, respectively. These results suggest that DHEA and a4- dione cm be transformed into active androgens in peripheral tissues in the hamster. In condusion, the steroidogenic enzymes, namely steroid sulfatase, 3& HSD, 17p-HSD, and Sa-reductase, which are required for the formation of active androgens, are present not only in the classical endocrine tissues such as the gonads and adrenals, but also in a series of peripheral intracrine tissues in the hamster. Therefore, the hamster provides a good animal model to investigate control of the formation and action of androgens in peripheral tissues.

Using the same animal model, we then investigated the antiandrogenic effects of the non-steroidal antiandrogen flutamide as well as the Sa-reductase inhibitor finasteride in male hamsters. These compounds were applied topically on the right flank organ and right ear. A four-week treatment decreased the size of the flank organ on both sides, as weli as the weights of the prostates and seminal vesicles, thus indicating that both flutamide and finasteride exert tbir effects systemical)y. Moreover#we have examinecl the antiandrogenic effects of a series of steroidal compounds synthesized in our medieinal chemistry division Arnong them, EM-250 (17p-alkynyl substituted testosterone derivative) wa! found to be a pure topical antiandrogen while EM-402, one of 17B-(N. alkylformamido)-substituted 4-methyl-4-azasteroids, was fond to be a poten topical Sa-reductase inhibitor.

Furthermore, we have studied the responsiveness of the growth of thc androgen-sensitive Shionogi tumor to androgen deprivation. In thi! experiment, we observeci that small tumors were highly sensitive to androger deprivation, while loss of responsiveness developed with increasing tumor size This indicates that, for optimal efficacy in the treatment of prostate cancer androgen blockade should be given at the early stage when the tumor has i small volume and is still organ-confined. Finally, we have investigated thi inhibitory effect of treatment with flutamide and finasteride, or with i combination of both compounds, on Shionogi tumor and prostate growth ii mice. Our data have demonstrated that an additive inhibitory effect can bc achieved with the combination treatment. These data suggest that sud combination therapy could be beneficial in the treatment of androgen-sensitivc diseases where optimal inhibition of DHT formation and action is desired. PREFACE

Although my graduate studies took a relatively short peroid of my life, the thought of ending this special and important stage is difficult. During my graduate studies, 1 acquired the ability to carry out independent research in the fields of EndOCIiI\C)Iogy and Phatmacology. This has been made possible through the generous help and guidance of my supervisor and colieagues. 1 shall never forget the people with whom 1 worked nor their cooperation and help during my study at the Research Center of Molecular Endocrinology at Laval University.

First, 1 wodd iike to express my deep gratitude and appreciation to my supervisor, Dr. Femand Labrie, for his expertise, time, patience and guidance throughout my Ph.D. training. With Dr. Labrie's personal instructions and support, I was able to step on the scientific research podium and accomplish my PhD. study .

1 am greatiy indebted to my jury members: Dr. Leone110 Cusan, Rudi Neri, Roland Tremblay and othet professors: Alain Belanger, Van Luu-The, Claude Labrie, Georges Pelletier, Richard Pouiin, and Jacques Simard for giving me the support and advice to enable me to complete this thesis.

Acknowledgments must also be given to al1 the In Vivo Study group (in alphabetical order): Drs. Jacques Couët, Michel Flamand, Jim Gourdon, Shengrnin Li, Ceiine Martel, Claude Trudel, Messieurs Yvan Labrie, Shouqi Luo, Andr4 Petitderc, Michel Rouleau, Alain St-Pierre, Milos Stojanovic, Mesdames Gina Gravel, Nancy Lemieux, France Létourneau, and Antigone Sourla for their ftiendly help and collaboration. 1 thank the Chemistry group for their timely provision of the compounds 1 studied, especially Drs. Shankar M Singh, and Yves Merand. 1 also thank Mesdames Louise Desy, France Lepire, Jacynthe Malenfant, Johanne Ouellet, Dr. Jean Côte, Messieurs Simon Caron, Roger Lachance, adGüles Leblanc for their skilled technical assistance.

1 would iike to thank othet graduate students, postdoctoral students and coîieagues, for thw cheerfui companionship and coilaboration in the laboratory. 1 appreciate the dynamism and quality of the work of Dr. Gilles Charpene Mesdames Aline Douville, Louise Doyon, Joyce Gardiner, Eiaine Leclerg José Pouiin, H4lène Rodrigue, Lise Thériault, and Marléne Walsh. 1 would like t thank the group of Medical Iiiustration and Photography for their quality an efficient work: Mesdames Anne Borgeat, Kathleen BUand Sylvia Kursteine: Messieurs Gilles Chabot, Marc Auger and Bertrand Vaillancourt. 1 also than Madame Carde Brault for her assistance in searching the iiteratures.

Finally, 1 wish to express my gratitude to my family, Dad, Mom, sisters, anl brothers for their never ending love and support through al1 these years c sdiool. 1 would iike to give my special thanks to Xun, my husband and my bei friend, for always being there; and to my precious littie boy Chenchen who is m inspiration to finish writing this thesis. TABLE OF CONTENTS

RÉSUMÉ COURT ...... LONG SUMMARY ...... +..:...... , TABLE OF CONTENTS ...... v LIST OF ABBREVIATIONS......

CHAFïER 1 INTRODUCTION ...... 1.0 Introduction-Intracrinology ...... 1.1 The formation of androgens ...... 1.1.1 Introduction ...... 1.1.2 Steroidogenic enzymes for andmgen formation ...... 1.1.2.1 3&hydroxystetoiddehydrogenase/~5-~4 isomerase (3B-HSD) ...... 1.1.22 l7B-hydroxysteroid dehydrogenase (l78-HçD)...... 1.1.2.3 5a-Reductase ...... 1.1.2.3.1 The properties of Sa-reductase...... 1.1.2.3.2 The distribution of Sa-reductase ...... 1.1.2.3.3 The physiology and pathology of 5a- reductase ...... 1.1.2.4 Steroid sulfatase and sulfotransferase...... 1.1.3 Classical steroidogenic tissues ...... 1.1.3.1 Adrenal cortex ...... 1.1.3.2 Testis ...... 1.1.4 Intracrine peripheral tissues ...... 1.1.4.1 Introduction...... 1.1.4.2 Skin...... i ...... 1.1.4.3 Prostate ...... 1.2 The action of androgens...... 1.2.1 Extracellular transportation of androgens...... 1 1.2.2 Androgen receptor ...... : 1.2.3 Androgen action and its regulation ...... : 13 Androgen-sensitive diseases ...... : 1.3.1 Acne ...... : 1.3.2 Hirsutism ...... : 1.3.3 Male pattern baldness ...... 1.3.4 Benign prostatic hyperplasia (BPH)...... : 13.5 Prostate cancer ...... 2 1.4 Inhibitors of androgen action ...... ; 1.4.1 Introduction ...... 1.4.1.1 For skin disorders ...... 2 1.4.1.2 For prostate cancer ...... 2 1.4.2 5a-Reductase inhibitors ...... c 1.4.2.1 Nonsteroidai inhibitors of 5a-reductase ...... c 1.4.2.2 Steroidai inhibitors of 5a-reductase...... r 1.4.3 Antiandrogens ...... 5 1.4.3.1 Non-steroidai antiandrogens...... 4 1.4.3.2 Steroidai antiandrogens ...... 4 1.5 Experimental models...... 4 1.5.1 Hamster ...... 4 1.5.1.1 Introduction ...... 4 1.5.1.2 Hamster flank organ ...... 4 1.5.1.3 Hamster ear ...... 4 1.5.2 Shionogi mammary carcinoma ...... 4

CHAPIER II THE IMPORTANCE OF LOCAL STEROID BIOSYNTHESIS IN THE HAMSTER ...... e 2.0 Introduction ...... ei 2.1 Distribution of steroidogenic enzymes in peripheral intracrine tissues in the hamster ...... e 2.2 Adrenal steroid precursors exert potent androgenic action in the hamster sebaceous glands of Bank organs and ears ...... € 2.3 DHEA administered percutanously exerts systemic effects in the hamster ...... 1C

CWTER III THE CONTROL OF ANDROGEN ACTION IN PERIPHERAL TISSUES IN THE HAMSTER ...... 13 3.0 Introduction ...... -13 3.1 Local and systemic reduction by topical finasteride or flutamide in hamster fiank organ size and enzyme activity ...... 13 3.2 A pure topical antiandrogem EM-250 ...... 17 3.3 Activity of 17 13-(N-alkyl /aryiformamido)- and 1713- [(N-alkyl/aryl) alkyl / ar ylamidol-4-me thyl-4-aza-3-0~0-5a-androstan-3-onesas Sa- reductase inhibitor in hamster Bank organ and ear ...... 20

CHAPTER IV THE CONTROL OF ANDROGEN ACTION IN SHIONOGI TUMORS ...... 23 4.0 Introduction ...... -23 4.1 Large shionogi tumors loose their responsiveness to androgen deprivation ...... 23 4.2 Additive in vivo growth inhibitory effects of flutamide and finasteride on androgen-sensitive Shionogi 115 carcinoma ...... 25

CHAPTER V CONCLUSION...... îû 5.0 Conclusioa ...... a

REFERENCES ...... 29

LIST OF ABBREVIATIONS

3a-di01 5a-androstane-3a,l7~ol 3B-di01 5a-androstane-3p,17p-diol SB-HSD 3$-hydroxysteroid dehydrogenase/~5-~4isomerase 4-MA N,N-die thyl-4-methyl-3-0x0-4-aza-Sa-androstam+one-l7f3- carboxamide 17p-HSD 17&hydroxysteroid dehydrogenase 170H-PREG 17a-hydroxypregnenolone 170H-PROG 17a-hydroxyprogesterone ABP Androgen binding protein ACTH Adrenocorticotropic hormone A-dione Sa-androstane-3,17-dione ADT Andros terone AR Androgen recep tor complex BPH Benign prostatic hyperplasia Cl9 19 carbones c21 21 carbones CBG Cotticosteroid binding globulin cDNA complementary deoxytiùonucleic acid CPA Cyproterone acetate A*-dione Andros tenedione AS-di01 Andros t-5-ene-3$,17&diol DHEA Dehydroepiandrosterone DHEA-S Dehybpiandrosterone-sulfak DHT Sa-Dehydzotes tosterone DNA Deoxyribonucleic aad DNase Deoxyribonuclease DOC Decixycorticos terone El Estrone El-S Estr one-s ulf ate E2 17bstradiol Epi-ADT Epiandrosterone FAR-1 7a Flank organ androgen regdated-l7a gene FIN Finasteride (MK-906) FLU Flutamide FSH Follicle-s tirnulating hormone ICso Concentration that inhibits 50% of test subjects LH Luteinizing hormone LHRH LH-releasing hormone MK-906 (Finasteride, Proscar, N-[l,l-dimethyIethyl]-3-oxo-4-aza-5a- androst-lene-17p-carboxamide) mRNA Messenger ribonucleic acid NADi Nicotinamide adenine dinudeotide (oxidized form) NADH Nicotinamide adenine dinucleotide (reduced form) NADP+ Nicotinamide adenine dinucleotide phosphate (oxidized form) NADPH Nicotinamide adenine dinudeotide phosphate (reduced forrn) ODC Ornithine decarboxylase ORCH ûrchiectomy Pl50 Cytochrome Paammatase Pm= Cytochmme Pasideduin deavage Pmcl1 Cytoduome PaIl-hydroxylase Pa91 CyWwome PaIl-hydroxyiase P450 17a Cytochrome P4s 17a-hydroxylase/ 17-20 1yase PBP Prostatic binding protein PH Logarithm of the hydrogen ion concentration PREG PregnenoLone PRL Prolactine PROG Pregesterone PS A Prostatic specific antigen RNA Ribonucleic acid RNas Ribonucleic nuclease SBP Sex hormone binding protein SH 115 Shionogi carcinoma 115 SHBG Sex hormone binding glubulin T Testosterone (testo) TLC Thin-layer chromatography Tris Tris(hydroxymethy)aminomethane tRNA Transfer ribonucleic acid

Miscellaneous

Centigrade Base pair Twice daily Curie Dalton Kiiobase Inhibitory constant Michaelis cons tant mg Milligram min Minute Miaogram Microiiter Micromolar Mol (mole/liter) Nanogram Subcutaneous Second CHAPTERI INTRODUCTION A discovery of major importance in the field of sex steroids and thei physiological and pathological roles, is that hurnans and some other primate are unique among animal species in having adrenals that secrete large amount of the inactive precursor steroids dehydroepiandrosterone -A), its sulphat, (DHEA-S),and androstenedione (Addione), which couid be converted intc potent androgens and estmgens in peripheral tissues (Adams, 1985; Labrie et al 1985). In fact, adrenal precutsors are responsible for 30-50% of androgens in me] and 75% of estrogens More menopause and almost 100% after menopause il women (Labrie, 1991b).

Despite the early description of circulating DHEA-S in the circulation (Baulieu 1960), its biological function has so far received little attention. It is knowi however, that adrenal secretion of DHEA and DHEA-S increases durinj adrenarche in children at the ages of 6-8 years, elevated values of circulatinj DHEA-S are maintained throughout addt life, and its circulating concentratio] in adult men and women is higher than any other steroids except cholesterol The term "prehormones" or "precursors of active hormones" has beei introduced to describe the transformation of adrenal steroids into activc hormones in the late 1960's (Baird et al., 1968). It wasn't until 1988, that the worc "intracrine" was first coined to describe the formation of the active androgen testosterone (T) and dihydrotestosterone (DHT)from the inactive adrena precursors DHEA and ~4dionein the orchiectomized rat mode1 (Labrie et al, 1988a). In such a system, locally produced androgens and/or estrogens exer their action inside the cells where their synthesis took place without beini released into the extraceliular space. This terminology is complementary to thc weU-known autoaine, paracrine and endocrine activities where a hormone act at the surface of the producing cells (autocrine activity), influences neighboriq cells (paracrine activity) or is first released into the circulation and, followiq transport, acts on distant target tissues (dassical endoaine activity). Since then the investigators in our laboratory have focused their research on intracrinolog] induding the elucidation of different types of rat and human steroidogenir enzymes (3B-HSD, 17B-HSD, and Sa-reductase), the demonstration of thc widespread tissue distribution of these enzymes, and the exploration of thc novel approab to treat horxnone-sensitive diseases. Since acne, hirsutism, male pattern baldness, benign prostatic hyperplasia @PH), and prostate cancer are hYJy androgen related disorders, the inhibition of the formation and the action of androgens may be a logical and effective way to treat these diseases. In addition to the classical steroidogenic tissues, namely, the testes, ovaries, and adrenals, a series of human peripheral tissues, such as the skin and the prostate, possess the enzymatic systems required for the formation of active androgens and estmgens from a relatively constant supply of precutsor steroids provided by the adrenals @.abrie, 1991b). In the past years, most therapeutic approiches were aimed to control the steroid formation by the classical steroidogenic tissues, for example, the gonadal sex steroids can be easily inhibited by the administration of CHRH agonist. It is now clear that we should focus in the field of intracrinology to better understand the physiological mechanisms conkoung local steroid formation. We will then be in a position to develop novel therapeutic approaches which take into account the high proportion of steroids made locally, and which are responsible for the growth and functions of normal and abnormal cells and tissues. The androgen associated diseases mentioned above are thus primary candidates for approaches based upon control of intracrine activity. Other than its therapeutic efficiency, an important advantage of endoaine (and even more, intracrine) therapy is the usual absence of serious side effects. The field of intrauinology provides new possibility in the development of therapeutically approaches aimed at producing more specific and potent inhibitors of peripheral steroidogenic enzymes, in addition to blocking their action at the receptor level with antiandrogens.

Based on the above information described, we are then interested in investigating the importance of adrenal precursors, the distribution of steroidogenic enzymes, and the inhibition of the formation and the action of androgens in peripheral tissues. Thus, the following chapter wiil describe certain basic knowledge concerning the field of intraainology: steroidogenesis including the enzymes involved; the classical endocrine tissues as well as some peripheral intracrine tissues; the androgenic actions and the anàrogen- dependent disorders; the inhibitors of androgenic actions; and finally, the animal models which were used in the researeh. The biosynthesis of steroid hormones in steroidogenic tissues begins with cholesterol as shown in Fig. 1 (Hechter et al., 1953). The major portion oi cholesterol used cornes from the blood. Its source is in either intestinal absorption or hepatic synthesis. Moreover, steroidogenic tissues synthesizi cholesterol de nmo from ethyl acetate (Hechter et al., 1953), or derive most 01 their cholesterol directly from low density lipoproteins (Brown et al., 1979, Gwynne and Strauss, 1982), or from high density lipoproteins (Halkerston et al., 1961; Gwynne and Strauss, 1982). Side-chain cleavage of diolesterol in mitochondria is the first step, involves a complex enzymatic system ol hydroxylation at C20 and C22, foUowed by the action of a desrnolase, and thua results in the formation of pregnenolone (PREG) (Halkerston et al., 1961)# These steps are catalyzed by cytochrome Pd50 side-chah cleavage enzyme (P4soscc). By the catalysis of 3B-hydroxy-5-ene steroid dehydrogenase/AS-~4 isomerase (3B-HSD), PREG is then converted to progesterone (PROG) which is further converted to Il-deoxycorticosterone (DOC) by cytochrome Pd50 21- hydroxylase (p450c21). DOC is transformed by cytochrome Pd50 Il-hydroxylase (P450cll) to corticosterone which will then be converted to aldosterone. in addition to this pathway, PREG can also be further transformed into 17a- hy droxypregnenolone (170H-PREG) by cytochrome P450 17a-hydroxyIase/17-20 lyase (p45o 17 a). Seventeen-OH-PREG is further converted to 1 7 a - hydroxypregesterone (170H-PROG) by 3P-HSD, and 17OH-PROG cari also be transformed into Il-deoxy-cortiso1 by P4soc2l or to ~4-dioneby Pd50 17a, respectively. SeventeenUH-PREG can also be transformed to DHEA by Pd50 17a. DHEA is the major precursor of sex steroids, and is further transformed inta andros t-5-ene-3B-l7~-diol (~S-diol)by 17p-HSD or ~4-dioneby 38-HSD, respectively (Ewald et al., 1965). Both AS-di01 and A*-dione can be transformed into testosterone O after the action of 3$-HSD or 17$-HÇD, respectively. ~4- Dione can aiso be transformed to estrone (El)by cytochrome Pd50 aromatase (* aro). bth T and El unbe converted into estradiol (Ez) due to &e action of Pd% aromatase or 17P-HSD, mpectively. T cm also be converted into Sa- dihydmtestosterone (Dm by Sa-ductase, and DHT, can then be further Figure 1. Biological pathway of steroid synthesis converted to androstane-3a,l7B-diol (3a-diol) os androstane-3$,17Bdiol (3P diol). Furthermote, DHEA and El are in a reversible equilibrium with thei sulfate, DHEA-S and El-SJ due to the functions of sulfotransferase aiid sulfatase.

In brief, large amounts of steroid precursor DHEA-S seaeting by adrenals i directly transformed into DHEA by sulfatase ,DHEA is converted to Acdione b, 3f3-HSD, and then to T by 17B-HSD. Finally T is transferred to most activ androgen DHT by Sa-reduatase in pdpheral tissues (Lébrie, 1991b; Labrie et al 1993). More detail discussions of the four enzymes mentioned above are give: in the following sections.

1.1.2 -vmes for a-n formation

The membrane-bound enzyme, 3P-HSD catalyses an essential step in th' transformation of al1 5-pregnen-3P-ol and 5-androsten-3P-01 steroids into th' corresponding ~4-3-keto-steroids,namely PROG as well as al1 the precursors a androgens, estrogens, glucocorticoids and mineralocorticoids. In fact, 3B-HSI catalyzes the conversion of PREG to PROG, 170H-PREG to 170H-PROG, DHEl to A4-dione, and AS-di01 to T. in addition, 3P-HSD is responsible for th( interconversion of 3b-hydroxy- and 3-ketoba-androstane steroids. 3P-HSD is , non-P450 enzyme and needs NAD+ as a cofactor.

The structures of the cDNAs and genes encoding two types of human 3$-HSI (type 1 and type II) have recently been elucidated (Luu-The et al., 1989b; Lachancc et al., 1990; Lachance et al., 1991; Rhéaume et al., 1991; Labrie et al., 1992b). 3P HSD is found not only in classicd steroidogenic tissues, namely the placenta aàrenal cortex, ovary and testis, but also in a series of peripheral tissues, such ai skin, adipose tissues, breast, lung, endometrium, prostate, Liver, kidney epididymis, and brain (Milewich et al., 1991; Simard et al., 1991a; Dumont et al. 1992b; Labne et al., 19921; Pelletier et ai., 1992). The type II 3$-HSD is expressec rnainly in the adtends and gonads while type 1 3P-HSD is preferentiall~ expressed in peripheral intracrine tissues. The human 3$-HSD genei cortesponding to human cDNAs type 1 and type II contain four exons and threc introns within a total length of 7.7-7.8 Kbp. These genes were located by in sifi hybridization at the pll-pl3 region of chromosome 1(Bérubé et al., 1989).

Multiple 3B-HSD forms have been characterized in the rat and mouse (Zhao e al., 1990; Bain et al., 1991; Zhao et al., 1991). The structures of four types of rat 3P MD cDNAs whida al1 encoding a 372-amino aad protein have been elucidatec (Zhao et al., 1990; Naville et al., 1991; Simard et al., 1991b; Zhao et al., 1991 Simard et al., 1993a). The type 1and II 3pHSD proteins expressed in the adrenal ovary, testis, kidney, and adipose tissue she93.8% homology. The type ïIï 38' HSD, mainly expressed in the liver, does not display the 3$-HSD activity witl As-hydroxysteroid precurson but is rather a bketosteroid reductase usin1 NADPH as a cofactor, which catalyzes the conversion of 3-keto-satqratec steroids such as Dmand dihydroprogesterone into their corresponding 3P hydroxy metabolites. Furthermore, type IV 3$-HSD is the exclusive forn expressed in the skin. The different types have different Km values for the samc substrate, however, eadr type of the enzyme can use either PREG or DHEA as i substrate with sidar Km values (tabrie et al., 1992a). Rat type 1 3P-HSD ah shows 17$-HSD activity with Sa-andrortane steroids but not with EU El, A44 &one, or T (de Launoit et al., 1992).

The widespread distribution of 3P-HSD expression indicates that this enzyme i: likely to play an important role in the intracrine formation of sex steroids ir peripheral target tissues. Moreover, it has been demonstrated that the patient: with the disease of dassical3p-HSD deficiency is due to point mutation(s) in th human type II 3b-HSD gene while the human type 1 3B-HSD gene is apparentlj normal. Thus, the existence of a normal type 1 3$-HSD gene in these patient: should ensure peripherat steroidogenesis. In fact, the 3$-HSD defiaent patient! who survive the early postnatal period marked by severe deficiency ir glucocorticoids and mineralocorticoids, showed normal adrenarche, hail growth, and secondary sexual characteristics induding, in some cases, fertilit) (Rhéaume et al., 1992; Simard et al., 1993b; Sirnard et al., 1994).

The enzymes of the 17-D gene family are responsible for the interconversions of DHEA and AS-diol, ~4-dioneand testa, estrone (El) and estradiol(E~),A-dione and DHT, ADT and 3a-diol as well as epi-ADT and 38- diol. 17B-HSD is the only steroidogenic enzyme which catalyzes a reversible step, namely the oxidation and the reduction reactions. The reduction reaction needs NADH or NADPH as a cofactor while the oxidation reaction needs NAD+ or NADP+ as a cofactor.

The structure of a human placental 17p-HSD cDNA and its corresponding gene has been isolated and characterized (Peltoketo et al., 1988; Luu-The et al., 1989a; Luu-The et al., 1990; Peltoketo et al, 1992). The gene encodes a predicted protein of 327 amino acids. This enzyme, designated as type 1 17kHSD, has been found to be a member of the short-chah alcohol dehydrogenase superfdy (Baker, 1991; Ktozowski, 1992) and was localized on the qll-q12 region of chromosome 17. The type 1 17b-HSD gene consists of six exons and five introns within a genornic DNA fragment of 3.3 kb pairs. The recombinant enzyme encoded by the type 1 17b-HSD gene, which functions in dimeric form; catalyzes the interconversion of El and E2 and, to a much lower extent, DHEA and AS-di01 (Dumont et al., 1992a). The type 1 17p-HSD gene was assigned to the 17qll-q21 region by in situ hybridization (Luu-The et al., 1989a). More recently, the structure of another type of 17B-HSD cDNA was repotted (Wu et al., 1993) which encodes a predicted protein of 387 amino acids with a molecular weight of 42,782, which is most likely associated with the membranes of the endoplasmic reticulurn. This enzyme, designated as type ïï 17P-HSD, catalyzes the interconversion of El and E% T and ~4-dione,DHT and androstanedione, as well as 20a-dihydroprogesterone and PROG. It is also a member of the short-chain alcohol dehydrogenase superfamily and shares approximately 20% sequence identity with the cytoplasmic enzyme encoded by the type 1 17PHçD gene (Luu- The et al., 1989a). The type II 17B-HSD gene contains five exons and is endosed within a genomic DNA fragment of > 40 kbp (Labrie et al., 1994), and is located on the 16q24.1-24.2 region (Durocher et al., 1994).

Seventeen p-HSD is widely expressed and active in human tissues. The presence of type 1 IV-IIÇD mRNA has been shown by ribonudease protection assay in the RNA extracted from placenta, breast ovary, endometrium, ZR-75-1 cells, LNCaP prostate cancer cells, adipose tissue, skin and prostate (Luu-nie et al., 198%; Marte1 et al., 1992). The enzymatic activity studies have demomtrated that 17b-HSD activity i; presented in 17 tissues examined in rat, 15 tissues examined in human (Martel et al., 1992), and 25 tissues examined in rhesus monkey (Martel et al., 1994).

1.1.2.3.1 The properties of Sa-reducfase

The Sa-reductase (NADPH: ~4-3acidoreductase)is a membrane bond enzyme and catalyzes irreversibly the conversion of ~4-dione,T, and other 4+ne+keto- steroids to the corresponding Sa-dihydro-3-keto-steroids in the psesence of its cofactor NADPH. The best known role of this enzyme is the transformation of T into DHT, the most potent androgen responsibIe for the differentiation of the male external genitalia and prostate as well as virilization at puberty in the extranuclear cornpartment of target cells (Anderson and Liao, 1968; Bruchovsky, 1972; Bruchovsky and Meakin, 1973b). Sa-Reductase is a hydrophobic microsomal protein located on the endoplasmic reticulum and nuclear membrane.

Two types of human steroid Sa-reductase, chronoIogically identified as type 1 and type 11, have.been isolated from a human prostate cDNA library (Andersson and Russel, 1990; Andersson et al., 1991). The structure of the human type 1 Sa- reductase gene has been elucidated in 1991 (Jenkins et al., 1991), however, it appears that this gene is not responsible for Sa-reductase deficiency and is not the major mRNA speues expressed in the .huma.prostate. In fact, type 1 Sa- reductase is the predominant form expressed in human shwhile type II is the major form detectable in the prostate (Andersson and Russel, 1990; Andersson et al., 1991; Labrie et al., 1992c; Luu-The et al., 1994). LateIy, the structure of type II gene has also been elucidated (Labrie et al., 1992~).The type II Sa-reductase gene contains five exons and fout introns and shows splicing sites identical to those of the type 1 gene. The encoding region of the type II Sa-reductase gene shares 57% homology with that of type 1 Sa-reductase gene. The enzymes have never been purifieci by any sources due to the unstable nature of the enzymes during purification, and the absolute structures have not been deterrnined yet (Harris et al., 1992). It is known that the Sa-reductase activity measured in the same tissue fron: various speaes is quite different and is changeable along with the pH, however the optimum pHs for most of Wozymes of Sa-reductase is in the range of pH 5.E to 9.0. For instance, the optimal pHs of Sa-reductase in the nuclear anc miaosomal fractions of rat seminal vesicles were around 5.7 to 5.9 (Suzuki anc Tomaoki, 1974), whiie the human scalp Sa-reductase had a broad pH optirnu centered at pH 7.0 (Harris et al., 1992). Furthermore, two distinct human skir Sar-reductases with peaks of activities at either pH 5.5 or pH 7-9 were describeC (Moore and Wiison, 1976). The pH 5.5 form was found only in fibroblasts fron genital ski.whereas the pH 7-9 form was presented in all fibroblasts assayed. A more detail study was the comparison of the prostatic Sa-reductases in th4 homogenates (particulate preparation, 5-10 mg protein/ml) of rat, dog, and human prostates (Liang et al., 1985). The optimum pHs for the enzymes art significantly different for these three species. Rat prostatic Sa-reductase had a broad optimum around pH 7, and the dog enzyme had a broad optimum around pH 6, while the human enzyme had a sharp optimum at pH 5' Recently, as successed in the isolations of cDNAs of human type 1 and II Sa. reductase isomases, the optimum pHs of two isozymes were sequentiall~ investigated. Human type 1 Sa-reductase had a broad optimum pH 6-8.5, whii~ type II had a sharp optimum at pH 5.0 (Andersson et al., 1991; Normington and Russell, 1992).

1.1.2.3.2 The distribution of Sa-reductase

After the incubation of T or A*-dione with rabbit liver homogenate, a series oi Sa-steroids were obtained. It was the first time able to identify the metabolites by paper chromatograph (Kocldcian and Stidworthy, 1954). Two years later, DH'ï was first identified as the major metabolite of the incubation of T with rat liver homogenate (Rubin and Dorfman, 1965). These two independent experiments indicated that the Sa-reductase activity is presented in liver tissues of rabbits and rats. Since then, the Sa-reductase activity has been observed in various organs and tissues. That homogenates or miaosomes of mouse (Jagarinec et al., 1967) and human livers (Lisboa et al., 1965) presented potent enzyme activity have also been reporteci. Later studies have demonstrated that hurnan (wilson and Waîker, 1969), hamster (Voigt and Hsia, 1973b; Takayasu and Itami, 1982), and rat (Monsalve and Blaquier, 1971) skin is another abundant Su-reductase containhg tissues; furtherrnore, human (Fansworth and Brown, 1963), rai (Anderson and Liao, 1968), and dog (Gloyna et al., 1970) prostates, as weli as rai (Bruchovsky and Wilson, 1968; Suzdci and Tomaoki, 1974) and dog (GloyM ei al., 1970) seminal vesides, are the most kr-reductase active organs. Other orgam and tissues also showed noticeable enzyme activity: rabbit, rat, and humar blood; rat and human neural tissues; Pabbit, dog, rats, and human, skeletal muscles; dog, rnouse, and rat submwllary glands; rat and chid bones; rat and hurnan oral mucus; and the intestinal mucus and urinary bladder of dog: (Koddcian and Arimasa, 1976).

Skin: The slices of human skin from several different sites (such as labia, majora, scrotum, prepuce, and clitoris) were first tirne observed to be able tc convert T into DHT in very high rates (Wilson and Walker, 1965). The distribution of androgenic inetabolites in human skin has been comprehensively studied through using tissues isolated either by direci dissection of axillary skin or by dissection of colIagenase-digested forehead and axillary skin. AU sliced skin tissues (epidermis, sweat glands, sebaceous glands, hair foliicles and dermis) of males and females were fould to have Sa-reductase activity (Hay and Hodgins, 1978). In forehead skin 40-66% of Sa-reductase activity was presented in the sebaceous glands, while in the axilla 50-70% of the activity was concentrated in the sweat gl&ds (Hay and Hodgins, 1978). The Sa- reductase activity was also described in human neonatal foreskin (Voigt et al., 1970), isolated human beard hair follicles (Fazekas and Lanthier, 1971), hamster flank organs (Takayasu and Adachi, 1970; Voigt and Hsia, 1973b), hurnan skin fibroblasts (Morimoto et al., 1991; Kaufman et al., 1993), and men scalp (Harris et ai., 1992).

-: -: The prostate is one of the most Sa-reductase active organ in mamrnalian. In the early study, the incubation of minced or slicesd tissues of human benign hyperplastic prostate with (4-l4C]-T or [1,23H]-T yielded major Sa-metabolites, such as DHT (35%), Sa-andastan-3,17-dione (Sadone, 7.7%) and Sa-andostan-3a017j3diol (Sa-Diol, 18%) (Fansworth and Brown, 1963; Ciorgi et d.0 197l), these results were conformed later by another study ushg more improved enzyme preparation method (Houston et al., 1985; Houston et il, 1987). Then it has been reported that the Sa-reductase activity presented in the rat prostate nudei (Anderson and Liao, 1968; Bruchovsky and Wilson, 1968). The enzyme activity was also found in immature, matute and hypertrophic dc prostates (Gloyna et ai., 1970), in the hypertrophic human prostates (Siiteri et a 1970), and in the rat prostate miaosomes (Liang et al., 1983).

Seminal The &st evidence of potent Sa-reductase activity presented : the seminal vesicles was obtained from the administration of [1,2-3HJ-T i normal and functionally hepatectomized male rats, as a result, high yield of 11, 3H]-DHT was formed in this tissue (Bru&msky and Wilson, 1968). Other i vitro studies with this organ comprising the works of using tissue slices (Gloyl et al., 1970), ceU-free homogenates (wilson and Walker, 1969) and subcelldi fractions such as nudear fraction (Wilson and Gloyna, 1970; Suzuki an Tamaoki, 1973), miaosomal fraction and cytosol fraction (Suzuki and TamaoE 1973).

1.1.2.3.3 The physiology and pathology of 5a- reductase

The normal activity of the Sa-reductase maintains the routinely T-mediate biological functions, such as anabolic actions (musde mass .increase, pen enlargement, scrotum enlargement and vocal cords enlargement) an spermatogenesis (male sex drive and performance) of human, as well as tk DHT-mediated effects, such as increased facial and body hair, acne, scale ha recession and prostate enlargement (Metcalf et al., 1990).

The absence of the Sa-reductase activity in human, that is, Sa-reductase geniti deficiency, provides a useful mode1 for studying the biological importance t this enzyme. The deficiency of Sa-reductase in males results in incornplel differentiation of the extemal genitalia at birth (Imperato-McGinley et al., 1974 At puberty, the patients have normal to elevated levels of T in plasma, an virilization occurs, but the prostate remains small and there is no acn (Imperato-McGinley et al., 1974; Walsh et a., 1974). Mal pseudohermaphroditism is also accompanied by low levels of DHT, inaease T/DHT ratio after hCG stimulation, deaease production of urinary Sa-reduce metabolites of'T, decreased urinary Sa-reduced metabolites of Cpl and Cl steroids other Uw T;inaeased plasma LH and FSH levelr, as well a dirnkished Sa-reductase ictivity in tissues (Imperato-McGinley et al., 197! Wilson, 1980). On the other hand, females with Sa-reductase deficiency appea to have no dinical symptorns (ïmperato-McGinley et al., 1974), although thc enzyme also metabolizes other ~4-3-oxosteroids, such as PROG anc cor ticos terone.

Abnormal high Sa-reductase activity in human results in excessively higl levels of DHï in peripheral tissues, which is implicated in the pathogenesis a ben@ prostatic hyperplasia (BPH), prostate cancer (Sansone and Reisner, 1971 Meikle et al., 1980), acne (Sansone and Reisner, 1971), female hirsutisn (Kuttenn et al., 1977), and male pattern baldness (Bingham and Shaw, 1973).

1.1.2.4 Steroid sul&tase and su

Many steroid hormones such as cholesterol, PREG, 17$-PREG, DHEA, and El an in reversible equilibrium with their sulfate esters, which are water soluble anc easily transportable. The steroid sulfates can be directly synthesized fron cholesterol sulfate, however, this is a minor pathway, while the major pax results from the sulfatetion of free steroids by sulfotransferase (ST) (Baulieu e al., 1965; Milier et al., 1981). The enzyme ST uses the sulfate contained in th( coenzyme PAPS (3-phosphate adenosine 5'-phospho-sulfate). Several forms O ST have been isolated and characterized from rodent, guinea pig, and bovin Livers or adrenal tissues (Ogura et al., 1990; Demyan et al., 1992; Oeda et al., 1992 Wilborn et al., 1992; Driscoll et al., 1993). These enzymes iepresent i heterogeneous family of isoenzymes with distinct, but overlapping, substrat( specificity. Three cDNAs encoding rat liver hydroxysteroid STs werc characterized and shown to have considerable sequence similarity (Ogura et alq 1990). Unlike the heterogeneous family of rat steroid STs, human DHEA-ST ha been showed to have oniy one form which has been purified from human live tissue (Falany et al., 1989). This enzyme can also catalyze most, if not dl, bil, acid sulfation activity in human liver cytosol (Radominska et al., 1990) Recently, the human adrenal .fotm of DHEA-ST was purified and shown to bc physiologicaily, immunologicaily, and kinetically similar to human live DHEA-ST (Corner et al., 1993). More recently, two cDNA clones encodinl human placenta1 and human liver estrogen ST, and the structure anc expression of the human placenta Sï gene have ahbeen repotted (Aksoy et al. 1994; Bemier et al., 1994). The steroid sulfates can be hydrolyzed into certain neutral steroids b microsome enzyme sulfatase. Considering the high levels of DHEA-S in th circulation in humans and other primates (Labrie, 1991b); the steroid sulfatas may play an important role in the formation of active androgens and estrogeri in the peripheral tissues. The enzyme has been detected occasionally in nor vertebrates and lower vertebrates, however, biochemical studies mainly hav been restricted ta the mammalian steroid sulfatases. in mammals, steroi sulfatase activity is found in almost al1 tissues as a highiy insoluble enzp associated mainly with membranes of the endoplasmic reticulum. It is not dei yet if the same steroid sulfatase can hydrolyze al1 the sulfates, or one enzym havirig a specific active site. It has been reported that hurnan placenta1 (Vaccar et al., 1987) and rat (Kawano et al., 1989) arylsulfatase C (ASC-), El-S, and DHEA S-sulfatase activities appear to reside with a single enzyme, wMe three differer enzymes in the-sheep brain have been suggested as responsible for sulfatas activities (Balasubramanian, 1976). Data obtained in primates indicate that i the baboon liver, a single enzyme catalyzes the hydrolysis of the three substrat€ (Ruoff and Daniel, 1991). The monkey brain (Macaca radiota) has been reporte' to possess an arylsufatase with MC- and El-S-sulfafase activities, and a alkylsulfatase possesses only DHEA-S sulfatase activity (Lakshmi anl Balasubramanian, 1981). The sulfatase activity has been described in man peripheral tissues in rat, human, and rhesus monkey (Martel et al., 1994). . . 1.1.3 -c tissues

The adrenal cortex seaetes mainly two kinds of steroi-ds: mineralocorticoids an, precursors of ;ex hormones, which are essential for human life. The adreni steroid precwors indude DHEA-S,DHEA, and ~4-dione(Labrie et al., 1988i Labrie et al., 1989). The adrenal cortex can be histologically divided into thre parts: 1) the outer zone, or zona glomerulosa, which has an 18-hydroxylase bu is defiaent in 17a-hydroxylase, consists of ceils involved in the synthesis a mineralocorticoids that are responsible for the maintenance of sodium balanc and blood volume; 2) the middle area, or zona fasciculata, which is separated b: venous sinuses, is able to secrete glucocorticoids which plays an important rol in the metabolism of lipids and glucose, as well as adrenal sex steroid precursor This is the largest zone and its cells are kranged longitudinally in colurnns; ara 3) the inner part, the zona reticlrlatis, where the ce11 columns become interlace in a network and secretes precutsors of sex steroids.

Recent studies have indicated that the mitabolism of steroids in the adreni cortex is a continuous process from the outer zone to the inner zone and that a three zones are involved in the seaetion of corticosterone or cortisol. Howeve enzymatic formation of aldosterone is limited to the zona glomerulosa, whil enzyniatic metabolism for the formation of cortisol and adrenal sex steroi precursars is Mted to the two imer zones (Chu and Kimura, 1973; Buckle and Ramadiaridran, 1981; Milier, 1988).

It is well documented that the synthesis and secretion of adrenal hormones ar mainly under the control of adrenocorticotropic hormone (ACTH Hypophysectomy leads to profound atrophy of the zona fasciculata aiid zon reticularis of the adrenal cortex (Miller, 1988).

The testis is the main organ responsible for the synthesis and secretion c androgens in man. In adults, this steroidogenic organ consists of two mai: parts: 1) seminiferorrs tubules, which contain spematic elements at al1 stages c maturation, and Sertoli cells, which are attached to one another and form barrier at the periphery of the serniniferous tubes. These cells control the flob of nutrients coming from the interstitial spaces, and are not destined to becom germinal cells. They also secrete numerous proteins, such as the best studiei androgen binding protein (ABP), the seaetion of which is controlled mainly b: FSH and T. Inhibin is also produced by Sertoli cells, which has several comple functions not yet dearly understood. Sertoli cells also aromatize a portion of * to E2. T cornes directly from Leydig cells by diffusion 2) Leydig cells (interstitia cells), which are stimulated by LH and responsible for the synthesis anr secretion of andtogens (Mathes et al., 1983).

Testes sectete T and srnail amounts of E2. Most of the secreted T is dilufed ii the peripheral circulation, however, a small Baction is direct1y transpor ted tc the seminiferous tubules and genital tract. It is now kpown that this local effec is indispensable for the maintenance of spermatogenesis and is also involved ix the activation of sperxnatozoa (Hansson et ai., 1974).

Studies on the testis have revealed that there are two androgen synthetii pathways in this organ: .

1) PREG -PROG - I70H-PR= -~4-dione - T 2) PREG -17OH-PREG -DHEA - dio one -T

The first pathway is more important than the second one in both synthesis anc seaetion of androgens in the testis (Miller, 1988).

Although the steroidogenic organs described above are involved in thc synthesis and secretion of steroids, recent studies have shown that periphera tissues also play an important role in the synthesis and secretion of steroic hormones, especially sex hormones (Labrie, 1991b), and the details will bc discussed in the foilowing section. -

Some early studies have illustrated that in human and other primates, adrenali secrete large amounts of the precursor steroids DHEA and DHEA-S.which arc inactive by themselves, however, both of them can be converted into A4-dionc and then into potent androgens and estrogens by tissue-specific steroidogenic enzymes in peripheral tissues (Adams, 1985; Labrie et al., 1985). In fact, th plasma DHEA-Slevel in adult men is 100-500 tirnes higher-than that of T, thui providing high level of the substrate required for conversion into androgens ir the peripheral.tissues. This high. level of steroids synthesized in the periphera tissues control biological processes in various target tissues.

The formation of active sex steroids from the inactive adrenal precursors DHEA DHEA-S,and/or ~cdioneldy. in the same ceiis whew synthesis tbk plaa withkt king releaseâ in the extrateiiular space, has &en recently described u intracrine activity that corresponds to an economical system which requires minimal amounts of hormone to exert maximal function. in classical endocrine systems, large amounts of hormbnes are needed with only a small fraction used in regulation while the rest is degraded mainly in iiver and kidney. The ovaries and testes are the exclusive sources of androgens and estrogens in lower mammals. The situation is completely different in higher primates, especially the human, where a large proportion of the active sex steroids is synthesized locally hmprecursor steroids seaeted in large amounts by the adrënals, thus, @ring autonomy to the target cells and the possibility of adjusting steroid formation as well as metabolism to local requirement. (Labrie, 1991b). The major importance of steroid biosynthesis by peripheral tissues is clearly indicated by the widespread distribution of expression of key steroidogenic enzymes, namely 3B-HSD, 17P-HSD, 5 a-reductase and /or aromatase, in human peripheral normal and/or cancerous tissues, including the prostate, breast, endometrium, adipose tissue, lung, and liver (Abd-Hajj, 1975; Milewich et al., 1977; Abalain et al., 1988; Lachance et al., 1990; Lacoste et al., 1990; Mate1 et al., 1993; 1994).

As mentioned above, both steroidogenic and peripheral tissues are involved in the synthesis and secretion of steroids, however, the content of steroids in dassical steroidogenesis and target celis is not the same. This leads to the hypothesis that steroid synthesis and seuetion in dassical steroidogenic and peripheral tissues have different regdatory mechanisms.

The regulation of steroidogenesis in dassical steroidogenic tissues appears to involve two major factors: central and peripheral factors. The central factor is mainly under pituitary control. Where the anterior gonadotrophic cells as well as ACïH-seaeting ceiis are implicated in the control of the gonadal and adrenal cortex activities (Yates and Urquhart, 1962; Fortier, 1966; Garren et al., 1971). The peripheral factor involves that the hormones secreted by steroidogenic tissues control the decretion of pituitary hormones by negative feedback. On the other han& except of same central factors, such as PRL and ACïH which appear to be involved in the regulation of the biospthesis of sex hormones in the peripheral tissues (Lorence et al., 1990b; Martel et al., 1990a; Martel et al., 1990b), so far, the data on the regutation of local steroid formation are limited. It is clear that the main effect of central or peripheral factors on steroidogenic .tells is to act on enzymatic systems which play a leading role in steroidogenesis (Hsueh et al. 1984; Erickson et al., 1985; Richards et aï., 1987)-

The skin covers the entire body and accounts for 16% of total body weight Recent investigations have revealed that the skin is niot ody a major site fol androgen action, but ahfor androgen metabolism.

The skin consists of a stratified, cellular epidennis, and an underlying demis O: connective tissues. Below . the dermis is a fatty layer, the panniculus adiposc tissue, usually designated as subcutaneous. In most mammals this is separatec from the rest of the body by a flat sheet of striated muscle, the panniculu: carnosus, but this layer is vestigial in man. There are two main kinds of huma1 skin. C;labrous_skin found on the palms and soies, is grooved on its surface bj continuously alternating ridges and sulci, in individually unique configuration! known as dermatoglyphics. It is characterized by a thick epidennis divided intc several well-marked layers, including a compact stratum corneum, by thc presence of encapsulated sense organs within the dermis, and by a la& of haii foliicles and sebaceous glands. skin, on the other hand, has both haii follides and sebaceous glands but la& encapsulated sense orga&. ïhe majoi two components interested in our studies are sebaceous glands and hair follicles which are presented in the dermi; of the skin (Montagna and Parakkal, 1974).

Sebaceoits glands are present everywhere on the skin except on the palms anc soles. On the skin, the sebaceous glands are found in association with haii structures. They lie in the dermis, and their excretory duck open into the ne&! of hair follides. When several glands are connected with one hair, they lie at th same level. The sebaceous gland is a holocrine gland which forms its seaetior by decomposition of its cells and are replaced by dividing undifferentiated basa ceiis balancing the las. The sebaceous glands in man are aggregates of various shed acini that empty into a duct. Each acinus is attached to a common exaetoq duct composed of cornifying, stratified aquamas epithelium which iz continuous with the wall of the pilary canal and indirectly with the surface oj the epidermis. The wbaceous differentiation is the orderly synthesis segregation, and accumulation of lipid droplets which culminate in enlarged misshapen cells which form sebum. Each sebaceous lobule possesses a peripheral layer of Nboidal, deeply basophilie cek that usually contain no lipid droplets, while the more centrally located cells contain lipid droplets. Numetous studies have established the dependence of sebaceous gland development on androgens. The evidence b that sebum production levels on castrated men are considerably lower than in intact man (Emanuel, 1936; Pochi et al., 1962; Hamilton and Mestler, 1963). Moreover, the administration of T to castrated males (Pochi et al., 1962), children (Rony and Zakon, 1943; Strauss et al., l962), or postmenopausal women in whom sebaœous seaetion k mrmally low (Smith and Bnui~t,1%1; Strauss et al., 1962) results in a significant inaease in sebaceous @and activity. On the other hand, administered systemically, estrogen elicits a reduction in the size and seaetion of sebaceous glands both in men and in women (Jarrett, 1955; Strauss et al., 1962; Peirone and Fenoglio, 1965; Strauss and Pochi, 1968; -bnelli aqd Alessi, 1970).

At birth, the skin is covered in hair of two main varieties. One is short, fine and unpigmented, and is termed veUus. The other is the hair which is long, thick and pigmented. Both types are found in three main groups: non-sexual, ambosexual and male sexual hair. The hair of the scalp, eyebrows, eyelashes and distal extremities is non-sexual and grows continuously after birth. In both sexes the androgen production of early puberty resuits in the growth of terminal hait in the axiilae, lower pubic triangle and distal extremities. Growth of thebeard, upper pubic triangle, and the coarse hair of the trwik and limbs are dependent on adult male levels of circulating androgens. Androgens also paradoxically cause a male pattern bddness in individuals with a genetic disposition. Thus, human h& follicles are responsive to androgens (Takayasu, 1979).

It is well recognized that except the formation of DHT from T of testis origin, the skin possesses ali the enzymes required for transformation of the inactive steroid precursors of adrenal origh, namely, DHEA and DHEA-S. into active androgens and estrogexk (Baillie et al., 1966; Po& and Strauss, 1969; Labrie, 199 lb). The immunohistochemical localization showed that S a-reductase is located in the various layers of the epidermis, in the sebaceous glands, as weU as in the sweat glands (Luu-The et al., 1994); and 3s-HSD is located in the sebaceous @ads (Dumont ct ai., 1992a). 1.1.4.3 Prostate

Prostate is a large gland which surrounds the bladder neck and the first part c the urethra in the midline. It is in association with the seminal vesides, is major secretory contributor to seminal plasma. The prostate is a compat musculoglandular organ and consists of two main parts: the submucosal an( prostatic glands. Cancer of the prostate develops generally in the peripher; zone of the major prostatic ghd(Dyrn, 1977).

The prostate, sllnilar to the &in, is not only a site for androgen action but alsi for androgen metabolism. It possess al1 the enzymes required for activ androgen formation. As described previously type II 5a-reductase and type 17$-H!3D are the major forms detectable in the prostate (Luu-The et al., 1994 Luu-The et al., 1989a; Martel et al., 1992), while human type 1 3P-HSD i expressed mainly in the placenta and peripheral tissues such as prostate anc skin (Rhéaume et al., 1991).

Steroids are transported primarily in the plasma. Although low concentration have been found in erythrocytes, their biological importance rernains unknown Steroids, as water insoluble lipids, for most part, are bound to proteins in orde to be disçolved in plasma (Westphal, 1971). Among several binding proteins albumin, preçent in the highest concentration, which is approximately 1000 tc 10,000 times greater (4Og/L) than all otheis, has several binding sites for almos a11 hormones of low molecular'weight. Most other proteins have lower afftiit: for hormones than albumin while a limited number of plasma proteins, usuaq designated as transport proteins, specifically bind hormones, with a highe: affinity than albumin does (K

Testosterone is bound to three plasma proteins: SBP (K~=0.8nM at 40C and 2 nM at 370C), CBG (0.2 pM), and aibumin (02 mM). As canheen seen, the major portion of T is bounded by SBP in normal men. The free fraction repremts approximately 2% of the total hormone in plasma. Moreover, it has-been demonstrated that the extracellular transport of androgens was alxi performed by an androgen binding protein (ABP), which has high affinity for DHT (Kg = 1.25 nM) and T (Kr, = 0.5 nM) and which is presented in rat testis and epididymis (Ritzén et al., 1971; Hansson et al., 1972; Ritzén et al., 1972; French and Ritzen, 1973). ABP is produced in the testis, seaeted into testicular fluid and carried to the epididymis by the efferent ducts (French and Ritzén, 1973). The concentration of ABP in efferent duct fluid (4-8 X 10-8 M)is sufficient to bind T and/ or DM'in a concentration of 13-26 ng/ml, assuming on a binding site per molecule ABP.

After their diffusion into target cells, the steroids may then bind to their cognate receptors and thus trigger an arrsy of intrinsic nuclear processes to initiate RNA synthesis and exert their biological actions. Although mosf of amino acid sequences of steroidal hormone receptors have been elucidated during the last decade, it took about thirty years to sîructurally identify the protein sequence of the androgen receptor (AR). Androgen receptor belongs to glucocorticoid family induding the glucocorticoid, PROG, aldosterone, and. androgen receptors. The structure is similar to other nuclear receptors, namely, estrogen family containhg estradiol receptor, and non-steroid family contaixkg. the Ts, retinoic acid, and 1,25-DHCC nuclear receptors. The prototypical steroid receptor has three major domains: an amino-terminais of unknown function, a carboxy- terminal whidi is the stemid-binding region, and a central DNA-binding domain. Prostate is the first organ where the presence of AR was localized (Liao anc Fang, 1%9). The suggestion of DHT receptor protein involved &I the actions o. androgens in wide ranges of target tissues and is supported by the finding o. similar proteins in many androgea-sensitive tissues, such as seminal veside! (Stem and Eisenfeld, 1%9; Tveter and Akag, 1%9; Uao et al., lm),hair foliidei (Fazekas and Sandor, 1973), humskin (Mowszowicz et al., 1981), sebaceou! and preputial glands (Bullock and Bardin, 1970; Adachi and Kano, 1972 Eppenberger and Hsia, 1972; Mainwaring and ~an~an,1973), utew (Jungblut e al, 1971), kidney (Gehring et ai., 1971; DUM et al., 1973), brain (Kniewald et al. 1969; Debeljuk et al., 1972; Sar and Stumpf, 1972; Kato and Onouchi, 1973; Sa1 and Stumpf, 1973; Kato et a., 1974), and androgen-dependent mouse mammaq carcinoma (Shionogi tumors) (Yamaguchi et al., 1971; Bruchovsky and Meakin 1973a; Mainwaring and Mangan, 1973; Labrie and Veilleux, 1988; Iabrie et al. 1988b).

1.2.3 aqjj its remtion

After the transportation by the ABP, androgens act in one of three moleculai forms: T, DHT and estradiol (E2).'It is known that the Sa-reductase in th6 peripheral tissues converts T to DHT which is the true active androgen in a1 adult tissues, except muscle (Gehring et al., 1971; Dunn et al., 1973); that is ,T iz usually a prehormone. 'Nonetheless, T does have some direct effects in severa! embryonic structures and in adult musdes. Fhlly, T can be transferred to E2 bj an aromatase in peripheral tissues; this is the form in which T acts on certair parts of the brain (Bolander, 1989).

The normal functions of T and DHT can be summarized into two parts: 1: Differentiation of male external genitalia in fetus. T promotes the developmenl of the male interna1 genitalia such as, epididymides, vasa deferentia and seminal vesicles. While DHT participates in the differentiation of external male structures, namely, penis, scrotum, and prostates. 2) Androgen action after post- puberty. As nientioned above, T induced the male sex drive, performance, muscle mass increase, penis enlargement, vocal cord enlargement, and spermatogenesis. Dm,on the other hand, mediated increased facial and body hair, am, r&lp hair recession, prostate enlargement. Androgens stixnuiate both Linear growth and skeletal rnaturity; that is, although the longitudinal growtl of bones is accelerated, the growth plate shrines and eventually becorne: obliterated when the epiphyses and diaphyses fuse. This fusion marks th( permanent termination of linear growth. Growth is also facilitated by thc anabolic actions of androgens, espedally by increased protein synthesis, which i most rnarked in muscles. T is aromatized to estradiol in central nervous systea nuclei, in Leydig and Seroli ceiis, and adipose tissue to act presumably via thc estrogen receptor to produce effects on sema1 behavior, spermatogenenesis, anc other functions (Mwradian et.al., 1987). Androgens may also have biologica effects as antiestrogens by competing for the estrogen receptor (Zava anc Mffiure, 1978; Casey and Wilson, 1984). In nonprimates, androgens may havc two other important targets: the rnammary gland and the hypothalamus. In th, rnammary gland, DHT destroys al1 or part of the epithelium and thwarts nipplc formation. In the hypothalamus, T, after conversion to E2, changes' the seaetor! pattern of the gonadotropins from the cyclic female pattern to the tonic maic pattern (Bolander, 1989).

As discussed above, androgens corne from adrend and testicular origins. So thc regulation of the androgen formation is under two control systems. Thc biosynthesis of adrenal androgens is controlled by ACTH. The secretion of thi pituitary hormone is regulated by corticotropin releasing factor (CRF). CRF itsel is under dual control: a long lwp feedback control regulated by the level of fra cortisol ,and regulation by the higher brain centers ( a short loop feedback by thc concentration of ACTH has also been proposed). This latter control consists of i 'clock' regulating a circadian rhythm, with highest level of ACTH observec early in the morning , and lowest level during the afternwn. In addition, thir system is involved during periods of stress, causing the stimulation of adrena cottex secretion. ACTH stimulates not only the synthesis of adrenal androgens but also that of cortisol and to a lesser degree, that of aldosterone, although thc adrenai androgens and aldosterone have no effect on the feedback control O ACTH. On the other hand, the secretion of T by interstitial cells in testis i! regulated by luteinizing hormone (LH). The mechanism of action of thi! hormone appears to be the same or very similat to that of ACTH. LH is secretec under the influence of hypothalamic gonadotropic releasing hormone (GnRH) Both LH and GnRH are negatively controlled by T. The seaetion of follidc stimulating hormone (EH)is also regulated by GnRH, but it appears to bt contmlled additionally by inhibin, a peptide produced in seminiferous tubules, The complexity of this tegulation is illustrated by the fact that spermatogenesiz depends on FSH, and also that it is sensitive to T produced locally by Leydig cells under the influence of LH. When steroid levels have been restored, feedback inhibition is re-established (S.mith et al. 1983). -

Arne is an infiammatory disease of the pilosebaceous unit, which mainly affeck the face, chest, back and shoulders-the sites of maximum density of sebaceouc glands. Almost everyone develops some ame during adolescence but the rangc of clinical expression is enormous. It varies from the transient presence. of a fe~ comedos and papillae to a severe disabling and debilitating condition marked b) persistent deep papules, nodules and cysts. These latter patients may also bc affected by crippling hidradenitis suppurative or dissecting folliculitis of thc scalp (Cunliffe, 1989). Up to 30% of teenagers have acne of sufficient severity tc require medical treatment (Cunliffe et al., 1981), but the severity usually declinec in their early menties. Sometirnes acne may continue into the fourth or fiftk decades and it is not particularly rare for patients to present for the first time ir the fifth, sixth or even seventh decades. Except acne vulgaris, there are man] other varieties of acne, some being caused by contact with chemical substances follicular keratinization, bacteria, and inflammation.

There are several factors involved in the development of acne lesions Increased rate of sebum production is believed to be one of the most importani factors. As a group, patients with acne secrete more sebum than normal individuals and the severity of acne is related to the degree of seborrhea. Sebun production is directly dependent on the size and rate of growth of the sebaceouc glands, which is under the control of androgeni hormones. However, there il evidence that acne is not due to increased sebum production alone (Pochi and Straus, 1964). Androgens may be involved in two ways: the first, excessivi lwels of andmgens may Mve the sebaceuus glands. ' In many patients there m borderline elevations in total serum T (Rosenfield, 1986; Troupe et al., 1988) or, more comrnoniy, reduced leveis of SHBG which implies inaeased free T levels Other androgen abnodities such as raised levels of DHEA Gd DHEA-S have been reported. In fact, in man, the increase of DHEA coincides with the preadolescent increase on the contribution of sebum to surface lipids (Raxnasastry et al., 1970). Furthesmore, DHEA or DHEA-S augments sebum seaetion in man (Pochi kdStrauss, 1969; Drucker et al., 1972). The second, thc glands themselves may be particularly sensitive to normal levels of androgem (endsrgan hypermitivity). The theory of end-organ hypersensitivity has been proposed as an explanation for the lack of correlation between androgen levels and acne kverity. The evidence is that the activity of Sa-reductase is increased in the skin of .patients with acne (Sansone and Reisner, 1971).

Hirsutism is defined as the presence of excessive coarse hair 'growth in those areas of the body that are relatively sensitive to androgens and that correspond to male pattern hair growth (Leung and Kiefer, 1989; Baker and McFarland, 1990; Pang and Riddick, 1990). ~heandrogen-dependent areas include the upper lip, chin, cheeks, chest, lower abdomen, and inner aspects of the thighs. By definition. hirsutism develops in only girls and women (Rittmaster et al., 1987; Bailey-Pridham and Sanfilippo, 1989). Hirsutism may- result in social embarrassment, poor self-esteem, and psychological distress. Idiopathic or constitutional hirsutism is the most common cause of hirsutism. The hirsutism can also be cawd by pharmacological agents.

Idiopathic hirsutism is believed to be one of the following'reasons: elevated serum T level, hair foiiicle hypersensitivity to androgens induding increased Sa-reductase activity (Rittmaster et al., 1987; Bailey-Pridham and Sanfilippo, 1989). dekased SHüG leading kicreased concentration of free T. The higher T concentration may be due to: excess sex hormone precursors secreted by the adrenals (Bates, 1981; Chrousos. et al., 1982; New et al., 1983; Ebling and Rook, 1986; ~immondet al., 1986; &bar, 1989). mdrogen-secreting tumors (Cole, 1992), and hyperpmlaetinemia (Schwartz and Flink, 1985; Sanfilippo, 1992). The definition of baldness is that terminal hairs are graduaily replaced by fine unpigmented velius hair. Vittually al1 adult males experience some degree 01 thinning hair with aging. For th& reason, male pattern baldness should not bc considered a pathologie condition. The hair loss usuaily starts with recession 01 the frontal hair he. This can progress to balding over the vertex and finaUy tc complete loss on the top of the scalp, leaving the hair at the sides and occipu! relatively intact. These stages of male pattern alopeaa have been docurnented (Hamilton, 1942; Hamilton, 1951). It is now recognized that women develo~ patt~edbalding in the same manner as men (Venning and Dawber, 1988) and in a more diffuse vertical style (Ludwig, 1977). In fact, 13% of premenopausal and 37% of postmenopausal women show signs of male-pattern hair los5 (Venning and Dawber, 1988). Baldness in womert is a relatively common encountered cornplaint. Hair loss in premenopausal women cm be divided intc two different types. The most common type is generally not attributed tc increased androgen production and is known as "diffuse alopeua" withoul fronto-parietal recession (Ludwig, 1977; Venning and ~awbir,1988; Valavara, 1990). The second type, on the other hand, is aswciated with inaeased androgen production and is so caued "patterned androgenic alopecia". This type 01 baldness usually CO-existswith other virilizing features, especially hirsutism, acne and seborrhea (Adachi et al., 1970; Futterweit et al., 1988).

It is known that androgens play a role in this process, but the role is poorly defined. This phenornenon was first demonstrated in studies of eunuchs. U being castrated before puberty, these men had no male pattern hair loss. When given androgew Late in life they become bald rapidly (Hamilton, 1942; Hamilton, 1951). Moreover, men with Sa-reductase deficiency do not deveiop male-pattern baldness. Although, serum T levels are similar in balding and nonbalding men, but hair follicle Sa-reductase activity is higher in the scalp oi balding men (Bingham and Shaw, 1973; Schweikert and Wilson, 1974). Enlargement of the prostate to the extent that it produces obstruction of the urethra is an almost universal finding in aging men (Moore, 1943; Moore, 1944; Ubelhor, 1962; Walsh, 1979). During human life the prostate undergoes three cycies of growth. The fht is a slow inaease in size More the puberta1 period. With the inaease in anàrogen production during puberty the prostate attains a weight of about 20 g. The third cycle occurs in the fifth to the seventh decades, when the prostate triples in size and weight (Swyer, 1944). By age 70180 years, most men (>BO% ) have palpable or histologie evidence of BPH (Grayhack et al., 1975). About 9%of them progressing to clinically detectable enlargement of the gland, and the dinical syndrome of bladder outlet obstruction will develop in approximately half of these men (Isaacs, 1990). Because of refinements in prostatic surgery, BPH is not a leading cause of death, but it remains a major cause of morbidity in the elderly men (Grayhack et al., 1975). There is a rernarkable species specificity in that prostatic hyperplasia occurs only in man and dog. This relation implies that some critic difference in porstate physiology between these and other species must be crucial to its development.

DHT pla ys a central role in male phenot ypic differentiation (Siiteri and Wilson, 1974) and in the postnatal development of the human prostate (Peter et al., 1977). Serum DHT Ievels are high in elderly men (Horton et al., 1975). DHT contents are increased in the periurethral region of BPH (Siiteri and Wilson, 1974; Gelier et al., 1976),.especially in the nudear fraction of BPH stroma, which is greater than in prostatic tissue from young men (Meikle et al., 1980; Krieger et al., 1983). In studies of canine experiments for potential ability to induce BPH (Moore et al., 1979), DHï and 3a-di01 are potent growth factors of prostate. DHT accumulation is observed in the pmstaie by the treatment of both 5a-reductase inhibitors and antiandrogens. Furthermore, in vivo (Morimoto et al., 1980) and in vitro (IPsacs et al., 1983) stuclies of androgen kinetics suggested that the accumulation of DHT in prostatic tissue is a causative factor in the development of BPH and results from a shift in the overall balance of androgen metabolism, which favors an hcrease in the net fornation of DHT'. In dogs, the only major anima mode1 for BPH (Berg, 1958a; Berg, 1958b), DHT levels are significantly increased 'in BPH tissue, although T levels are similar in normal and hyperplastic prostate tissue (Gioyna et al., 1970; Uoyd et al., 1975; Meikle et al., 1979; Meikle et al., 1981; Ewing et al., 1983; Carpenter et al., 1990). The Sa- reductase enzyme activity is ais0 inaeased in BPH tissue by comparing with nord dog prostate. If a dog is castnted and replaced with the administration of DHT, BPH can be induced (Walsh and Wilson, 1976; Moore et al., 1979, Horton, 1984) .

Prostate cancer is the most Erequent cancer in men and is the second mosi important cause of male cancer death in the United States (Boring, 1995). In fact, in 1995, it is predicted that 44,000 men will die from this disease, while 240,000 new cases of prostate cancer will be discovered in the United States alone (Boring, 1995). The lifetime adds of developing prostate cancer is 1 in 11 men. Moreover, as the popdation ages, it is expected that the impact of prostate cancer will increase further (Chisholm, 1981).

Among al1 hormone-sensitive cancers, prostate cancer is recognized as being the most sensitive. nie best-knawn characteristic of prostate cancer is its marked sensitivity to androgen deprivation. In fact, using surgical or medical castration, as originaiiy suggested by Huggins and Hodges in early 1940's (Huggins and Hodges, 1941), which oniy blocks testicular androgens, a temporary respome is observed in 60-8036 of patients (Veterans Administration Cooperative Urological Research Group, 1967). Such a high rate of positive respowes is already quite impressive rince, testicular androgens account for only 60% of total androgens in 65-year-old men, thus leaving approximately 40% of androgens free to continue to stimulate prostate cancer (Labrie et al., 1985; Labrie, 1991b; Labrie, 1991a). In other words, with prostate cancer, one must am at the best achievable response and not simply be satisfied with any type of positive response which is relatively easy to achieve but is also rapidly followed by the relapse of the cancer. In many experimental data and recent dinical trails, it has been observed that combination therapy exerts a maximal uihibitory effect on prostate cancer growth (Labrie et al., 1994). It has ben desaibed in previous sections that skin disorders, such as acne hirsutism, and male pattern baldness are dependent on an excess of androgen! or increased sensitivity to androgens (e.g., increased Sa-reductase activity in tht skin). Therefore, a logical treatment of the above diseases could be achieved bj either influence the most potent androgen DHT formation with Sa-reductast inhibitors, or block the peripheral androgen effect through competitivt inhibition of the formation of the hormone-receptor complex with tht antiandrogens, as shown in Fig. 2.

Oral administration of systemically active antiandrogens such as flutamidt (Cusan et al., 1990), spironolactone (Matias and Orentreich, 1983; Goodfellow ei al., 1984; Marchetti and Labrie, 1988), cyproterone acetate (Burton et al., 1973, Winkler and Schaefer, 1973), and 17a-methyl-B-nortestosterone (Zarate et al., 1966) could conceivably be an effective means of treatment, but is not the besi choice in practice, because of the possibility that these drugs might reduce libidc and impair spermatogenesis in. men and feminize male fetuses in pregnani women. Thus it would be of great therapeutic value to develop drugs that havi an antiandrogenic effect without causing the above mentioned systemic sidc effects. The search of such compounds is still a fascinating area of research.

Prostate cancer is the most frequent cancer in men. In more than 50% ol patients, this cancer is discovered only at an advanced stage with bone metastases, and the only efficient treatment available is endocrine therapy (Huggins and Hodges, 1941; Lab.de et al., 1985).

After the original report in 1941 about the hormone sensitivity of prostatg cancer (Huggins d Hodges, l~i),m~pprerrion of testidar andmgens har been the cornerstone of treatmnt of advanced prostate cancer, usually achieved Figure 2. Biologicai pathways of udtbgen action and the sites of inhibition of androgen actions by Sa-redudase inhibitors and antiandrogens.

TESTE ADRENAL

l RNA SYNTHESIS PERIPHEFIAL TARGET TISSUE by surgical castration or treatment with high doses of estrogens (Veteran! Administration Cooperative ~rblo~icalResearch Group, 1967; Mettlin et. al. 1982). It should be pointed out that, despite being regarded as standard first-lim hormonal treatment for metastatic prostate cancer, orchiectomy .s not beer proved to prolong patient survival (Paulson et al., 1984). ~owker,castration i! well recognized as improving quality of life in a high proportion of patients Aithough surgical castration presents serious psychological problems for maq patients, high doses of estrogens, on the other hand, have seriour cardiovascular complications which can lead to death in 15% of patients during the first year of treatment (Glashan and Robinson, 1981). in 1977, a study ha! discovered that chronic administration of LHRH agonist to intact male ratr caused a blockade of T seaetion by the testes, followed by atrophy of the ventra prostate and seminal vesicles (Auchair et al., 1977). It has been then achievec the first medical castration using a LHRH agonist in a man with prostate cancei in 1980 (Labrie et al., 1980). LHRH agonists act at the level of the anterioi pituitary gland by blocking the secretion of bioactive LEZ, thus depsiving th testes of their physiological stimulus and causing an arrest of androgen secretior and spermatogenesis (Labrie et al., 1980; St-Arnaud et al., 2986). Serum testidai androgens can thus be reduced easily to castration levels during ckvonic treatment of men with LHRH agonists. It is also important to recognize that, although LHRH agonists offer a more acceptable method of castration that is fie€ of the important side effects of estrogens (Glashan et Robinson, 1981). We cannot expect to improve the prognosis of prostate cancer beyond the results achieved with orchiectomy, since the effects of LHRH agonists are limited similarly to blockade of testicular androgens. Moreover, as LHRH agonists induce a temporary rise in T seaetion during the first 5-12 days of treatment, with the risk of tumor exacerbation or flare-up of the disease during this period, they should never be administered alone without the protection oi simultaneous treatment with a pure antiandrogens (Labrie et al., 1986a; Labrie el al., 1986b). Although castration @y orchiectomy or LHRH) causes a 95% reduction in serum T concentration, a much smaller effect is seen on the only meaningful parameter of androgenic action in the prostatic tissue: the intraprostatic concentration of DHT. In fact, after elhination of testiculai androgens by medical or sur@ castration, the intraprostatic conkration of DHT temains at about 40% of that measured in intact men. As another measurement of the importance of adrenal andrdgens in adult men, the serum levels of the main metabolites of androgens Sa-androstane-3a,17&diol (3a-diol), ADT, and their glucurckidated derivatives are reduced by only 50-70% (Moghissi et al., 1984; Bélanger et al., 1986) following castration, thus reflecting the high level of adrenal precursor steroids converted into DHT in castrated men. Contrary to the previous erroneous belief that the testes are responsible for 95% of total androgen production in men, it is now well demonstrated that the prostatic tissue efficientIy transforms the inactive steroid precursors (Labrie, 1991b). Since the adrenal gIands are responsible for close to half the total androgens in adult men, as a iogical consequence, combination therapy with a pure antiandrogen given in association with medical or surgical castration should be logically used to block simdtaneously the androgens of both testidtr and adrenal origins. The complete blockage of androgen action is the combination of castration, a pure antiandrogen, and a Sa-reductase inhibitor.

Sa-reductase selective1y catalyzes the irreversible conversion of 4-ene-3- oxosteroids into the corresponding Sa-3-oxosteroids in the presence of its coenzyme NADPH, especially in the transformation of the major adult male circulating androgen T into the more potent androgenic metabolite DHT in the extranuclear compartment of target ceïls. Since elevated DHT concentration in target cells plays an important role in etiology of certain diseases, inhibition of androgen action by Sa-reductase inhibitors is a logical site. According to the chernical structural characteristics, the inhibitors of Sa-reductase are divided to two major categories: non-steroidal and steroidal inhibitors.

Zinc sulfate, the oniy inorganic sait that has ken reported among al1 inhibitors of Sa-reductase and androgen receptor, has been first demonstrated existing the inhibitory effect on Sa-ductase in human skin homogenate (Stamatiadis et al., 1987). The inhibitory activity of zinc sulfate together with azelaic aad on Sa- reductase was studied in human skin by using an in vitro assay with [1,23~]-1 as substrate (Stamatiadis et al., 1988). An additive effect of the two inhibitors wa! observeci. Vitamui 86, the precursor of pyridoxal 5'-phosphate (PU,a coenzymr of phosphorylase), potentiated the inhibitory effect of zinc, but not of azelaic acid, suggesting that two different mechanisms are involved. When the thret substances were added together at very low concentrations which has beer shown to be ineffective alone, 90% inhibition of Sa-reductase activity wa! obtained.

Azelaic acid (l,7-heptanedicarboxylic acid), a Cg straight chai.fatty acid, has beer reported having inhibitory effect on acne &lgaris after topical applicatior (Nassaro-Pono et al., 1983). One hundred patients with acne vulgaris treated foi 3-9 months by twice daily application of the cream had significant improvemeni in every case. The mechanism might be explained by later in oilro study of thii organic acid on inhibition of human skin 5a-reductase (Stamatiadis et al., 1988) The inhibition was detectable at concentrations as low as 0.2 rnk and about 70% of inhibition was detected when the concentration was at 0.5 mM and, 85% at 1.C mM, and complete inhibitory effect was observed at the concentration of 3.C mM.

From a fermentation broth of a strain of Strepfornyces sp., two phenazim derivatives (designated as WÇ9659A and B ) were purified by solvent extraction foliowed by silica gel chromatography (Nakayama et al., 1989b). The IC50 values of WS9659A and B for partially purified rat prostate Sa-reductase were 5.0 X 1û-7 and 1.0 X 10-5 M, respectively. Radiolabelled receptor binding assay of rai prostate for androgen receptor suggested that W59659A had no affinity for this receptor (1% > 3 X 1~ M). Furthermore, WS9659A was tested subcutaneously in immature castrated rats, and showed the inhibitory effect on the growth ol the ventral prostates induced by T propionate (Nakayama et al., 1989a).

Riboflavin (vitamin Bz), another conunon natural -phenazine product, was isolated from fermentation of a strain of yeast, and the ICso value of its inhibitory activity for rat prostate Sa-reductase was 5.0 X 10-6 M. Among th4 riboflavin analogous tested, lumiflavine and proflavin hemisulfate were mon active, with IC50 values of 7.0 X 10-7 and 9.3 X 10-7 M, respectively (Nakayama e al., 1990).

Among al1 the compounds in this subcategory, ONO-3805 has been reportec with an ICm value of 1.2 x 10-7 M for the inhibition of rat prostatic Sa-reductasi (Nakai et ai., 1988).

Benzoquinolinones, the newest type of non-steroidal inhibitors, weri synthesized starting from Ztetralones (Jones et al., 1993). Among of them LY191704, the tram-isomer, is an extremely potent inhibitor of human type ~ steroid Sa-reductase with an ICSOof 8.0 x 10-9 M, while its cis-isorner mea 5P, relationship in 4-azasteroids) hold an IC5o of 4.1 x 10-8 M, is about 5-times les! active than LY191704. In contrast, the benzoquinolinones is a class of weal inhibitors of the hwnan type II enzyme. Even through, they represented a nove opportunity to intervene clinically in conditions due to local overproduction oi DHT' via the type 1Sa-redcutase (Jones et al., 1993).

Steroidal inhibitors of Sa-reductase could further be divided, according to th6 structural characteristics of the A-ring, into four subcategories: riaturd steroidal derivatives, 3-carboxylic acid and SphosphUnic/3-phosphinic acid derivatives, 4-nitrogen or sulfur containing substituted derivatives, and 4-azasteroidal derivatives. In 1970, 23 steroidd hormones were first assayed for inhibiting the conversior of [4-14C]~to Dmby human neonatal foreskin miaosomal preparations, th enzyme had a Km of 1.1 x 106 M for T (Voigt et al., 1970). The most poteni inhibitor was the one of the natural substrate of Sa-reductase enzyme, PROG, which restrained the transformation of T into DHT up to 93.3%. The PROC itself was converted into Sa-pregnane-3,2(Fdione, and competed effectively witk T for the active site. The Ki for PROG was .approximately 0.7 x 10-ï M. Othe] potent inhibitors tested were deoxycorticosterone acetate, deqxycorticosteron~ uid Ad-dione with the inhibitory percentages of 65.8%. 84.7% and 76.4%. respectively; while PREG, DHT, ADT, DHEA, and E2 had no significani inhibitory effects (lesr than 5% of inhibition) (Voigt et al., 1970).

The first potent inhibitor of steroid Sa-reductase was 4-androsten-3-one-17p. carboxyiic acid (Voigt and Hsia, 1973a), which was identified as a metaboiite ot deoxycorticosterone in perfusates of rat iiver (Levy and Maloney, 1961) and bovine adrenal (Picha et al., 1952). The compound was reported as a cornpetitive inhibitor and showed a 87.7% of inhibition for the miaosomal preparation 01 human skin, and a Ki of 1.1 x 104 M which is similar to the Km fomd for T.

A series of lob-alkyl (methyl, ethyl, propyl, cyclohexyl and phenyl) derivatives of androstane, El and EL originally prepared as antiandrogens, were tested for their inhibitory effects on the rat prostate Sa-reductase (Sudo et al., 1981). Among them, TSAA-291 (16~thyl-17~hydroxy-4-estren-3-one),was the firsi compound which had dual action on competitive inhibition of Sa-redcutase activity and andmgen receptor mplex formation, and showed a Ki of 1.4 x 106 M on Sa-reductase of purified rat prostatic nudei fraction (Sudo et al., 1981). The molecules fïrst found to be noncornpetitive and possibly irreversibl inhibitors of epididymal Sa-reductase (Robaire et al., 1977) were 4R-5,1( secoestra-4,5-diene-3,10,17-trione and 4R-5,10aeco-19-norpregna4,5-di~ 3,10,20-trione, which were derived form the mechansim-based irreversibl inhibitors of 5sne-3-ketosteroid isomerase (Covey idRobinson, 1976).

The primary evidence of a dramatic inaease in the affinity of Sa-reductase an an inhibitor with Sa-juncture of A-/B-ring and sp2 hybridization at C-3 and C- position was obtained hmthe inhibition of rat ventral prostatic Sa-reductas w i th (5a,20R)-4-diazo-21-hydroxy-20-methylpren-ne (RMI 18,341) (Metcà et al., 1980). Then, diazoketone has been reported to be a potent tirne-depender inhibitor of Sa-reductase with a Ki of 3.5 x 10-8 M (Blohm et al., 1980).

Another potent time-dependent inhibitor of Sa-reductase was 2',3'a tetrahydrofuran-2'~pù0-17-(6-meChylene-4-androsten-3-one)(L 612,710), whicl was isolated fkom rat microsorne, and caused the highest percentage (81%) a inhibition among ten selected steroidal derivatives (Dupuy et al., 1978).

Recently, the modifications of the A-ring and together with the substitution o 1Ffunctionality of steroid nudei have becorne fashiow to develop more poten mechanism-based inhibitois of Sa-reductase. Among of rnany, the 3-carboxyli~ (Metcalf et al., 1989) and 3-phasphanic acids (Lavy et al., 1991) were prepared am were examined as cornpetitive inhibitors of the enzyme. From the beghhg of the last decade, hundreds of 17B-substituted-4-azasteroid have been reported as inhibitors of Sa-reductase. Among them 2',3'a tetrahydrofuran-2'-rpiro-17(4-a~a-5a-androstan-3ne)s(Liang et al., 1984), 4 MAPC (sodium 4-methyl4-aza-k~x~Sa-pregnane-2O(S)-carb0~y1ate(Kadoham et al., 1984), L-651,580(methyl 3-0x04-methyl-4-aza-Sa-androst-l-ene-17b carboxylate) (Brooks et al., 1991b), AMPD (4-aza-4-methyl-Sa-pregnane-3,2( &one) (Bertics et ai., 1984), and L445,066 [(5a)-23-rnethyl-4-â2a-21-norchol-l~nt 3,îOdione] (Depre et al., 1992) were well stuciied. However, &MA and MK-90 (finasteride) are the most extensively investigated of numerous 17B-substitutel 4-azasteroids, and will be discussed in more detail in the following sections.

itv- of 4-M4.At the beginning of the 1980s, both in vitro and i vivo inhibitions of Sa-reductase, receptor binding, and nuclear uptake c androgens in the male rat prostate by 4-MA were first reported independently b two research groups (Brooks et al., 1981; Liang and Hesis, 1981). The inhibitio. by 4-MA was cornpetitive with T, but noncornpetitive with NADPH (Liang an1 Hesis, 1981; Liang et al., 1984). 4-MA was particularly effective at the optimal pl of the enzyme (pH 5.5) and had.no effect in the alkaline range (pH 9.0) (Leshi and Wilson, 1982). 4-MA,rneantime, was demonstrated to have no significan androgenic, estrogenic, progestational, antiprogestational or antigonadotrophi activity (Liang et al., 1983). Except the inhibitory activity on Sa-reductase, 4-MA was also reported to be a potent inhibitor of 3P-HSD activity with a Ki value a 5.6 x 10-8 M (Takahashi et al., 1990). While the enzyme activity of 17p-HSD wa not inhibited by 4-MA (Takahashi et al., 1990).

d m-9%.4-MA was replaced in development by WOt h4K-906 was reported as having an IC5o value of 1.0 x 10-8M with humai prostatic Sa-reductase and 6.8 x 109 M with the rat prostatic enzyme, but it ha( no affinity (1- > 1.0 x 1C@ M) with the rat androgen receptor in vitro (Liang e al., 1985; Rasmusson et al., 19%). h nad male volunteers, MK--906 suppressed 50% of plasma DHT levels at 24 hours after a single administtation of dose as low as 0.5 mg (approximately 0.0: mg/kg), and DHT levels returned to normal within 5-7 days. Fulïy dinica effects of MK-906 with normal male volunteers in a blinded placelmcontrolled at daily oral doses of 25,50 and 100 mg @art 1) and 0.04,0.12,0.2 and 1.0 mg (par1 2) were investigated. Results from part 1showed a significant reduction in DHI and ~4-dioneat the three doses (Gormley et al., 1990; De khdpper et al., 1991) while LH, FSH, cortisol, E2 levek remained unchanged. Serum lipids, induding total cholesterol. low density lipoprotein, high density lipoprotein, and triglycerides were also not affected by the treatment. Results from part 2 again exhibited significant reduction in DHï at al1 doses: DHT levels returned ta pretreatment'values within 14 days after the last dosage. Significant inaeases ol T concentrations were observed only in the 1.0 mg group and only during the first 8 days of treahent. The T/DHT ratio inaeased with all doses and returned to baseline when drug was discontinued (Gormley et al., 1990).

A placebo-controlled study in patients awaiting definitive surgery with clinically significant BHF was undertaken to determine the ability of MK-906 to suppress intraprostatic DHT levels (Geller and Franson, 1989). After 7 days of treatmenl with 50 mg of MK-906, once daily, prostatic DHT decreased by 92% and intraprostatic T levels increased from 0.26 f 0.18 ng/g tissue to 1.91 ng/g tissue. This study has first demonskated that MK-906 caused significant inhibition of Sa-activity in the hyperplastic hurnan prostate, with a moderate, redprocal increase in intraprostatic T (Geller and Franson, 1989). Moreover, for patients with BPH, MK-906 was found to decrease prostate volume by a mean of 28% over a penod of 6 rnonths without causing' dinically significant adverse effects (Stoner, 1990). The selective inhibition of Sa-reductase by MK-906 provided a novel approach to BPH therapy by reducing prostate sue without affecting T- dependent processes such as fertility, muscle strength, and libido. Although DHï seerns to be the main androgen stimulating and maintainhg the growth of BPH, it is unclear whether the androgen dependence of prostate cancer is also related mainly to DHï because the application of MK-906 to the Copenhagen rat Dunning prostatic adenocarcinoma mode1 R-3327 had no effect on the growth of the Du~ing-Htumar; however, castration and diethylstilbestrol in the control group did sigdcantfy suppress the growth of Dunning tumors (books et al, 1991a). On the ather hand, clinical experience applying Sa-seductase inhibitors to prostate cancer patients is very LLmited (Gormley, 1991). The first study reporteci was the 12-week application of MK-906 (10 rng/daily) to 28 untreated patients with a symptomatic stage D prostate cancer. Among prostatic volume, PSA, and prostatic acid phosphatase (PAP)investigated, oniy the significan decrease in PSA was observed, plus a 77 to 79% reduction in plasma DHï however, the effect does not seem as pronounced as usually seen with castratioi or total mdrogen blockade in this group of patients (Presti et al., 1992). In th second clinical study, 150 men experienced a raised in PSA level after radia prostatectomy were treated with MK-906. After MK-906 treatment, a significan decrease in PSA was measured, and more pronouncedly in patients whc initiaily had lowatage tumors (Andriole et al., 1994).

In a recent dinical report (Diani et al., 1992), MK-906 was administered ordy a 0.5 mg daily, alone or in combination with topical2% minoxidil for 20 weeks tc determine the effects on scalp hair growth in balding adult male stumptai macaque monkeys. The combination of MK-906 and minoxidil generatec significant augmentation of hair weight (additive effect) compared to eithe drug alone, while MK-906 and minoxidil elicited a significant .devation in hai weight compared to vehide alone. Serum T was unchanged, whereas serun DHT was significantly depressed in monkeys that received either MK-906 or thi combination of MK-906 and minoxidil (Diani et al., 1992).

The term antiandrogen is "prevent androgens from expressing theu activities a target sites" by either impeding the binding of androgens to their receptor or b] decreasing the number of available binding sites. Thus, the antiandrogen usec should be a compund having specificity and affinity for the androgen recepto~ while not possessing any androgenic, estrogenic, progestational, ghcocorticoid or any other hormonal and antihormonal activities.

The study of antiandrogens was started in earlier 1960's, since then, hundreds O, natural or synthesized compodds of various classes, of which the majority i: steroidal derivatives, were reported as androgen antagonists (Neumann et al. 1967; Neumann et ai., 1%8). The biologic& activities, pharmacologiul uses, and clinical applications of antiandrogens were well reviewed (TindaU et al., 1984 Neumann and ToFf1986; Moguilewsky and Bouton, 1988; Sciarra et al., 1990) Antiandrogens can be divided into two catalogs according to their biologica effects: topical antiandrogens and systeniic antiandrogens. The topica antiandrogens must be active topically and act thorough cutaneous androge~ receptors and has to be devoid of systemic side effects. In contrast, systemi antiandrogens inhibit androgen action in ali target tissues and not only at th desired target site. According b the chernical structural characteristics antiandrogens can be catalogued into two main classes: non-steroidal inhibitors such as flutamide, and sieroidal inhibtors, such as cyproterone acetate (CPA).

Non-steoridal inhibitors of the AR can be divided into four sub-categories: N 3,4,(5)-substituted phenyl isobutyramide derivatives, des-A-steroida derivatives, dodecahydrophenanthrene analogous, and miscellaneous chemica structures.

ide nenvativa. . .

Flu- The best known non-steroidal antiandrogen is N-(4'-nitro-3' triflurorrnethylpheny1)-2-methylpropionamide (Sch 13521, Flutarnide). It wa: originally synthesized to test on Staphylococcus aureus for its anti-bacteriostatil activity in vitro (Baker et al., 1967). The antiandrogenic activity of flutamidc was discovered later with antagonized effects of Testo, TP, ~4-dione,and DZF! on the seminal vesicles and ventral prostate growth. The weights of semina verides and *ristate gland were reduced after the treatment of flutamide witl the doses of 1 to 50 mg, once daiiy for 1to 3 weeks in intact male rats (Neri et al, 1972). The similar results were' observed in the flutamide treatec hypophysectomized rats supplemented with hCG (Neri and Peets, 1975). Nc androgenic, estrogenic, anti-estrogenic, corticoid, progestational, anti progestational, or antigonadotmphic activities were observed. The reduction O the prostate size of aged dogs with BPH wâs also seen at 6 weeks, and thc prostate remained reduced at 3,6, and 12 months after initiation of drug therapj with flutamide (Neri and Peets, 1972; Neri and Peets, 1975). Flutamide wai demonstrated to inhibit the formation of the nudear 3s pmtein-androger cornplex. The treatment .with Autamide induced a direct inhibition of th genital tract. On the hypothalamo-pituitary axis, flutamide acted as a puri antim&ogen,'which was able to inhibit the negative feedback of T, produced ai inaease in LH synthesis as welî as in the secretion of FSH and PRL (Viguier Mattinez et al., 1983; Sciarra et al., 1990).

of F-. The metabolism of flutamide has been studied ii man after both oral and topicai administration (Katchen et al., 1975; Katchen e al., 1976). After a single oral dose of radiolabeied flutamide, the compound wa rapidly ibsorbed. the plasma radioa&vity concentration' reached maximua levelç two hour later. Of the extractable radieactivity in plasma, flutamide wa, accounted only 3.7%, and more. than 10 other metaboites were made up the tes portion. The hydroxylated derivative (OH-flutarnide, Sch-16423) is the majo plasma metabolite (Neri et al., 1979). After S days, the total radioactivity Wai measured 45.5% in urine and 4.7% in feces (Neri et al., 1979).

When topical administration of flutamide in. a solvent of ethanol/propyleni glycol (1:l) is used, 16% of the applied drug was absorbed whüe 56% wai recovered from the application site after 5 hours (Katchen et al., 1975). Plasmi fiutamide level peaked within 4-6 hours and declined to 6% of peak level at 2r hours after administration. At least 10 plasma metabolites have been found. Th same metabolites were seen foilowing oral and topical doses, but the rates of thi metabolites were different. One hour after the oral administration, flutamidr was almost completely converted to other compounds. The concentration o OH-flutamide was about 10 times higher than that of flutamide (Katchen et al. 1975). On the other hand, flutamide and OH-flutamide had approximately equa plasma concentrations 4-8 hours after a topical administration.

Sch 16423, OH-flutarnide showed about 1.5 times mord potent inhibitory activity than flutamide in vivo (Katchen et al., 1975; Neri e

RU 38882 was about 25 times les potent than CPA when given by subcutaneou! or ord administrationsf but it was about 100 times more potent than CPA aftei topical application in the intact rat sicin. It decreased the density of the srnad endoplasmic reticulum vesicles of sebaceous glands in a dose dependen manner, and unlike CPA, did not moMy the prostate weight following topicz application (Bouton et al., 1986).

Compounds in this subcategory were deveIoped in the early studies c antiandmgens. The first compound investigated was RO 2-7239 ,which showel antiandrogenic and antimyotrophic activities, inhibited the hypertrophy of th seminal vesicles, ventral prostate and levator ami induced by TP in th immature castrated rats, at the doses that equal to TF (Randall and Selitto, 1958: hother compound was RO 5-2537 (Boris, 1965). At the dose of 40 times of T, i &O possessed progestationai and very weak antiestrogenic properties (Boris Stevenson, 1966).

According to the structural natures of steroidal skeletons, steroidal inhibitors a AR could be catalogued as three sub-categories: progesterone derivative1 testosterone derivatives, and A- and D-ring modified steroids.

mne acwThe most famous anti-androgen, an analogoue a progesterone is cyproterone acetate (CPA, 17a-a cetox y-1 a-2a-meth ylene-6 chloro-pregna-4,6-diene-3,20-dione),which was the fust compound which ha been demonstrated to has a dose-dependent inhibitory effect on th1 masculinized female rat fetuses caused by the administration of TP (Neumani and Kramer. 1964), and thus was originally patented as a progestin (Weicheri 1966) . Late studies have demonstrated that CPA reduced the gross prostatic sizi and prostatic epithelial height in more than half of the men who received CPl orally at the dosage of 50 mg, once daily (Scott and Wade. 1969). Moreover, thi oral administration of 100 mg of CPA has been applied to over than 600 patienu during the day 5 to 14 and ethyl estradiol dwing the day 5 to 25 of the femali cycle, thewith rymptom of acne and seborrhea showed the first and bes responriveness with an. aîmost 100% success rate, while the hirsutism wa, improved in 60% to 80% of the subjects (Harnmerstein et al., 1975). CPI decreased sex steroid synthesis in the testes without suppression of LH seaetioi and appeared to block steroid-receptor binding in the prostate (Geller et al., 1975: But this compound did not have effect on the hamster sebaceous glands afte topical application ( Bouton et al., 1986), and clinically ineffective when appliec to human skia

S.. . As newly developed CPA analogous, TZF 4238 and CMA inhibited effectively the growth of androgen-induced canin1 prostate. and showed at least five times more potent than CPA (Takezawa et al 1992).

The best studied in this group was SKF 7690 (17 a-methyl-B-nortestosterone (Josko et al., 1960), which had no hormonal activity and inhibited the formatioi of DHT-receptot complex in the cytosol of rat ventral prostates (Sudo et al- 1979). . Its affinity for AR was about 11%of DHï (ïindall et al., 1984). Anothe receptor antagonist is R2956 (Azadian-Boulanger and Bonne, 1974), whid significantly antagonized the TP-binding with -an 8:l dose ratio and showet virtually no androgenic activity (Raynaud et al., 1977). 6 Dehydroretrotestosterone has been reportèd to be also an antiandrogei (Kassenaar et al., 1960).

Topical application of the 17a-propyltestosterone (Win 17665) caused a dose related tegression of the hamster flank organ and guinea pig supracaudal glanc in mature male animals (Ferreri et al., 1978; Brooks et al., 1991b). Histologica examination confirmed that 17a-propyltestosterone decreased the size of thc hamster sebaceous glands and was reversible on cessation of treatment Another interesthg example of this family is 17a-propylmesterolone (SH 434) which significiently reduced bot. sebaceous gland size and sebogenesis of thc Syrian hamster in a dose dependent mannet, and there was no influence or plasma T concentration or testicular weight after topical administratior (Luderschmidt et al., 1984.). Furthermore, neither l'la-propyltestosterone no] 17a-propyl-mesteralone inhibited the prostate weight, the contralateral flad organs and the sebaceous glands following topical application on the one side c hamster flank organ which indicates that both compounds possess bttle or n systemic activity (Brooks et al., 1991). However, Seminal veside weight wa reduced after the repeated subcutaneous administratio~of 100 mg/kg of 17a propyltestosterone (Ferreri et al., 1978). -

A number of A-ring modified androge antagonists has been reported. The topical antiandrogenic activity of L-651,58 was as potent as that of Win 17665 and more than that of SH-434. The topici administration of effective doses of L-651580 caused few, if any, systemic effeci (Brooks et al., 1991b). Other A-nor androgen antagonists include A nortestosterone (Weisenborn and Applegate, 1959), A-norprogenterone (Lem€ et al., 1960), and 3-hydroxy-A-nortestosterone(Levin, 1965; Levin and Diass 1965)...... *. *. Spironolactone (SC 9420) (Cella and Tweit, 195! 1961), the well known aldosterone antagonist, has been reported to be a androgen antagonist in rats -and men (Covol et al., 1975; Pita et al., 1975).

Androgens stimulate the growth of sebaceous glands, hairs and increase th vascularity of the skin in men (Edwards et al., 1941; Hamilton and ~ontagm 1950). Abnormal function of androgens could cause androgen-dependent ski disorders, namely acne, hirsutism, and male pattern baldness. In order t investigate the mechanisms involved and the control of these problems, ther is a need for a practically comparable animal mode1 for study of activity of th sebaceous gland and hair foilicle. The most Çommon models used are th costovertebtal sebaceous organs of hamsters (Takayasu and Adachi, 1970; Frost

8 al., 1973; Gomez and Frost, 1975; Pochi, 1975), gerbils (Q~Mand Gray, 1%5), and the sebaceous glands deep inside the ventral surface of earlobes in rabbits (Mi.k and Kîigman, 1975). The preputiai glands in rats (Ebhg, 1963; Ebling, 1974) hm been used for planimetric studies of normal sebaceous glands (Ebling, 1963), and the 6ebaceous glands of larger animals such as calves (Smith and Jenkinson 1975) or sheep (Downing et al., 1975) have also been used. A suitable sebaceoui gland model rhould fulfill several requirements: the gland should bt morphologically similar to those of man with a sebaceous follicle witl infundivulum, lobdes, and a piIary unit; they shodd be androgen sensitive; il should be economic to use. The Syrian hamster Wi many of these goals. Ir this animal, study can be made not only of the skin, in general, but also oi special cutaneous areas. That the easy accessibility, relatively inexpensive tc purchase and care for, and easy to work with alsu make hamster a good, mosi frequently used model, especialiy for topical anti androgen research.

The hamster flank organs are paired, oval, darkiy pigmented structures IocateÉ on each flank of the animal, which is also called costovertebral organ. Thes4 pigmented costovertebra spots have been reported to be maIe secondary se, characters in 1940's (Kupperman, 1944). The organ in the mature male hamstei measures approximately 6 mm in diameter, is heavily pigmented and coverec with coarse, dqk hairs. The organ of the female animal is less developed ù about 2 mm in diameter. It is lightly pigmented, with only very dark hairs and is at times difficult to distinguish. The flank organ is composed of three androgen dependent structures: pigment cells, hair follicles, and sebaceous gland3 (Hamikon and Montagna, 1950). Castration caused a rapid atrophy of this orgar as well as underlying sebaceous glands. These changes could be readily reversa by injection of testosterone pmpionate (TP), which stimulated growth and pigmentation of flank organs in immature males and fernales. The chiei changes observed follming androgenic stimulation are: 1) mqssive inaease ir the amount of sebaceous tissues as a result of enlargement of existent aiveolj and the. budding of additional alveoli; 2) increase in the size of individwl ubaaais cells; 3) w~ointime stajning of the acidophilic rytoplasrnic granules, 4) appearance of conspicuous Lipid droplets rwounding the nucleus; 5: progressive increase in the size of droplets and in the number of celis thi contain large droplets; 6) coalescence of droplets to form a lipid mass that lies o the side of the ce11 between the nucleus and the lumen of the acinus; 7) pyknos: of the nuclei and their displacement in the direction opposite to the lumen c the acinus; 8) arrangement of the glandular cells in strata according to sizc staining reactions, and content of lipid droplets; 9) increase in the number c deeply acidophilic cells near the Iumina of the acinus; 10) enlargement of tk lumina of the acini, acinar ducts and hais follicles, dong with the distention t these lumi- with sebum and cellular debris; and Il) openings in the walls c adjacent acini that allow a continuity of contents. Androgens stimulate th growth of coarse black hairs in the custovertebra spot and eventually result i coarsening of hairs of the general skin. Melanogenesis is enhanced in th costovertebral spot, especially in the hairs and in the dermis near th pilosebaceous apparatus (Hamilton and Montagna, 1950).

Recently, Seki and his colleagues have isolated a cDNA library from flan organs of male hamsters and screened the library by a different hybridizatio method using cDNA probes from normal and, castrated males (Seki et al., 1991 The expression of cDNA clone, termed FAR-I 7a, was found to be highl sensitive to androgen. FAR-17a mRNA of 3.8 Kb was reduced after castratio and reappeared after T treatment. Among several tissues examined, FAR-17 gene was expressed at a high level in flank organ and a low level in testis an earlobe. FAR-17a probe detected a few fragments in genomic DNA of hamste: mouse, suncus, pig, and human, suggesting that this gene is phylogeneticall conserved. The sequence of FAR-17a cDNA predicts a protein of 231 a- acid (27,216 doltons) having basic properties. The deducted protein has no signhcar homologies to proteins previously desaibed.

Systemic and local effects of topical antiandrogens can both be studied using thi model, and the presence of paired organs provides a built-in control for th study of topical applications to one side. Local effed should lirnited on the sid of application while systemic effects can be obçerved on both sides and on th prostate. 2.5.1.3 Hamsfer ear

The ventral side of the ear is richly studded with sebaceous glands, the gland closely resemble sebaceous follicles in human since they are fairly largg androgen sensitive, and have a turnover time that is similar to that of man. I. fact, each sebaceous gland unit consists of an infundibular canal, several larg lobulated acini with 6 to 10 layers of sebocytes, and one centraliy located pilar unit in the ear. Each earlobe contains numerous such glands and, with the low power objective, 8 to 15 glands can be seen in routine histologie section. Th glands in male hamsters are large -and have an average site of 0.175 mm*. îhis 1 comparable to 0.2175 mm2 whidt has been previously calculated for huma: forehead sebaceous glands (Plewig et al., 1971b). Basal cells and differentiatin, lipid cells are easily recognized and normaliy the undifferentiated ce11 pool i small (Plewig et al., 1971b; Plewig et al., 1971a). The thymidine incorporatio; data demonstrated that labeled basal cells are still seen in appreaable numbers a day 14 in male. Similar observations were made in hurnan sebaceous gland! The stratuxn comeum turnover time (or transit tirne) for Syrian hamster ea epidermis is 12 to 14 days, and therefore close to the estimated value for th sebaceous glands. In man, epidermal stratum corneum turnover time is also 1 days, at least for several areas of glabrous skin (Plewig and Ludersdunidt, 1977).

However, the control of the endogenous androgen secretion of the Syriai hamster by photoperoidieity must be considered (Luderschmidt et al., 1984b: Long photoperoids in summer lead to sexual development with high T level ii plasma. In autumn and winter, the reproduction activity is reduced, the plasm T concentration is diminished, and the sebaceous gland activity is reduced to minimum (Luderschmidt et al., 1984b). Moreover, studies in animal mode1 should stimulate our interest in the mechanism of action of steroid hormon effects on the skîn. Although the potential for a new drug is exciting, translatioi to human use must be made with caution. The screening of potentialll antiandrogenic compounds shed some îight in their mechanisms of action However, antiandro8enici& in the hamster, for example, would not necersad; imply equal efficaq in humans. in addition, the success or failure may not be ii the selection of the antiandrogen but in the selection of the vehicle in which i is deiivered. For example, different vehicle may show absolutely ciifferen results. Finally, the antiandrogenic effect on one skin structure, such as th, sebaceous gland, is not necessarily equivalent to the effect on another, such as the hait follicle. Thus, a topical antiandrogen that may be effective in acne therapy may not work in treating hirsutism or baldness.

Androgen-dependent mouse mammary carcinoma SC115 was established in 1964 by Minesita and Yamaguchi (Minesita and Yamaguchi, 1964). The original hunor arose spontaneously as an rndrogen-independent i de no car ci no ma of mammary origin in a female DS mouse and grew equaliy well when transplanted into male and female mice. After passage in male DS mice for 19 generatioxq the tumor was found to be androgen dependent, defined by its failure to grow in female or castrated male mice and by its ability to grow in female or castrated male mice given androgen. It was also shown that even up to 70 days after transplantation, androgen administration caused growth of tumors that had been dormant without visual growth in the intact female hosts. The same evidence was obtained when the tumor was implanted in newbom mice. Growth was seen only in the males after sexual maturation. Development of the tumor in the intact male was markedly suppressed by the administration of estrogen. Howwer, it has been shown that pharmacological dose of estrogen stirnulated the tumor growth (Noguchi et al., 1985). This suppressive effect in the tumor growth was obvious in the castrated and then androgenited male as well as in the androgen treated female. Cortisol and progesterone administration also showed effect on the tumor growth, but in a lesser degree. Moreover, after initiation of growth by androgen, the subsequent fate of SC115 turnor following androgen removal dependent on the size of the tumor: small, medium, and large tumors showed complete, temporary, and no regression, respectively (Kitamura et al., 1978). In contrast with the partial and no androgen dependency of medium and large tumors in situ, *or seeds taken from the medium and large tumors before androgen removal grew only in males when transplanted into male and female mice (Kitamura et al., 1978). It has been demonstrated that, in SCllS, bth T and DHT can bind to specific androgen receptor, though the binding affinity of DHT is several leshigher than that for T (Yamaguchi et ai., 1978). Sa-Reductase activity has been obse~edin the shionogi tumor homoge~teand slia (Yamaguchi et al., 1978). Ceils derived from this tumor retain their androgen responsiveness in vivo and in ceIl (Kitamura et al., 1978; Labrie and Veilleux, 1986; Labrie and Veilleux, 1988; Labrie et al., 1988b; Labrie et ai., 1988~).The culture cells can be transformed into Shionogi mice to develop a visible tumor. In fact, the intact mice whkh were inoculated with SC115 cells developed tumors after 12 days of inoculation (Bélanger et al., 1985). Moreover, adrenal precursors may also have a positive effect on the tumor growth (Minesita and Yamaguchi, 1965; Goff and Belanger, 1984; Bélanger et al., 1985). THE IMPORTANCE OF LOCAL STEROID BIOSYNTHESIS IN TH1 HAMSTER In the past 15 years, a discovery of major importance in the field of sex steroic and their physiological and pathological roles is that humans and some othi primates are unique arnong animal speaes having adrenals that secrete lari amounts of the inactive precursor steroids DHEA, its sulfate (DHEAS),and A dione, which could be converted into potent androgens and estrogens in tl peripheral tissues. In fact, adreiial precursors response 30-50% of androgens I man, 75% of estrogens before menopause, 100% after menopause in women. 1 order to better understand the important role of adrenal steroids in tf formation of active androgen Dm,we have studied 'the tissue distribution 4 enzymatic activity of steroidogenic enzymes involved in sex steroi biosynthesis, namely steroid sulfatase, 313-HDS, 17G-HSDf and 5a-reductase in 1 tissues of both male and female hamsters, and the results are shown in the fir article. in the second article, we have investigated the effects of adrenaI steroic UHEA and ~4-dioneby constant releasing from silastic implants on androgei sensitive parameters cornparhg with active androgens T and DH'F. In the thii article, we have demonstrated the systemic stimulatory effects of DHEA and B changes of enzymatic activitieç in the skin, prostate, and uterus aftc percutaneous administration in both male and female hamsters. The da, confirmed the importance of local steroid biosynthesis in the hamster and ah ruggested that hamster is a good mode1 for investigating the control of tl formation and the action of androgens. Distribution of stemidogenic enzymes in pedpheral intracrine tissues t the hamster

Cailin Chen, Céline Martel, Femand Labrie.

MRC Group in Molecular Endocrinology, CHUL Research Center and Laval University, Québec, GIV 4G2, Canada ABSTRACT

It is welî recognized that human adrenals seaete large amounts of th precursor steroids dehydroepiandrosterone @HEA) and especially DlEA-sulfat @HEA-S), which requires the presence of steroid sulfatase, 3B-hydroxysteroi( dehydrogenasd~s-~4iwmerase (3&HSD), L7B-hydroxysteroid dehydrogenas (17B-HSD), Sa-reductase, and aromatase to form the active androge~ dihydrotestosterone @HT) and estrogen 17i3-estradiol (E2)in the gmds as wej as extragonadal tissues. Hamster is a common model used in recent years fa screening the antiandrogenic compounds. In order to inaease our knowledg about this animal model, we have studied the tissue distribution of th' enzymatic activity of the above-mentioned steroidogenic enzymes in both mal' and female hamsters. The activities of steroid sulfatase using both DHEA-S ani E1-S as substrates, 313-HSD using DHEA as substrate, 1713-HSD using both A4 dione and El as substrates, Sa-reductase using testosterone as substrate, anc aromatase using Ar-dione as substrate were measured in a11 the tissue examined, namely adrenal, testis, ovary, liver, kidney, ventral and dorsa prostates, seminal vesicle, uterus, vagina, ventral and dorsal skin, ear, and fiad organ. The widespread tissue distribution of the steroidogenic enzymes indicate the importance of the local formation of active androgens and estrogens ii peripheral intracrine tissues in the hamster. The hamster flank organs are complex structures, located on each fiank ol the animal, and consist mainly of sebaceous tissue [l,î]. Like sebaceous glands ir other species, the hamster fiank organs are highly androgen dependent, Castration of male hamsters results in a deaease in the volume and weight ot these organs, while systemic administration of testosterone (testo) oi dihydrotestosterone (DHT)restores their normal size and weight. The local antiandrogenic activity of a compwnd applied topically to one flank organ of a castrated androgen-treated male hamster can be evaluated by its ability tc counteract the stimulatory activity of androgens on its size or weight, on the sizé of underlying sebaceous glands, on DNA synthesis, namely thymidin6 incorporation [2,3], or on incorporation of labeled acetate into the lipids [4]. The effect should be limited on the treated side. The systemic antiandrogenic activi9 can be determined either by evaluating its effecîs on the contralateral untreated fiank organ and/or by measuring the prostate weight. Also, due to its relatively inexpensive to purchase and care for, easy of access and easy to work with, th3 animal model has been widely used to test the potency of androgens and antiandrogens, especially the antiandrogens for topical use related to th€ androgen associated conditions such as acne, hirsutism, and male patterE baldness [4-7,

Humans, along with other non-human primates, are unique in havine adrenals that secrete large amounts of the precursor steroida dehydroepiandrosterone (DHEA)and especially DHEA-sulfate (DHEA-S), wbich are converted into androstenedione (A4-dione) and then into potent androgem and estrogens by steroid sulfatase, 3a-hydroxysteroid dehydrogenase/~5-~~ isomerase (38-ZISD), 1713-hydroxysteroid dehydrogenase (1713-HÇD),Su-reductase, and aromatase in peripheral target tissues [8,9]. Having a high secretion rate ol adrenal precursor sex steroids 'is thus completely different from animal models currently used in the laboratory, nameiy, rats, mice, guinea pigs, hamsters etc., in which the formation of anàrogens and estrogens takes place exclusively in the gonads [9,10]. Due to the uneasy of access of human spedmen, most studies luve been done on experimental animals. It is thus important to use a good animai model which b dose to human situation: Previously, it has been nported thal constant release of adrenal steroids DHEA and Ad-dione by silastic implants caused an important stimulation of androgen-dependent gene expression an prostate growth in castrated rats [11,12]. We have aiso observed that tl implantation of DHEA or ~4-dionein orchiectomized hamsters reversed tl inhibitory effect of the castration on flank organ and prostate growt (unpublished data). It has been suggested that exogenous sex steroid precurst cm be transformed into active androgens and estrogens in peripheral tissues j rat and hamster, although, these animal5 do not seaete large amount adreni precursors [8,9,13].

In order to increase our knowledge about the local steroid biosynthesis i hamster, we measured the activities of the steroidogenic enzymes involved i sex steroid synthesize pathway in gonadal as well as peripheral intracrine tissui of the male and female hamsters. MATERIALS AM) METHODS

Male and female adult Syrian hamsters (110-12Og) were purchased from Charles-River Laboratories (St-Constant, Quebec, Canada) and kept in a lighl (14h light/day) and temperature (22 f 1°C) controlied environment for one week before saaifice. Animals were küled by cervical dislocation. The tissues were quickly removed, freed from fat and other adhering tissue, frozen in dry ice, and kept at 180% until used.

Frozen tissues were homogenized with a Polytron in phosphate buffer (2C mM KH2PO4, 0.25 M suaose, 1 mM EDTA, pH 7.5) containing protease inhibitors (1 mM phenylmethylsulfonyl fluoride and 5 &ml each of pepstatin A, antipain and leupeptiri) and centrifuged for 30 min at 1OOû X g to remove ce11 debris. Ali procedures were performed at 4°C and the supematants were saved in eppendoff and stored at -80°C before assayed. Protein content of tissue homogenate was meastlred by the method of Bradford using bovine seruni albumin as standard [14].

Sa-red *I Aiiquots of the 1ûûO X g supernatant were incubated for 150 min at 37OC water bath in a total volume of 0.5 ml phosphate buffer (12.5 mM KH2PO4,I mM EDTA, pH 7.5) containing 0.5 PM of 14C-labeled substrate and 1 mM of t. appropriate cofactor(s). Three O-HSD activity was measured with [4-'4C]-DHEA (S.A. 51 mCi/mmoI) and NAD+; estrogenic 17S-HSD activity was measured using [4-14C]-El (S.A.51 mCi/mol) while androgenic 17B-HSD activity wa2 studied using [4-14C]-d4-dione, both substrate being used with the cofactorr NADH + NADPH; Sa-redudase activity was tested with [4-l%]-testosterone (S.A 51 mCi/mmol) and NADPH. Labeled radioactivity was purçhased from Neu England Nuciear/Dupont (Markham, Ontario, Canada) and purified by thir layer chromatography (TLC) before use. The enzymatic reactions were stopped by chilling the incubation mixture in an ice-water slurry and adding 3 ml oj diethyl ether. The component were then rnixed, separated by centrifugation and fiozen in a dry ice ethanol bath. The organic phase was kept while thr remaining frozen aqueous fraction was reextracted once with eh. The twc organic phases were then pled and evaporated to dryness under a nitroger Stream. Ail components were then separated on TLC (60Fw silica gel, E. Merck Darmstadt, F.R.G.) using the solvent system containing 4:l (v/v) toluene acetone before autoradiography for 48h. The metabolites revealed b: autoradiography were identified by cornparison with standard steroid charaderized by HPLC as describeci previously (151. The TLC areas correspondiq to the substrate and product(s) of the enzymatic reaction, namely DHEA, A4 dione, El, E2, testo, Dm,androstane-3,17-dione, androsteneSa, 178-diol, an{ androstene-38,17$dol were collected and transferred into scintillation via1 containing 0.5 ml ethanol. After adding 10 ml scintillation liquid, th radioactivity was measured in a spedrophotometer. Control samples weri processed identically except that buffer was used instead of the homogenate.

AU measurements of enzymatic activities were performed with individus tissue samples obtained from 6 male or 6 female hamsters as described above The recovery of [IdCl-radioactivity varied between 80 to 90% of the radioactivit; initially added as substrate and was taken into consideration to calculate thi rates of product formation. Moreover, dilutions of the homogenate were uset for some tissues in order that, under the conditions of the assay, the maxima percentage of conversion of the substrate into product(s) was inferior to 25% The formation of hydroxylated products and water-soluble metabolites, such a; sulfates and glucuronides was not studied.

Aromatase activity was measured with a radiometric assay, which takêi advantage of the stereospecific loss of hydrogen from the C-lB position o androstenedione duting aromatization [Io]. The incorporation of tritium fron [lB, 28=)H(N)1-Ar-dione into 3Hz0 is thus proportional to the amount O estrogen formed. Briefly, tissue homogenates were incubated with 10 nM [lB, 2i3 3~(N)]-~uone(S.A. 4-43 mCi/mmol.) and phosphate buffer (12.5 mM RX2PO4 1mM EDTA, pH 7.5) in a total volume of 0.5 ml containing 1mM NADPH as i cofactor was incubated for 150 min ai 3w. The reaction was stopped by addini 0.5 ml cold active charcoal solution for 4 h at 4OC. The mixture was ther separated by centrifugation and aliquot (800N)of the resultant supernatant wa! transferred into scintiîiation viais and measured in a spectrophotometer aftei adding 10 ml scintillation cocktail. Aliquots of the supernatant were incubated in a shaking water bath for 150 mi at 37"C, in 0.5 ml Tris-acetate buffer (0.1 M Tris-acetate, 5 mM EDTA, 10' glycerol, pH 7.0) containing 10 nM of [~.~.~.~~H(N)]-DHEA-S(S.A. 75. mCi/mmol) or [~.~~H(N)]-E~-S(S.A. 49.0 mCi/mmol). The enzymatie reactio - was stopped by chilling the incubation mixtur= in an ice-woter slurry, rdding 32 pM of DHEA-SO4 or El-SO4 to sahuate the enzyme and 4 ml of diethyl ether t extract the organic component. After œntrifugation, the test tubes were froze in a dry-ice+thanol bath. The liquid organic phase containing [SHI-DHEA c [3H]-El was transferred into scintillation vials containing 10 ml sanollatio fluid and the radioactivity was measured in a spectrophotometer. Results ai expressed as means f SEM in fmoles, pmoles or nmoles product formed/m protein/min for al1 the enzymes. RESULTS

The most potent androgen has been recognized to be dihydrotestosteronl (Dm), of which the biosynthesis from the precursor sex steroid DHEA-S i catalyzed by four enzymes, namely DHEA-sulfatase, 38-HSD, androgenic 17% HSD, and Sa-reductase. The first step for androgen formation is the hydrolysi of DHEA-S into DHEA by . DHEA-sulfatase. DHEA-sulfatase activity wa rneasued in all the tissues examined (Fig. 1).in male hamster, relatively higl levels of DHEA-sulfatase attivity were found in the adrenal, liver, kidney, an1 seminal vesicle, with respective values of 0.62 f 0.08, 1.63 f 0.12, 1.14 f 0.05, anc 0.62 f 0.09 fmol/mg protein/min, while lower Ievel of 0.066 fmol/mi protein/min was measured in the testes. On the other hand, comparable level of DHEA-sulfatase activity, ranging from 0.11 to 0.88 fmol/mg proteidmir were found in female hamsters tissues, the liver having the highest DHEA sulfatase activity (1.58 f 0.27 fmol/mg protein/min).

The second step in androgen synthetic pathway is the conversion of Dm into Ad-dione by the 3%-HSD. The levels of 3B-HçD activity measured in crud, homogenate obtained from different tissues of male and female hamsters art presented in Fig. 2. The highest levels of 3i3-HSD activity was measured in mal1 (327.07 f 20.9 pmol/mg protein/min) and female (407.66 i 22.03 pmol/m; protein/&) adrenals followed by the testis and ovary with respective 3B-HSI activity of 53.38 f 5.3 and 262.89 f 46.27 pmol/mg protein/min. These dati highly corresponded to the classical endocrine theory in which these targe organs can synthesize large amount of sex steroids. However, it is of interest tc find 31FHSD activity in many peripheral tissues. Thtee iEHSD activity of 102.7: f 8.36 and 60.57 f 5.32 pmol/mg proteinhin was measured in male and femd liver, respectively, while levels of 29.42 f 3.1 and 45.68 f 1.37 pmol/ml prbtein/min were observed in male and female kidneys, respectively. Lov levels of 38-HÇD activity ( ~0.1pmol/mg protein/min) were found in othe peripheral organs studied, namely hamster flank organ, ear, dorsal and ventra skin, dorsai. and ventral pnWtateû, seminal vesides, utents and' vagina.

Ihe mxt step in andmgen formation requkes the conversion of ~4-dion, into testosterone by the andmgenic 178-HSD. As illustrated in Fig. 3, androgeni 17B-HSD activity was measured in ail the male and female tissues examined. 1 fact, 170-HSD activity of 94.46 f 9.60 and 9.52 f 1.50 pmols/mg protein/min wz measured in the male liver and kidney, respectively, while activities of 576.9 58.79 and 100.78 f 14.27 pmols/mg protein/min were obtained in female livc and kidney. bwer levels of 17g-HSD activity was measured in the other tissut studied (0.1 to 5 pmol/mg protein /min).

The fuial step of the formation of active androgen requires the conversio of testosterone into DMby the Sa-reductase. in male hamster (Fig. 4), U' highest level of Sa-reductase activity was measured in the liver (71.50 I1.5 prnol) and kidney (68.68 f 5.44 pmol), while the lowest level of Sa-reductai activity was measured in testis (0.11 f 0.01 pmol/mg protein/rnin). Similar dai were obtained in female hamster with the highest levels of Sa-reductase activil measured in liver (135.30 f 16.11 pmol) and kidney (29.60 f 6.20 pmol) whil lower levels of Sa-reductase activity (0.40 to 2.25 pmol/mg protein/min) we~ detected in the other tissues. The present data suggests that the adreni precursors can be transformed into the most active androgen DHT in al1 tk male and female hamster tissues examined.

Five enzymes are involved in estrogen synthetic pathway, namely DHEl sulfatase, 38-HSD, estrogenic 178-HSD, aromatase, and estrone sulfatase. Th tissue distribution of DHEA sulfatase and 3e-HSD has already been describe above since these two enzymes are also needed for androgen formation. Ne] step of estrogen formation requires one key enzyme, namely the aromatasc which catalyses the transformation of A4-dione into estrone (El) and testo in1 estradiol (E2). In the male, the highest aromatase activity was measured in e: (5.80 f 0.70 fmol/mg proteidmin) followed by the dorsal skin (1.34 f 0.1 fmol/mg proteidmin). In the female, as expected, the highest level c aromatase activity was measured in the ovary at 6.50 f 0.70 fmol/m proteidmin followed by the eat (1.79 f 0.20 fmol/mg protein/min), while th levels of aromatase activity ranged from 0.08 to 0.36 fmol/mg protein/min i the other tissues.

It is weii known that the most active estrogen is estradiol. The next step i estrogen formation is thus the. conversion of El into E2 by the estrogenic 171 ED. As illustrated in Fig. 6, estrogenic 17B-HSD activity was measured in al tissues examined. in the male, the highest levels of 178-HSD activity wen measured in the adrend, testis, liver, and kidney with respective activities ol 3.47 f 0.37, 14.82 f 1.50, 11.14 f 1.50, and 18.27 f 0.83 pmoIs/mg proteinhin, while activities of 2.74 f 0.17,23.97 f 4.49,38.83 f 1.60, and 48.26 f 1.52 pmols/mg protein/min were obtained in the female adrenal, ovary, liver, and kidney respectively. Lower levels of 17MISD activity were shown in the other tissue! (c 1.0 frnol/rng protein/min).

&&O& cm &O be released foliowing hyârolysis of El-S by steroid suifatase In fact, as observed for the transformation of DHEA-S into DHEA, the enzymatil conversion of El-S into El was found in al1 male and female tissues examinec (Fig. 7) and the levels of enzymatic activity did not show significant variatior between tissues. DISCUSSION

in the present study, we demonstrated the widespread tissue distribution oi steroidogenic enzymes requked to hmfonn DHEA-Sinto the bioactive steroid! DHï and E2. In fact, all tissues examined, namely adrenal, testis, ovary, liver kidney, dorsal and ventral prostates, seminal vesicles, uterus, vagina, flad organ, ear, and dorsal and ventral skin possess the enzymatic activities necessaq to localiy produce DHT and Ez from the precursor Dm-S. Howevet there art significant variations of enzymatic activities for different enzymes or/anc different tissues with the method used. The purpose of this study is tc demonstrate the existence of the enzymes in these tissues rather than providr complete kinetic data. Since the protein concentrations lead to a 20-30% oi product formation for each enzyme in each tissue and pHs of the reaction buffei set at 7.5 which is close to' hurnan and most animal homeostasis, the data present should give a due to estirnate the levels of the enzymatic activities ir these tissues.

Steroid sulfatase activity catalyses the hydrolysis of sulfuric acid esters oi some neutral steroids such as DHEA or El. Circulating sterylsulfate are believed to be biologically inactive, as they do not bind to known steroid receptor, and are considered as an important reservoir of biologically active steroids. These negatively charged conjugates can gain access to and be hydrolyzed by the sterylsufatase. There is increasing evidence thaï steroid sulfation and desulfation play a primordial role in intracrine control of stetoid hormone action [9, 13, 17, 181. DHEA-S sulfatase has been detected in the hanovary, &in, and uterus 1191 as well as in benign prostatic hyperplasia (201, and MCF-7 human breas! cancer cells (211. The activity of El-S sulfatase has been reported in rat and human liver [22, 231 brain 1241, pituitary [24], and breast carcinoma [21, 25). The activity of sulfatase isolated from human placenta was observed using both DHEA-S and El as substrates (261. The present data demonstrated that all of the 14 tissues examined possess steroid sulfatase activity, however low levels of enzymatic activity were observed for both substrates in male and female hamsters. The present study does not provide data that indicate whether hydrolysis of El-S and DHEA-S results from the action of a single or twa düferent enzymes. Previously, we have observed that the enzyme affinity for El- S substrate was higher than for DHEA-Sin rhesus rnonkey tissues [27] . The membrane-bound enzyme 3fi-HSD catalyzes an essential step in t: transformation of al1 5-pregnen-3Sol and 5-androten-311-1 steroids into thc corresponding ~4-3-ketosteroids,namely progesterone as weU as al1 t: precursors of &drogens, estrogens, glucocorticoids and mineraiocorticoids. addition, 3L-HSD is responsible for the interconversion of 313-hydroxy- and keto-Sa-androstane steroids. In mammals, the 313-IISD isoenzyrnes are express not only in classical steroidogenic tissues, namely the adrenal cortex, ovai testis, and placenta [28-33], but also in several peripheral tissues induding f skin, liver, adipose tissue, i

The enzyme 17B-HSD is responsible of the interconversion of androstenedione (~4-dione)and testo, El and E2, IlHEA and 5-androsene-38,17 di01 (%ne-diol) whidi are essential to the development, growth, and functic of al1 tissues responsible for reproduction and fertility in men and wome Previously, we have reported the ubiquitous tissue distribution of 170-XiE enzyrnatic activity in human [36], rat [36] and rhesus monkey [27]. The prese results show that estrogenic and androgenic 17l3-HSD is also presented in i hamster tissues exarnined.

The enzyme steroid Sa-reductase catalyzes the reduction of 4-ene- oxosteroid to the corresponding 3-0~0-Sa-dihydrosteroids.Its best hown role the transformation of testo into DHT, the most potent androgen, which responsible for differentiation of the male external genitalia, prostate and 0th accessory sex organs. Sa-reductase may also be involved in androgen-depende disorders such as benign prostatic hyperplasia, acne vulgafis, female hirsutisi seborrhea, male-pattern baldness, and pseudohermaphroditism [37-431. It h been suggested that the ktbloekade of an&ogen action was achieved by usù an antiandrogen and a Sa-reductase inhibitor 144-463. There is no doubt that tl enzyme piays a major role in active androgen formation. in recent years, mu investigetors have wosked on the Joning of corresponding genes, expressio and tissues distribution of this enzyme [47-491 as well as inhibitois of 5( reductase [50-561. ~reviousreports have demonstrated the iresence of 5i reductase in hamster prostate and flank organ [57,58]. The present study shows a widespread tissue distribution of this enzyme in steroidogenic as weli as peripheral hamster tissues. nie presence of Sa-reductase activity has previously been reported in the prostate, fat, liver, skin, brain and adrenal in human, rat, or rhesus monkey [52,59-61].

The cytochrome P-450 mmatase enzyme converts ~4-dioneand testo to El and E2, respectiveiy. In the present study, using ~4-dioneas substrate, we have measured arornatase activity in ai1 the tissues examined. It is interesthg ta mention that high levels of activity were measured in the male skin, flank organ, and ear, while small difference between organs was observed in female hamster, with the exception of the ovary. It may indicates that the skin is the major site for metabolizing the estrogens in male. The presence of aromatase activity has previously been reported in skin, brain, adipose tissue, beast, cervix and endometriun [6268].

In conclusion, the steroidogenic enzymes, namely steroid sulfatase, 38- HSD, 178-HSD, Sa-reductase and aromatase, which are required for the formation of active androgens and estrogens, are present not only in classical but also in a series of peripheral intracrine tissues in hamster. The presence of these steroidogenic enzymes in peripheral tissues indicates the major importance of local steroid biosynthesis. The present results obtained in the hamster are similar to those found in human and other animals models which suggest that the hamster is a good model for stydies on intraainology (131 and espeaaily for development of novel therapeutic approaches for the.treatment of hormonal- sensitive diseases. REFERENCES

Hamilton, J.B. and Montagna, W.: ïheesebaceous glands of the hamtei Morphological effects of sndrogens on integumentary structure. Am. j Anat. 86 (1950) 191-233.

Frost, P. and Gomez, E.: Inhibitors of sex hormones: development a experirnental models. Adv. Bi01 Skin 12 (1972) 403413.

Frost, P., Giegel, J.L., Weinstein, G.D. and Gomez, E.C.: Biodynami studies of hamster flank organ growth: hormonal influences. J hues Dermatol 61 (1973) 159-167.

Lutsky, B.N., Budak, M., Koziol, P., Monahan, M. and Neri, R.O.: Th1 effects of a nonsteroid antiandrogen, Flutamide, on sebaceous gland. ) Invest. Dermatol. 64 (1975) 412-417.

Burdick, K.H. and Hill, R.: The topical effect of the antiandrogei chlormadinone acetate and some of its dinical modifications on th( hamster costovertebral organ. Br. J. Dennafol. 82 (Suppl. 6) (1970) 19-25.

Luderschmidt, C., Eiermann, W., Jawny, J., Bidlingmaier, F. and Ring, J. 17a-propylmesterolone (SH-434): an antiandrogenic sebosuppressivi substance not influencing circulating testosterone concentrations Experimental studies in Syrian hamsters. Nouynyn Schmiedebrgs Arch Phannacol. 328 (1984) 214-218.

~robks,J.R., Primka, R, Berman, C., Krupa, D.A., Reynolds, G.F. ant Rasmusson, G.H.: Topical anti-androgeniaty of a new 4-azaçteroid in thc hamster. Steroids 56 (1991) 428-433.

Adams, J.: Control of secretion and the function of C19-delta-5-steroids O the human adrenal gland. Mol. CeIl Endocrinol, 45 (1985) 1-17. hbrie, P., Dupont, A. and Belanger, A.: Cornpiete androgen bloclade foi the treatment of prostate cancer. In Important Advances in Oncolog! (Edited by V.T. de Vita, S. Heiiman and S.A. Rosenberg). J.B. Lippincott Phiiadelphia (1985) 193-217. Bélanger, B., Bélanger, A., Labrie, F., Dupont, A., Cusan, L. and Weth G.: Cornparison of residual Cg19 steroids in plasma and prostatic tissue c human, rat and guinea pig after caskation: unique importance c extratesticular androgens in men. 1. Steroid Biochem. 32 (1989) 695-698.

Labrie, C., Bélanger, A. and Labrie, F.: Androgenic activity c dehydroepiandrosterone and androstenedione in the rat ventral prostatc Endocrinology l23 (1988) 1412-1417.

Labrie, C., Sid,J., Piao, H.F., Bélanger, A., Pelletier, G. and Labrie, F Stimulation of androgen-dependent gene expression by the adren; precursors dehydroepiandrosterone and androstenedione in the rz ventrai prostate. Endocrinology 124 (1989) 274502754.

Labrie, F.: Intracrinology. Mol. Cell. EndoMnol. 78 (1991) C113-C118.

Bradford, M.: A rapid and sensitive method for the quantitation c microgram quantities of protein utilizing the principle of protein-dy binding. Aml. Biochenr. 72 (1976) 248-254.

Thériault, C. and Me,F.: Multiple steroid metabolic pathways in ZR-7: 1 human breast cancer cells. J. Steroid Biochem. Molec. Biol. 38 (1991) 15; 164.

Thomas, J.P. and Oake, RJ.: Androgen metabolism in the skin of hirsut women. 1. Clin. Endocrinol. Mefab. 38 (1974) 19-22.

Hobkrik, R.: Steroid sulfotransferases and steroid sulfatases: characteristic and biological roles. Can / Biochem Cell Bi01 63 (1985) 1127-1144.

Roy, A.: Regulation of steroid hormone action in target cells by soecifi hormone-inactivating enzymes. Porc Soc Exp Bi01 Med 199 (1992) 265-272

Hanhg, R.J., Hadcett, R, Boothroid, R and Canick, J.: Steroid sulphatas activity in the human ovarian corpus luteum, stroma, and folliclc comparisom to activity in other tissues and the placenta. 1 Steroi Bioch36 (1990) 175-179. Kiein, H., Molwitz, T. and Bartsch, W.: Steroid sulfate sulfatase in humar benign prostatic hyperplasia: characterization and quantification of thc enzyme in epithelium and stroma. J Steroid Biuch33 (1989) 195-200.

MacIndoe, J.: The hydrolysis of estrone sulfate anc dehydroepiandrosterone sulfate by MCF-7 human breast cancer cells Endocrino1 123 (1988) 1281-1287.

Prost, O., Ottignon, Y., Remy-martin, A., Vuitton, D., Miguet, P. anc Adessi, G.: Steroid sulfatase activities in nonmal and cirrhotic livers anc plasma levels of estrone sulfate, estrone and estradiol-178 in men Sferoids 43 (1984) 189-199.

Eriksson, L., Nilsson, B., carlstrorn, K., Oscarsson, J., Eden, S. and Vor Schoultz, B.: Secretory pattern of growth hormone regulates steroic sulfatase activity h rat Iiver. 1 Seroid BioJiem 33 (1989) 413-416.

Comolly, P. and Resko, J.: Estrone sulfatase activity in rat brain anc pituitary: effects of gonadectomy and the estrous cycle. J Sferoid Biochen 33 (1989) 1013-1018.

Pasqualini, J., Schatz, B., Varin, C. and Nguyen, B.: Recent data or estrogen sulfatases and sulfotransferases activities in human breas< cancer. J ~eroidBiochem 41 (1992) 323-329.

DiabbeIt, L. and Kuss, E.: Human placental sterylsulfatase interaction oi the isolated enzyme with substrates, products, transition-state analowues amino-acid modifiers and anion transport inhibitors. Biol Chm Hoppe Seyler 372 (1991) 173-185.

Martel, C., Mehr, M.H, Ga@, D., Simard, J. and Labrie, F.: Widespreac tissue distribution of steroid sulfatase, 3P-hydr ox ys ter0 id dehydrogenase/A 5-A 4 immerase (3P-HSD), 17B-HÇD Sa-reductase and aromatase activities in 'the rhesus monkey. Mol. Cell. Eniiocrinol. 104 (1994) 103-111.

Luu-The, V., Lachance, Y., Labrie, Cn, Ublanc, Thomas, J.L., Strickier, RC. and hbrie, F.: Puil le+ cDNA &chire and deduced amir~~acid sequence of human 3P-hydroxy-5-ene steroid dehydrogenase. Mol Endocrinol. 3 (1989) 1310-1312.

29. Lachance, Y., Luu-The, V., Labrie, C., Simard, J., Dumont, M., Launoit Y.d., Guerin, S., Leblanc, G. and Labsie, F.: Characterization of human 3P. hydroxysteroid dehydrogenase/~5-~4isomerase gene and its expressior in mammalian cas. J. Biol. Chem. 265 (1990) 20469-20475.

30. Lachance, Y., Luu-The, V., Verreault, H., Dumont, M., Rhéaume, E. Leblanc, G. and Labrie, F.: Structure of the human type II 3P- hydroxysteroid dehyd.rogenase/~5-~4isomerase (3b-HSD) gene: adre~ and gonadal speafiaty. DNA Ce11 Bi01 10 (1991) 701-711.

31. Rhéaume, E., Lachance, Y., Zhao, H.F.,Breton, N., de Launoit, Y., Trudel C., Luu-The, V., Simard, J. and Labrie, F.: Structure and expression of e new cDNA encoding the almost exclusive 3$-hydrox ys teroic dehydrogenase/~5-d4 isomeraw in human adrenals and gonads. Mol Endocrinol. 5 (1991) 1147-1157.

32. Simard, J., de Launoit, Y. and Labrie, F.: Characterization of the structure activity of rat type 1 and type iI 3$-hydroxysteroid dehydrogenase/A5 -A4 isomerase by site-directed mutagenesis and expression in HeLa cells. 1, Biol. Chem. 266 (1991) 14824-14û45.

33. Labrie, F., Simard, J., LUU-The,V., Bélanger, A. and Pelletier, G.: Structure, function and tissue-specific gene expression of 3P-hy droxy s teroid dehydrogemse/%ne-4-ene isomerase enzymes in classical and peripheral intracrine steroidogenic tissues. J. Steroid Biochern. Mol. Biol. 43 (1992) 805-826.

34. Lacoste, D., Bélanger, A. and Labrie, F.: Biosynthesis and degradation of androgens in human prostatic cancer ce11 lines. In Steroid Fomat ion, Degradation, and Action in Peripheral, Normal, and Neoplastic Tissue (Edited by H. Bradlow, L. Castagnetta, S. d'Aquino and F. Labrie). Ann. N.Y. Acad. Sa., (1990)389-391.

35. Dumont, M., Luu-The, V., Dupont, E., Pelletier, G. and Labrie, F.: Characterization, expression and irnmunohistochemical localiration of 3p-hydroxysteroid dehydrogenase /A~-A~isomerase in human skin. 1 Invest. Dermafol.99 (1992) 415421.

Martel, C., Rhéaume, E., Takahashi, M., Trudel, C., Couet, J., Luu-The, V Simard, J. and Labrie, F.: Distribution of 17s- h y drox y s t e r oi r dehydrogenase gene expression and activity and human tissues. )

Steroid Biochmi. Molec. Biol. 41 (1992) 597-603. . .

Labrie, F., Bélanger, A.,Simard, J., Labrie, C. and Dupont, A.: Combinatioi therapy for prostate cancer: endocrine and biological basis of its choice a new standard fint lire therapy. Cancer 71 (1993) 1059-1067.

Sansone, G. and Reisner, R.: Diffetential rates of conversion a testosterone to dihydrotestosterone in acne and in normal skin: 1 possible pathogenic factor in acne. J. lnvesf. Demafol. 56 (1971) 366-372.

Toscano, V.: Hirsutism: pilosebaceous unit dysregulation . Role a peripheral and glandular factors. J Endocrino1 Inoesf (1991) 153-170.

Kuttenn, F., Mowszowicz, I., Schaison, G. and Mauvais-Jarvis, Pt Androgen production and skin metabolism in hirsutism. J. Endocrinol. 7, (1977) 83-91.

Schweikert, H.U. and Wilson, J.D.:Regulation of human hair growth b steroid hormones: 1. Tstosterone metabolism in isolated hairs. J. Clin Endocrinol. Metab. 38 (1974) 811-819.

Imperato-McGinley, J., Guerrero, L., Gautier, T. and Peterson, R: Çteroic Sa-reductase deficiency in man: an inherited form of mal, pseud~hermaphroditisrn~Sciénce 186 (1974) 1213-1215.

Bringham, K. and Shaw, D.: The metabolism of testosterone by humai male scolp skin. 1Endmnol57 (1973) 111-121.

Matias, J., Malioy, V. 8nd Orentreich, N.: Synergistic antiandrogenic effect of topical combinations of Sa-seductase and androgen reœptor inhibitor in the hamster sebaceous glands. J Inmst Denna fol 91 (1988) 429-433. Labrie, C., Trudel, C., Li, S., Martel, C., Couet, J. and Labrie, F Combination of an antiandrogen and a Sa-reductase inhibitor: A furth~ step towaràs total androgen blockade? Edocrinology 129 (1991) 566-568.

Martel, C., Trudel, C., Couet, J., Labrie, C., Bélanger, A. and Labrie, F Blockade of androsténedione-induced stimulation of androgen-sensitiv parameters in the rat prostate by combination of Flutamide and 4-Ml - Moi. Cdi. Endocrirrol. 91 (1993) 43-49.

Andersson, S. and Russel, D.W.: Structural and biochemical properties c cloned and expressed human and rat steroid Sa-reductases. Proceedings 4 the Nathal Acadeniy of Sciences of the USA 87 (1990) 3640-3644.

Jenkins, E.P., Hsieh, C.L., Milatovich, A., Normington, K., Berman, D.M Francke, U. and Russel, D.W.: Characterization and chromosomi mapping of human steroid Sa-reductase gene and pseudogene an mapping of the mouse homoiogue. Genornics 11 (1991) 1102-1112.

Labrie, F., Sugirnoto, Y., Luu-The, V., Simard, J., Lachance, Y., Bachvaro~ D., Leblanc, G., Durocher, F. and Paquet, N.: Structure of human type 1 Sa-reductase. Endocrinolugy 131 (1992) 1571-1573.

Li, X., Singh, S. and Librie, F.: Synthesis ans in vitro activity of 17a-(E; alkul /arylformamido)-and 17i3-[(N-alkyl/aryl/arylamido]-4-rnethyl-4 rnethyl-4-aza-3-0~0-5a-androstan-3snesas inhibitors of human 5a reductases an dantagonists of the androgen receptor. 1. Med. Chem. 3 (1995) 11584173.

Rasmusson, G., Reynolds, G., Utne, T., Jobson, R., Primka, R, Berman, C and Brooks, J.: Azasteroids as inhibitors of rat prostatic Sa-reductase. Med Chem 27 (1984) 169û-1701.

Liang, T., Cascieri, M., Cheung, A., Reynolds, G. and Rasmusson, G Species differences in prostatic steroid Sa-reductase of rat, dog anc human. Endocrinui 117 (1985) 571-597.

Hirsch, K., Jones, C., Audia, J., Andersson, S., McQuaid, L., Stamm, N Neubauer, B., Penning, P., Toorney, R. and Russell, D.: LY191704: seledive, non-steroidal inhibitor of human steroidd Sa-reductase type PmNalt Acad Sci 90 (1993) 527705281.

Depre, M., Meeter, C., Hecken, A., van, L., OL, Buntinx, A. and çcheppe P.D.: Pharmacodynamies and rolerability of L-654,066, a steroid 5-alphi reductase inhibitor in man. Clin Phannacol Therap 52 (1992) 40942.

Lamb, J., English, H., Zavandoski, P., Rhodes, G., Johnson, R and bacs, ] Prostatic involution in rats induced by a novel Sa-reductase inhibitor, S & F-105657: role for testosterone in the androgenic response. Endocrini 13 (1992) 685-6944.

Lamb, J., Levy, M., Johnson, R and Isaacs, J.: Response of rat and huma prostatic cancers to the novel Sa-reductase inhibitor, Sk & F-10.565: Prostate 21 (1992) 15-34.

Takayasu, S. and Itami, S.: Hormonal control of Sa-reductase activity i hamster sebaceous glands. Endocrino1 Japon 29 (1982) 95-98.

Chen, C., Puy, L., Simard, J., Li, X., Singh, S. and Librie, F.: Local an systemic reduction by topical finasteride or flutamide on hamster flan organ size and enzyme acüvity. 1. laest. Dennafol. 105 (1995) 678-682

Herkner, K., Swoboda, W., Hodler, B. and Goedl, U.: Molecular biology ( androgen action: testoster~ne/dihydrotestosteronereceptor and androge 5-alpha-reductase in the human foreskin. J Steroid Biochem 24 (1986) 23! 243. tephart, E., Simpson, E. and Trzeciak, W.: Rat adrenal 5 alpha-reductas mRNA content and enzyme activity are sex hormone dependent. J Mi Endocrino16 (1991) 163-170.

Zyirek, M., Flood, C. and Longcope, C.: Salpha-reductase activity in r; adipose tissue. Proc Soc Exp Bi41 Med 186 (1987) 134-138.

Sawaya, M.E. and .Penneys, N.S.: Immunohistochemical distribution c aromatase and 3p-hydroxysteroid dehydrogenase in human hait follicl and sebaœous gland. 1. Cutmr. Pathl. 19 (1991) 309-314. 63. Roselli, C. and Resko, J.: Tesstosterone regulates aromatase activity in disaete brain areas of male rhesus macaques. Biol Reprod 40 (1989) 929- 934.

64. West, N., Roseili, C., Resko, J., Greene, G. and Brenner, R: Estrogen and progestin receptors and aromatase activity in rhesus monkey prostate. Endocrino1123 (1988) 2312-2322.

65. Kamiyama, H., Hirato, K., Akamatsu, T. and al, e.: Changes in steroid enzyme activities with age in human ovary. Nippon Sanka Fujinka Gaki ZClsshi 44 (1992) 419-426.

66. Miller, W.: Aromatase activity in breast tissue. J Steroid Biochem Mol Biol 39 (1991) 783-790.

67. Taga, S., Yoshida, N. and Sekiba, K.: Distribution and cydic change of aromatase cytochrome P-450 activity in human uteri. ] Steroid Biochem Mol Bi0137 (1990) 741-745.

68. Frost, P., Reed, M. and James, V.: The aromatization of androstenedione by human adipose and liver tissue. J Steoid biochem 13 (1980) 1427-1431. LEGENDS TO FIGURES

Figure 1. Tissue distribution of DHEA sulfatase activity in a series < steroidogenic and peripheral tissue from male and female hamster. Sulfatas activity was measured by the hydrolysis if PH] DHEA-S to DHEA. Imibatio were performed at 370C for 150 min with 10 nM labeled-substrate. Data ai present as means f SEM in fmoles DHEA formed/mg protein/& (log scale).

Figure 2. Tissue distribution of 3GHSD activity in a series of steroidogenic an peripheral tissue bmmile and female hamster. 3&=D activity was measure by formation of Ac-dione from [14C] DHEA. Incubation were performed at 370( for 150 min with 0.5 pM Iabeled substrate and 1rnM NAD+.Data are present a means f SEM in pmoles ~cdioneformed/mg proteidmin (log scale).

Figure 3. Tissue distribution of androgenic 17B-HSD activity in a series c steroidogenic and peripheral tissue from male and female hamster. 17L3-HS1 activity was measured by formation of testo from [le]~4-dione. Incubation wex performed at 37&2 for 150 min with 0.5 pM labeled substrate and 1 mM NADH NADPH. Data are present as means f SEM in pmoles testo for.med/m protein/min (log scaie).

Figure 4. Tissue distribution of Sa-reductase activity in a series of steroidogeni and peripheral tissue from male and female hamster. Sa-reductase activity wa measured by formation of DMfrom [14C] testo. Incubation were performed i 370C for 150 min with 0.5 pM labeled substrate and 1 mM NADPH. Data ar present as means f SEM in pmoles DHT formed/mg protein/rnin (log scale).

Figure 5. Tissue distribution of aromatase activity in a series of steroidogeni and peripheral tissue from male and female hamster. aromatase activity wa measured by formation of El from [l'Cl Addone. Incubation were perfonned a 370C for 150 min with 0.5 pM labeled substrate and 1 mM NADPH. Data ar present as means f SEM in pmoles DHT fonned/mg protein/& (log séale

Figure 6. Tissue distribution of estrogenic 17B-HSD activity in a series o steroidogenic and peripheral tissue from &de and female hamster. 17B-HS1 activity was measund by formation of E2 from P4C] El. Incubation wer performed at 370C for 150 min with 0.5 ph4 Iabeled substrate and 1mM NADH NADPH. Data are present as meaw f SEM in pmoles testo formed/m proteiri/min (log scale).

Figure 7. Tissue distribution of El suifatase activity in a series of steroidogen and peripheral tissue from male and female hamster. Sulfatase activity wi measured by the hydrolysis of El-S to El. incubation were performed at 37a for 150 min with 10 nM label4 substrate. Data are present as means f SEM : fmoles DHEA formed/mg protein/min (log scale). FIGURE 1

- -- Adrenal 1 Testis h D. Prostate b V. Prostate S. Veslcle Kidney I! Liver V. Skin O. Skin Ear ;6 Flank Olgan l' 3 DHEA-S DHEA l 1 1 .O 1 .1 1 10 DHEA FORMED (fmoUmg proteintmin)

Adrenal Ovary Ulerus Vagina Kidney Liver V. Skln O. Skin €sr Flank Organ DHEA-s OHEA

1 1 .cil - .l 1 10 DHEA FORMED (fmollmg proteinlmin) FIGURE 2

Adnnal 1 Testis f( D. Prostate

V. Prostate . . S. Vestcle P Kidney 1 uver 1 V. Skin 1 O. Skin b Ear

Flank Organ DHEA 4.010~~ 1 1 1 I 1 .O1 -1 1 10 100 1000 COIONE FORMED (pmoUmg protelnimin)

Adrenal Ovaty uterus Vagina KIdney Uver V. Skin D. Skin Ear Flank Orgin DHEA &DION€ 1 41 .t 1 10 100 1000 COIONE FORMED (pmoUmg protelnlmln) FIGURE 3

a Adrenal I Testis 4 D. prostate 3 V. prostate 1 S. Vesicle k Kiclney b Liver b V. Skin D. Skin I

Flank Oigm 4-DIONE TEST0 1 1 1 L I .O1 .1 1 1O 100 1000 ESTOSTERONE FORMED (pmollmg proteinlmin)

Adrenal Ovary Uterus VagiM Kidney Llver V. Skin O. Skin Ear

Flanik Organ 4-DIONE j TEST0

I I ni .1 i i'0 160 1 TESTOSTERONE FORMED (pmollmg proteinlmin) FIGURE 4

Adrenal Testis D. Prostate V. Prostate S. Vesicle Kidney Liver V. Skin D. Skin Ear

Flank Organ TEST0 DHT 1 .I i 100 DIHVDROTESTOTERONE FORMED (pmollmg proteintmin) 9 Adre~l Ovary Uterus Vagina Kidney Liver V. Skin D. Skin Ear Flank Organ TESTO DHT

DIHYDROTESTOSTERONE FORMED (pmoUmg proteidmin) FIGURE 5

a Adnnal b Tedis D. Prostate 1 V. prostate b S. Vesicle Kidmy I Liver P V. Skin 1 ' O. Skin F Ear 1 Flank Organ COIONE E, 1 I 1 .O1 .1 1 10 ESTRONE FORMED (ImoVmg proteidmin)

Admnal Ovaty Utenis Vagins Kidney Uver V. Skin 0. Skin EN

Flank Organ COIONE El 1 .or - .1 1 10 ESTRONE FORMED (fmollmg proteidmin) Adrtnal Testls D. Prostate V. Prostaie S. Vesicle KMney Lfver V. Skin O. Skin Ear Flank Organ .i 1 10 IOO ESTRADIOL FORME0 (pmol/mg proteidmin)

Adrenal Ovary Uterus Vagina Kidney Uver V. Skln O. Skln Ear ik Organ .I i iO ESTRADIOL FORMED (pmoUmg protein/min) FIGURE 7

Adnnal Teslis D. Prostate V. Pro-te S. Vesicle Kldnay Liver V. Skln O. Skin Ear Flank Organ -1 i IO ESTRONE FORMED (fmoVmg proteinlmin)

Adrenal O"W utenis Vaglnr Kidney Uw V. Win D. Skln

Flaink Organ E.-S

I 7 ~ -- .t 1 ESTRONE FORMED (fmoVmg protelnlmin) ADRENAL STEROID PRECURSORS EXERT POTENT ANDROGEMC Am01 IN THE HAMSTER SEBACEOUS GLANDS OF FLANK ORGANS AND EARS

Caiiin CHEN, Alain BELANGER and Femand LABRIE

MRC Group in Mofecular Endocrinology, CHUL Research Cenfer and Laval University, Québec, GIV 4G2, Canada

Endocrinology 137 : 1752-1757,1996 SUMMARY

To assess the effect of the androgen precursors dehydroepiandrosterone (Dm)and androstenedione (4+ndione) on androgensensitive parameters in the skin of the hamster, these two steroids were released from silastic implants inserted subcutaneously in castrated male hamsters. The pigmented area of the flank organs, the size of the underlying sebaceous glands, [JHI- thymidine incorporation into these sebaceous glands, and the size of the ear sebaceous glands were measured. The decrease in Bank organ size caused by orchiectomy was partiaiiy reversed by DHEA and completely reversed by 4- dione, testosterone 0,or dihydrotestosterone (DHT)implants. Similarly, the size of the sebaceous glands of the flank organs was reduced 88.1% by castration, while DHEA, 4-ene-dione, T and DHT implants reversed the effect of ordriectomy. Orchiectomy also deaeased the size of the sebaceous glands of the ears; DHEA, 4-enedione, Ttand DHT implants inaeased their size to 50.6%, 81.9%, 91.6%, and 105.8% of the values found in intact hamsters, respectively. Parallel results were observed on the labeling of flank organ sebaceous glands with PH]-thymidine as well as on prostate weight. Serum concentrations of T and DHT became undetectable in castrated hamsters and were increased to intact or slightly elevated values in anirnals receiving implants of DHEA, 4-ene-dione, or T. The present results show that DHEA and 4-me-dione are potent stimdators of androgen-sensitive parameters in the sebaceous glands of both the flank organs and ears in the hamster and illustrate the importance of extragonadal or peripheral intraaine formation of active steroids. It is suggested that the castrated hamster supplemented with adrenal precutsor steroids is a good mode1 that can closely mimic the human situation, where adrenal steroids play an important role in androgen formation and action in peripheral tissues. INTRODUCTION

Androgens play a predominant sole in the development, growth, and secretoty activity of the sebaceous glands. In fact, early studies have demonstrated that the active androgens T and Dmstimulate the sebaceous glands and that estrogenr exert =me inhibitory action (1,2). Excess androgenic action is wuknown to be associateci with frequent and cosmetically importani diseases, nameiy acne, sebomhea, hirsutism, and androgenic dopecia. The flad organ as well as the inner ear of the male hamster contain highly developed clusters of sebacmus glands that are widely used as models of human sebaceous gland activity (3,4). These mdels are in fact the mort widely used to assess the potentiai antiandrogenic activity of topically applied compounds (5-11).

To understand more fdy the role of androgens in the sebaceous glands, it is important to note that the skin possesses au the enzymes required for the transformation of the steroid precusors of adrem1 origin into active androgeril and estrogens and that these enzymes are expressed at a particularly high level in the sebaceous glands (12-14). In fact, humans are unique, along with some other primates, in having adrenals that secrete large amounts of the inactive precursor steroids dehydroepiandrorterone (DHEA)and especialIy DHEA-sulfate (DHEA-S),which are transformed into active ancirogens and/or estrogens in peripheral target tissues (15, 16). In fact, it is estimated that about 40% ol androgens in men while, in women, 75% of estrogens before menopause, and almost 100% after menopause are made in peripheral target tissues from the adrenal precursots DHEA and DHEA-S (15) .

The skin is the largest organ in the human body; therefore, the presence of all these steroidogenic enzymes (12-14, 17-19) makes it likely that this tissue synthesizes a sigruficant proportion of the sex steroids in humans, the main role of these steroids is a local or intracrine action (15). It thus becomes important ta gain a better understanding of the specific role of the individual adrenal steroids in the control of periphral androgenic action, especially in sebaceous glands which can be used as a &el. We have thus conducteci a study in the hamster using silastic implants to deliver constant levets of the adrenal steroids DHEA and Ilendone in order to mimic the contribution of the adrenals in the human. The parameters of androgenic action investigated were the flank organ size, the sebaceous gland size in both the fiank organs and ears, as weii as PH] thymidine incorporation in the sebaceous glands of the flank organs. MATERIALS AND METHODS

AnimaIs

Male addt Syrian hamsters (110-120 g) were purchased from Charles River (StXonstant, Québec) and housed 4 per cage in a Light (14 h light/day, lights on at 06:OO)- and temperature (22 f 1°C)-controlled environment. The animals had ad libitum access to hamster chow (Agway Country Foods Inc., Syracuse, NY) and water.

The animals were divided randomly into the following groups (8 animals per group): a) intact; b) orchiectomized; c) orchiectomized implanted with DHEA (1Slanllong implants); d) orchiectomized irnplanted with 4-ene-dione (6- cm-long implants); e) orchiectomized implanted with T (6-cm-long implants); f) orchiectomized hamsters implanted with DHT (4.5-cm-long implants). Castration was performed via the scrotal route with the animals under ether anesthesia. The implants were inserted subcutaneously into the dorsal area of the hamsters at the time of orchiectomy and left for 4 weeks. Control animals received silastic implants containing cholesterol. The flank organs were then shaved with an animal clipper, and the length and width of the darkly pigmented oval spot were carefully measured with a vernier caliper (Fisher Suentific). The surface area of flank orgaw was calculated using the foUowing formula: area = n (L/2 x W/2) = 3.14 (L X W)/4. Animals were killed by decapitation. Truncal blood was collected for measurement of plasma steroid hormones. The ventral and dorsal prostates were removed, dissected free from fat and connective tissue, and weighed. The flank organs and ears of the hamsters were removed quickly and fixed for morphological and histological studies.

DHEA, bene-dione, T, and DHî were purchased from Steraloids Inc. (Wilton, NH). The silastic implants (Dow Coming) were of the following sizes: 0.078" inner diameter, 0.125" outer diameter for DHEA and hne-dione, anr 0.062 inner diameter, 0.125" outer diameter for T and DHT.

The flank otgans and ears hmhamsters (5 animals per group) were fixe( in 4% paraformaldehyde for 48 h before immersion in 15% sucrose phosphat buffer for 24 h. The fhk organs were then cut in the middle in the rostral t( caudai direction, and a small section between the first and second cartilage ridge of each ear was cut and ernbedded in Tissuetek (Miles Inc., Diagnostic Divisior Eikhart). Frozen tissue was eut within a ayostat into 8-pm-thick 8ctions anc stained with hematoxylin and min. The surface area of the sebaceous glands a the flank organs was drawn on the photographs with a graphie tablet (Hewlett Padcard 9111A graphics tablet). The surface area of the ear sebacwus glands wa measured under optical miaoscopy using a grid and expressed in arbitrary unik The results are expressed as percentage of the intact control group.

Three animais in each group received 10 pCi [methyl3H]-thymidine (8: Ci/mmol, Amersham) diluted in 100 pi notmal saline injected subcutaneousl: into the flank organs; the animais were sacrificed 2.5 h later. The flank organ were cut in the midàle and fixed in Bouin's fluid (75% of picric acid, 25% of 370, formaldehyde, and 1% glacial acetic acid) for 2 days. The tissues were thel dehydrated in a séries of ethanol solutions and embedded in paraffin. Seven micrometer-thick sections were cut, deparaffined, and exposed tc autoradiography for 3 weeks. six consecutive sections of each sample weri analyzed. The total number of labeled celis on each slide was counted and thi mean of six slides from each flarik organ was calculated. The results ari expressed as percentage of the intact control group.

Serum DHEA, hne-dione, T, androsterone, and DHT concentratiom were measured in duplicate by radioimmunoassay after diethyl ether extractio~ followed by dvomatography on LH-20 columns using antisera developed an characterized by our laboratory (20).

Al1 results are expressed as means f SEM. Statistical significance wa deterrnined according to the multiple-range test of Duncan-Kramer (21). RESULTS

As illustrated in Fig. 1, implants containing the adrenal androge~ precursors DHEA and 4-enedione induced a marked stimulation of the surfaci area of the flank organ pigmenteci areas in orchiectomized hamsters. in fact, thc 75.8% decrease wface area of the flank organs resuiting from orchiectomy wa reversed by DHEA implants to 69.2% of intact values wMe 104.4% of intac values was observed following 4-ene-dione implantation. Moreover, the T anc DHT implants mpletely prevented the effect of orchiectomy.

Fig. 2 shows that simüar effects were observed on the surface area of th, sebaceous glands that are the highly predominant component of the hamste flank organs. Four weeks after orchiectomy, the size of the sebaceous glands ii the flank organs was reduced by 88.1% (P < 0.01 vs intact), whereas the DHW and 4lene-dione implants caused increases in the sebaceous gland size to 73.00, and 83.6% of intact values, respectively (P < 0.01 for al1 groups versus contra orchiectomized); Moreover, T or DHT implants completely reversed the effect o castration on this parameter.

Since the size of the sebaceous glands in the hamster ear is a well recognized parameter of androgenic response, we aIso measured the effect of th same endocrine manipulations on this parameter. As shown in Fig. 3, th( results obtained on the surface area of the ear sebaceous glands are almos superimposable to those observed on the surface area of the sebaceous glands O the flank organs. More preciseIy, orchiectomy caused a 95.1% (P < 0.01 vs intact decrease in the size of the ear sebaceous glands, whiie the implants of DHEA, 4 ene-dione, T, and DHT inaeased their size to 50.6%, 81.9%, 91.6%, and 105.8% o the intact values in castrated hamsters, respectively (P < 0.01 for al1 group versus orchiectomized control).

We next investigated the effect of the same endocrine manipulations 01 [3H]-thymidine incorporation into the sebaceous glands of the flank organs. A illustrateci in Fig. 4, the total numbet of labeled œiis in représentative sections a the hamster flank organ decreased by 81.9% in the orchiectomized groq whereas the Dm,benedone, T, and DHT implants compIeteIy preventei the effect of orchiectomy. Fig. 5 shows [3H]-thymidine incorporation il sebaceous glands of flank organ in the intact male hamster. Under ligl microscopy, most labeled ceIls are in the basai layer of the sebaceous glands an represent about 15 to 25% of total basal cells. Labehg was also seen in the bas; layer of the epidermis and in hair follides (data not shown). The silver grair were seen exclusively in the nuclei of the corresponding labeled ce&.

TO compare the effect of the steroid implants on the sebaceous gland with the effect observed on another typical androgen-sensitive tissue, w measured the effect of the same treatment on ventral and dorsal prostati weights in the same animals. As illustrated in Fig. 6A, ventral prostate weigl was reduced by 37.5% (P < 0.01) from 47.5 f 2.32 mg in intact animals to 29.7 1.62 mg 4 week after castration. Constant administration of active androgens and DHT as well as the adrenal precursors DHEA and 4-ene-dione cause significant increases in ventral prostate weight. In fact, 109.1%, 127.1%, 144.65 and 139.6% of intact contro1 values were measured following DHEA, 4-ent dione, T, and DHT implants. The corresponding chenges in dorsal prostal weight can be seen in Fig. 6B: 4-week castration deaeased dorsal prostate weigk by 86.1% while treatment of castrated hamsters with DHEA, 4sne-dione, T, an DHT increased dorsal prostate weight to 97.9%, 107.2%, 118.5%, and 113.6% c the values found in intact hamsters.

It is of interest to see in table 1 the senun levels of DHEA,4-ene-dionef 1 DHT, and androsterone achieved with the above-described treatments. Th serum DHEA concentration was elevated ftom undetectable levels (e0.2 nM) i castrated animals to 55.2I 6.6 nM in animals bearing DHEA implants. On th other hand, as expected, treatment with 4-ene-dione, T, or DHT had no effect O serum DHEA. Serurn 4lêne-dione, on the other hand, was increased fror undetectable levels in orchiectomized animals (4.5 nM) to 16.5 f 0.53 nM i: orchiectomized animals bearing 4+nedone implants. Implants of DHEA anc T, on the other hand, increased sérum 4ene-diane to 5.5 f 0.04 nM and 3.1 : 0.59 nM, respectively (P < 0.01), thus reflecting 3B-HSD and/or 17mDactivitj respectively; DHT had no effect on the serum levels of these studies. The 6e.r~~ levels of the androgen metabolite androsterone were reduced from 2.3 f 034 nh to undetectable leve. after orchiectomy but were increased to 6.6 f 0.69 nM, 15. f 1.98 nM, 5.6 f 0.22 nM, and 3.9 f 0.44 nM in castrated animals implanted witl DHEA, kndione, T, and Dm,respectively. As iliustrated in same table, T implants caused a marked increase ir semT from undetectable levels in casttated hamsters to 48.6 f 10.9 nM; DHEl anci 4edoneimplants, on the other hand, led to serum T levels of 7.1 f 2.72 nM and 19.0 f 0.70 nM, respectively. Semm DHT concentration was elevated ti values not sipificantly different from intact controls in castrated animal! bearing implants of DHEA, Q-enedione,or T; higher value of 6.9 f 1.42 nM wai observed with the DHT impl&ts. DISCUSSION

The present data dearly demonstrate that the inactive steroid precursol of adrenal origin DHEA and knedione are potent stimulators of a series c androgen-sensitive parameters in the hamster, namely, in the sebaceous gland of the flank organs and ears as well as in the ventral and dorsal lobes of th prostate. Such data indicate that the hamster can be used as a modei applicable t humans to study the control of androgen formation and action in periphers tissues, especially in the sebaceous glands and prostate, two tissues that are th sites of important diseases in the human. In fact, cancer of the prostate, the moi frequent cancer in men and the second cause of cancer death in men in Nort America (22), is well known to be highly sensitive to androgens while benig prostatic hyperplasia affects most men after the age of 50 years. On the othe hand, acne, seborrhea, hirsutism, and androgenic alopecia are well known to b androgen-sensitive and are cosmetically important and frequent medica problems (23,24).

Because DHEA and Q-ene-dione are inactive steroids, the present dat dearly demonstrate the importance of extragonadal or intracrine formation c androgens (15). In fact, the word intraainology was first coined in 1988 tl describe the formation in peripheral target tissues of active androgens fror inactive steroid precursors (25). These locally synthesized steroids exert thei action in the same cells where synthesis takes place without release into th general circulation. The hamster adrenals, like those of al1 laboratory animal (except primates), do not seaete sigruficant amounts of DHEA. Our data indicat that exogenous supplementation of DHEA in the castrated hamster should b provided in order to be in a position to extrapolate the results to the human.

Before king active, the adrenal precursors DHEA and 4+ne-dione mus be taken up by the target tissues and be transformed intracellularly by thl enzymes 3p-hydroxysteroid dehydrogenase/~s-A*-isomerase (3p-HSD), 178 hydroxysteroid dehydrogenase (17WD), Sa-reductase, and/or aromatase inb active androgens and/or estrogens (15). These enzymes are present in humai (12-14,18019026,27). An interesting observation made in the present study is the lack of a dos correlation between senun 4-ene-diane, T, Dm, and androsterone levels an( their action on androgen-sensitive parameters in peripheral target tissues. Sud data indicate the major importance of intracellular steroid levels and th relative lack of reliability of .measurement of serurn steroid concentrations, th changes in intracellular T and DHT not being reflected in the serua concentration of the same steroids (16). Although 4-enedione cause stimulatory effects on al1 the androgen-sensitive parameters studied that ar comparable with those of the T implants, the serum levels of T were mucl lower in animals bearing 4-ene-dione implants than in those bearing ' implants. Such data suggest that a large proportion of active androgen synthesized in peripheral target tissues are metabolized locally before beinl released into the circulation. This is also supported by the data of table 1, whid show that the serum levels of the metabolite androsterone are lower in the 'I and DM-treated animals compared with those treated with the DHEA and 4 ene-dione implants, respectively. Such data suggest that the oxidative pathwa: of 17$-hydroxysteroid dehydrogenase is favored, as observed in other periphera target tissues (27-29).

It has been suggested that the pigmented spot and the sebaceous gland may respond differentially to hormonal stimulation, thus indicating that visua inspection alone of the hamster flank organs may lead to erroneous condusion about the size and activity of the underlying sebaceous glands (4). Under thc present experimental conditions, DHEA, 4-ene-dione, T, and DHT implant, reversed the reduction in the size of the sebaceous glands of hamster flanl organs and ears after orchiectomy, the effects being dosely parallel to thosi rneasured on the size of the flank organs. Comparable effects were alsc observed on [SHI-thymidine incorporation, a highly sensitive parameter O androgen action in sebaceous glands.

Prostatic weight has long been recognized as a reliable parameter O androgenic and antiandrogenic activity (30-32). In the present study orchiectomy led to 37% and .86% decreases in the weight of the ventral ant dorsal prostatic lobes. respectively, whereas DHEA, 4-ene-dione, 1, and DM implants completely reversed the effect of orchiectomy and sometime! stimdated these glands at levels above those found in intact animals. It i! interesting that DHEA reversed to 69.2%, 73.0%, and 50.6% of intact values the area of the pigmental spots, the size of the sebaceaus glands in flank organs, and the size of the sebaceous glands in the ears of castrated hamsters, respectively, while it reversed prostatic weight to 109.1% of intact values. These results might suggest that the steroidogenic enzymes in the two lobes of the prostate are more active than those in the sebaceous glands.

in adult men, the senun levels of the main metabolites of androgens, namely, Sa-androstan-3a-17Wio1, androsterone and their glucuronidated derivatives, are only reduced by 50 to 70% following castration, thus indicating that adrenal steroids contribute 30 to 50% to the pool of total androgens in adult men (15, 16, 33-35). As mentioned above, humans are unique among species, except other primates, in having adrenals that secrete large amounts of precursor steroids (15). It is however interesting to note that the deuease of cirdating adrenal and testicular Cl9 steroids during aging was not associateci to similar changes in DHT and its polar metabolites, thus suggesting an inaease of DHT formation by steroid target tissues during aging.

In sumrnary, the present data show that the precursor steroids of adrenal origin, which are now well recognized to play a major role in the overd1 secretion of androgens in men (15-16), are also efficiently transformed into the active androgens T, and DHT in peripheral androgen-target tissues in the hamster. Since the size of flank organs and the sebaceous glands in the flank organs and ears in the hamster have been recognized as good models of androgenic action (5-Il), the present data suggest that the present modification of the previously well-recognized model (5-Il), namely, supplementation in intact animals with implants constantly delivering the adrenal precursor DHEA and/or Q-ene-dione, offers a model that more closely mimics the human situation and takes into account the contribution of the adrenals. We thank France Lepire for her skiilhl technical assistance. This woi was supported by the Medical Research Council (MRC)of Canada, Societ d'investissement R&D Andros Inc. and Endorecherche. We also thank D Jacques Simard for his interest in this work. REFERENCES

Ebling FJ 1948 Sebaceous glands. 1. The effect of sex hormones on th sebaceous glands of the female albino rat J Endocrin01 5:297-302

Ebling FJ 1963 Hormonal control of the sebaceous gland in experimenti animals. Advances in Biology of Skin. The Sebaceous GIands. Pergamo Press, Oxford, vol 4:2001219

Hamilton JB, Montagna w 1950 The sebaceous glands of the hamstei Morphologicai effects of androgens on integumentaty structure. Am J Ani 86:191-233

Plewig G, Luderschmidt C 1982 Hamster ear model for sebaceous glands. Invest Dermatol 68:171-176

Goss GMAA, Wirtz P, Vorrnorken HJM 1982 An improved method fc evaluating antiandrogens. Dermatol Res 273333-341

Weissman A, Bowden J, Frank BL, Horwitz SN, Frost P 1984 Morphometri studies of the hamster flank organ: an important model to evaluat pharmacologic effects on sebaceous glands. J Invest Dermatol82522-525

Lutsky BN, Budak MtKoziol P, Monahan M, Neri RO 1975 The effeds of nonsteroid antiandrogen, Nutamide, on sebaceous gland. J Invest Dermatc 64:412-417

Schroder HG, Ziegler M. Nickisch K, Kaufmann ~,.~treb~FE 1989 Effect c topically .applied antiandrogenic compoundr on sebaceous &nds c hamster ears and flank organs. J Invest mat0192:769-773

Luderschmidt C, Eiermann W, Jawny J, Bidlingmaier F, Ring J 1984 17a propyimesterolone (SH-434):an antiandrogenic sebosuppressive substanc not influencing circulating testosterone concentrations. Experimenta studies in Syrian hamsters. Nauynyn Schmiedebergs Ar& Pharmacc 328:214218 Brooks JR, Primka R, Bennan C, Krupa DA, Reynolds GF, Rasmusson GI 1991 Topical anti-androgeniaty of a new 4-azasteroid in the hamstei Steroids S6:42&433

Bouton MM, Lecaque D, Secchi J, Tournemine C 1986 Effect of a net topically active antiandrogen (RU38882) on the rat sebaceous gland cornparison with cyproterone ace ta te. J Inves t Dermatol86:163-167

Baillie AH, Thomson J, Milne JA 1966 The distribution of hydroxysteroic dehydrogenase in human sebaceous glands. Br J Dermatol78:451-457

Pochi PE, Strauss JS 1969 Sebaceous gland response in man to th' administration of testostetone, A 4-andros te nedione ani dehydroisoandrosterone. J Invest Dermatol 52:32-36

Dumont M, Luu-The V, Dupont E, Pelletier G, Labrie F 199: Characterization, expression and immunohistochernical localization of 38 hydroxysteroid dehydrogenase/~5-A4isomerase in human skin. J Inves Derrnatol99:415-421

Labrie F 1991 ïntracrinology. Mol Cell Endocrinol78:C113-C118

Labrie F, Dupont A, Bélanger A 1985 Complete androgen blockade for th treatment of prostate cancer. In: de Vita, VT, Hellman, S and Rosenberg, S1 (eds.) Important Advances in Oncology. J.B. Lippincott, Philadelphia:193-211

Cameron EH, Baillie AH, Grant JK, Milne JA, Thomson J, Transformatioi in vitro of [7a-3H] dehydroepiandrosterone to PH] testosterone by skin fron men. Proceedings of the lOlst Meeting of the Society for Endocrinology London, 1966, J. Endocrinol. 35: p. 19 (abst.)

Ellis RA 1958 Ageing of the human male scalp. In: Montagna, W and Ellis RA (eds.) The Biology of Hair Growth. Academic Press, New York:469485

Hay JB, Hodgins MB 1978 Distribution of androgen metablizing enzyme in isolated tissues of human forehead and axillary skin. J Endocrinol 7929 39 Bélanger A# Caron S, Picard V 1980 Simultaneous radioimmunoassay c progestins, androgens, and estrogens in adult rat testis. J Steroid Biochex 13185190

Kramer CY 1956 Extension.of multiple range tests to' group means wit unique numbers of replications. Biometrics 12:307-310

Boring CC, Squin TS#Tong T 1993 Cancer statistics 1993. CA Cancer J CL 43:7-26

Burton JL, Johnson G# Libman S, Shuster S 1972 Skin virilism in wome: with hirsutisrn. J Endhl53349-354

Plewig G 1974 Acne vulgaris, proliferative cells in sebaceous glands. Br Dermatol90:623430

Labrie Cf Bélanger A, Labrie F 1988 Androgenic activity o dehydroepiandrosterone and androstenedione in the rat ventral prostatc End~~~inology123:1412-1417

Sawaya ME, Penneys NS 1991 hmunohistochemical distribution a aromatase and 3bhydroxysteroid dehydrogenase in human hair foiiicle and sebaceous gland. J Cutan Pathol19:309-314

Martel C#Rhdaume E, Takahashi M#Trudel C, Couet J# Luu-The V, Simarc J, Labrie F 1992 Distribution of 17P-hydroxysteroid dehydrogenase gen expression and activity and human tissues. J Steroid Biochem Molec Bio 41:597403

Thériault Cf Labrie F 1991 Multiple steroid metabolic pathways in ZR-75-' human breast cancer cells. J Steroid Biochem Molec Bi01 38:155-164

Couture P, TMriault C, Sirnard J, Labrie F 1993 Androgen recepbr-mdatec stimulation of 176-hydroxysteroid dehydrogenase activity b! dihydrotes tosterone and medroxy progesterone ace ta te in ZR-75-1 humai breast cancer celis. EndOCtiTY)logy 132179485 30. Poyet P, Labrie F 1985 Cornparison of the antiandrogenic/androgeni activities of fiutamide,.cyproterone acetate and rnegestrol aœtate. Mol Ce1 Endocrin01 42283-288

31. Labrie C, Simard J, Zhao HF, Pelletier G, Labde F 1990 Synthetic progestin stimulate prostatic binding protein messenger RNAs in the rat ventra pmtate. Mol CeU Enduaiml 68:169-179

32. Labrie C, Cusan L, Plante M, Lapointe S, Labrie F 1989 Analysis of th androgenic activity of synthetic "progestins" currently used for thi matment of prostate cancer. f Steroid ~iodrem28:379û84

33. Moghissi E, AbIan F, Horion R 1984 Origin of plasma androstanedio glucuronide in men. J Clin Endocrinol Metab 59:417-421

34. Bélanger A, Bmchu M, aiche J 1986 Levels of plasma steroid glucuronidei in intact and castrated men with prostatic cancer. J Clin Endocrinol Metaï 62812-815

35. BQIangerA, Candas B, Dupont A, Cusan L, Diamond P, Gomez JL,Labrie 1 1994 Changes in serum concentration of conjugated and unconjugatec steroids in 40- to 80- year-old men. J ChEndocrin01 Metab 79: 1086-1090 LEGENDS TO FIGURES

Figure 1. Effect on the size of flank organs of orchiectomy and treatment c castrated hamsters for 4 weeks with implants-of DHEA, 4-ene-dione, T, or DM Data are presented in mm2 as means f SEM. .,P < 0.01 vs castrated control.

Figure 2. Effect on the size of the sebaceous glands of flank organs a orchiectomy and treatment of castrated hamsters with implants of DHEA, 4+nc dione, T, or DHT for 4 weeks. Data are presented as means f SEM taking th intact control as 100%. **P c 0.01 vs castrated control.

Figure 3. Effect on the size of the ear sebaceous glands of orchiectomy anc treatment of castrated hamsters with implants of DHEA, 4-ene-dione, T, anc DHT for 4 weeks. Data are presented as means f SEM taking the intact control a 100%. **P< 0.01 vs castrated control.

Figure 4. Effect on [JHJ-thymidine incorporation in hamster flank organs a intact or orchiectomy or treatment of castrated hamster with implants of DHEA 4-ene-dione, T, or DHT. Sections were autoradiographed for 3 weeks. Si: consecutive sections were counted for each sample. Data are presented as mean f SEM by the total number of labeled cells in intact animals as 200%. "P < 0.0 vs castrated control.

Figure 5. Light rniaoscopic autoradiography of PH]-thymidine incorporation ir the different sebaceous glands of untreated mature male hamster. Seven-pm thick paraffin section exposed for 3 weeks. The labeled cells are located mainly ir the basal celis of acini of the sebaceous glands of the flank organ. + Labelled ceil H/E. x 500.

Figure 6. Effect on the weight of ventral (A) and dorsal (B) lobes of the prostatc of orchiectomy or treatment with implants of DHEA, 4ene-dione, T, and Dm Data are presented as means f SEM. -P < 0.01 vs castrated control. TABLE 1

Serurn stemid levels in intact, castrated, and steroid-supplemented hamsters

Treatment Steroids

DHEA androsteroe

Intact . a

DHEA

4-ene-di one

DHT

Serum DHEA, 4ae-dione, androsterone, Tland DHï levels in intact, castrate (CX), and castrated animals treated with DHEA-, 4-ene-dione-, T-, or DHï releasing implants for 4 weeks. Serum steroid levels achieved with thes treatments were measured by RIA after extraction and chromatography on LE 20 columns. Data are expressed as the means f SEM of individual samplc (n=B/group). The Limits of detection (LD) for serum DEA, 4-ene-dionc androsterone, T, and DHT were 0.27,0.20,0.50,0.19, and 0.12 nM,respectively. FIGURE 1

3 . . - - -

INTACT I I IMPLANT I 1 ORCHIECTOMY 1 SURFACEAREAOFSEBACEOUSGLANDSOF FLANK ORGANS (Oh OF INTACT) A A FIGURE 3 [3H]-THYMIDINE LABELLE0 CELLS (Oh OF INTACT) A 4 FIGURE 5 --- P

IN. IMPLANT 1 - DHEA ADMINISTERED PERCUTANEOUSLY EXERTS SYSTEMIC EFFECTS IN THE HAMSTER

îaiiin CHEN and Fernand LABRIE

Laboratoy of Molecuiar End~nology,CHUL Research Cmtw, Québec, Canar Adxenal steroid precursors, especially DHEA, play an important role in androgen and estrogen formation in man and other primates. The decreasing levels of these steroids could be associated with problems of aging in both men and women. The purpose of this study was to investigate the effect of percutaneou administration of DHEA and androstenedione (~4-dione)on the parameters of androgen and estrogen actions in orchiectomized (ORCH) or ovariectomized (OVX) hamsters. The steroids were administered at a dose of 5 mg of DEEA or 2 mg of dr-dione on the right flank organ, right ear, and dorsal skin, twice daily, for four weeks. 80% and 82% deaeases in the size of the left and right flank organs, respectively, were caused by castration while 48.7% and 40.2% reversals were achieved by DHEA and a complete reversal was obtained with ~4-dione.Parallel resdts were observed on the underlying sebaceous giands and [3H]-thymidine incorporation. 48.3%, 86.8%, and 70.4% decreases, respectively, were found in ventral and dorsal prostate, and seminal veside weights following castration while 65.1%, 96.8% and 103.2% reversals were obtained with DHEA and a 100% reversal was obtained with b4-dione. In the OVX hamsters, DHEA caused no sigdicant change in the size of the flank organs, while dldione caused an inûease to the value observed in intact male animais. On the other hand, OVX caused 79.7% and 56.9% decreases in ute~e and vaginal weights, respectively, while reversals of 20.8% and 34.1°/0 on the uterine, and 56.9%, and 77.5% on vaginal weights were observed following the treatment with DHEA, and ~4-dione, respectively. ORCH increased steroidogenic enzymatic activities, namely 3P-HSD, I7P-HSD, and Saoreductase in male hamster dorsal skin and prostate while the administration of DHEA and 44-dione had an inhibitory effect except for 38-HSD which was increased following DHEA treatment in the prostate. Moreover, aromatase activity was decreased in ORCH hamster dorsal skin and a further decrease was observed following DHEA treatment while ~4-dianehad no sigruficant change. Similar results were obtained in the skin and utenis in female hamsters except the aromatase activity was increased following OVX and DHEA reversed the effect in the skin. The present data suggest that DHEA and A4dione exert their main androgenic and/or estrogenic activitis after systemic absorption and not by a local action at the site of application . INTRODUCTION

The growth and differentiation of steroid target tissues are often directly regulated by potent steroids that circulate within the plasma cornpartment After their diffusion into target cas, these steroids may then bind their cognate receptors and, thus, trigger an uray of ktrinsic nudear processes [l]. in addition, a Iarge proportion of the hormonal steroids enter into the target cells in the forms of nonactive precursors, particularly dehydroepiandrosterone (DHEA] and its sulfate ester (DHEA-S)f2-81. DHEA, an abundantly secreted adrenal steroid, is converted into androstenedione (Addone) and then into poteni androgens and estrogens in peripheral tissues by the foiiowing enzymes: 3$- hydroxysteroid dehydrogenase/~s-~4isomerase (3$-HSD), 17p-hydroxysteroid dehydrogenase (17bHSD), Sa-reductase, and aromatase [BI.

Man is unique, with some other primates, in having adrenals that secrete large amounts of the steroid precursors. In fact, during the adrenarche ,the adrenal glands start to produce more androgens, predominantly DHEA and DHEA-S,which begin to rise rather sharply after age 7 in both sexes, continue ta increase more steeply from ages 10-11 until adolescence, and elevated values 01 circulating DHEA-S maintain through adult life [9]. in both sexes, the plasma levels of DHEA and DHEA-Sdeaease significantly with aging. -Infact, at 70 years of age, serum DHEA-S level is at approximately 20% of its peak value while il decreases by up to 95% at the age of 85-90 years [10,11]. Thus, the 70% ta 95% reduction in the formation of DHEA-S by the adrenals during aging results in a dramatic reduction in the formation of androgens and estrogens in peripheral tissues, and thus may influence the biochemical and physiological functiow of certain celis which depend upon sex steroiids.

Although the physiological role of DHEA has not been defined, however, a growing body of evidence, both epidemiological and experimental, suggests a strong inverse relationship - between alteration in the serum levels or exaetion of this steroid or its metabolites and a number of disease syndromes, such as osteoprosis [12,13], atherosclds, and cardiovascular base [14], breast cance~ [15,16], and a desof other conditions including obesity, autoinunune diseases, fatigue8muscle -8 diabetes, aglng, and 10ngevity [IO, 17-22] In order to achieve a better understanding of the specific role of individu, adrenal steroid in the control of pedpherd anbgenic and estrogenic activitie we designed the present study to investigate the effects of DHEA as weiï as & dione treatment percutaneously on the sexual accessory tissue weight, flan organ size, and the changes of steroidogenic enzymatic activities in male an female hamsters. Considering the relevance between plasma DHEA or DHEA decreases and aging, the orchiectomized (ORCH) and ovariectomized (OV) animals have been used. MATERiALS AND METHODS

Animals Male and female Syrian hamsters obtained from' Charles River hc. (St- Constant, Quebec), weighing 110-120 g were housed in light- (14 h/day, lights on at O6:OOh) and temperature- (22 f loC) controlled environment. The animals received hamster chow (Agway Country Foods, inc, Syraause. NY) and tap water ad libitum.

Materials The steroids DHEA (5-androsten-3bl-17-ones) and ~4-dione(4-androsten- 3,170dione) were purchased from Steraloids (Wiiton, NH).

The hamsters were randomly divided into intact ~0nh01,gonadectomized control, and DHEA and ~4-dionetreated groups. At the beginning of the experirnent, the ear was marked for identification purpose, and the area of both side of flank organs and in front of dorsal skin about 2 cm in diameter were shaved with animal clippers. The gonadectomy was performed under ether anesthesia and the treatment was started on the same day. The appropriate doses of DHEA and ~4-dionewere dissolved in 50% ethanol-50% propylene glycol (vol/vol) and JO pi was administered on the right side of flank organs, dorsal skin as weil as right side of the ears. Meanwhüe, the intact group and the left side of the flank organs and ears of the DHEA and ~4-dione-treatedgroup received the vehide only. In the morning of the last dose, the length and width of the darkly pigrnented oval spot outlining each flank organ were carefully measured with a vernier caliper (Fisher Scientific). Flank organ area was then calculated according to the following formula: area of ellipse = n (L/2 x W/Z) = 3.14 (L x W)/4. The animals were killed by decapitation. The truncal blood was collected for measurement of piasma hormones. The ventral and dodlobes of prostate, seminal vesides, uterus, and vagina were nmoved, dissected, freed from fat and connective tissue, and rapidly weighed. Then, the ventral lobe of pratate, uterus u weli as a pieœ of dorsal skin (avoiding the treateà ma) were frozen and stored at -ûû"C for enzymatic assays. Frozen tissues were hornogenized with a polytron in phosphate buffer (20 mM KH2PO4, 0.25 M suaose, 1 mM EDTA, pH 7.5) containing protease inhibitors (1 mM phenylmethylsulfonyl fluoride and 5 pg/rnl each of pepstatin A, antipain, and leupeptin), and centrifuged for 30 min at 10ûû x g to rerhove cell debris. Protein content of tissue homogenates was measured by the method of Bradford using bovine semalbumin as standard (231.

Aliquots of the 1000 x g supernatant were incubated for the inclkated time intervais at 37% in a total volume of 0.5 ml phosphate buffer (12.5 mM KWO4, 1mM EDTA, pH 7.5) containing 0.5 ph4 [IdCl-labeled substrate and 1 mM of the appropriate cofactor(s). Three PHÇD activity was measured with [I-l4C]-DHEA (S.A., 51 mCi/mrnol) and NAD+ while 17P-HSD activity was measured with [4- 14C]-~4-dione and NADH. (4-l4C]-Testosterone (S.A., 51.4 mCi /mol) and NADPH were used for measurement of Sa-reductase activity. Labeled radioactive steroids were purchased from New England Nuclear/Dupont (Markham, Canada) and purifiecl by thin-Iayer chromatography (TLC)before use. The enzymatic reactions were stopped by chilling the incubation mixture in an ice-water slurry and adding 3 ml of diethyl ether. The components were then minced and frozen in a dry icestfiano1 bath. The organic phase was kept while the remaining frozen aqueous fraction was reextracted once with either. The organic phase was then pooled and evaporated to dryness under a nitrogen Stream. AU components were then separated on TLC (60 F254 silica gel, E. Merck, Darmstadt, F.R.G.) using the sdvent system containing to1uene:acetone 4:l (v/v) before autoradiography of the plates for 48 h The metabolites revealed by autoradiography were identified by comparing with standard steroid characterized by HPLC as dedbed previously [24]. The TLC areas correspondhg to DHEA and ~4-dionefor 3b-HSDf Ad-dione and testo for 17P-HSD. Testo, and DHT including its further metabalites androstane-3a,l7$-diol and andmstane- 38,178-di01 were scraped and transferred into vials containing 0.5 mi ethanoi and then added 10 ml scintiliation liquid. The radioactivity was then measured in a spectrophotometer. Results are expressed as means f SEM in pmoles or nmoles product fod/rng proteinhin. Aromatase activity was measured with a radiometric assay 1251. Briefly tissue homogenates were incubated for 180 min at 37OC with 10 nM [1$,2B. 3H](N)]~4done(S.A. 44.8 Ci/mmol) in phosphate butfer (Icrr2PO4,lmM EDTA pH 7.5) containing 1mM NADPH. The reaction was stopped by adding a kxturt of cold active charcoal (2.5% charcoal in phosphate buffer, pH 7.5) for 4h at 40C

The charcoai.. was then removed by centrifugation and an aliquot of the resdtan supernatant was measured in a scintillation spectrophotometer.

IxaHaA Senun LH and FSH were measured by double-antibody RLA using rat LH-E 6 and FSH-1-6 for iodination, rat LH-RP-2 and FSH-RP-2 as standards, and tht rabbit antisera anti-rLH-S-8 and anti-rFSH-S-8 which are generoudy supplied bj the National pituitary program, Baltimore, USA.

RIA data were analyzed using a program based on mode1 If of Rodbard anl Lewald [26]. Data are expressed as means f SEM. Statistical sigruficance wa! measured according to the multiple-range test of Duncan-Kramer [2a. RESULTS

As iliustrated in Fig. IA, the 80.7% to 81.8% Ioss of surface areas of both flank organs was 40.25% to 48.7% reversed (pc0.01) by the application of 5 mg of DHEA on the right flank organ, right ear, and dorsal skin area, the effect being similar on both sides. ïhe application of androsienedione (~4-&one)at the dose of 2 mg, on the other hand, caused a complete reversa1 of the effect of orchiectomy on the size of both flank organs. It is interesthg to see in Fig. lb that while Addione caused a marked stimulation to the vaiue observed in intact male animals (33.38 f 1.9 mm2 to 35.71 f 1.9 m.2)or 4-5 tirne larger than that of intact femaies (5.72 f 0.39 mm2 to 5.23 f 0.41 mm2) of the size of the fiank organs in ovariectomized hamsters, DHEA had no significant effect at the dose used. Paraliel results were observed on the underlying sebaceous glands and PH]- thymidine incorporation (data not shown).

As shown in Fig. 2,48.3%, 86.8%, and 70.4% of deaeases were measured on ventral (panel A), dorsal (panel B) prostate and seminal veside (panel C) weights, respectively foiiowing castration. The twice daily application of 5 mg DHEA on the right flank organ, right ear and dorsal skin for 4 weeks in 30 microliters of 50% ethanol - 50% propylene glycol led to 65.1°h, 96.8%, and 103.2% (~~0.01)reversals and a 100% (pc0.01) reversa1 was obtained with ~4- dione treatment.

In the female animals, OVX caused 79.7% and 56.9% deaeases in uterus (Fig. 3A] and vaginal (Fig. 38) weights, respectively, while reversals of 20.8% and 34.1% (p<0.01) were observed on the uterine, and 56.9% and 77.5% (pc0.01) on vaginal weights after pe~taneoustreatments of 5 mg of DHEA and 2 mg of ~4- dione, respectively.

It is well known, the adrenal precurçur DHEA and Ac-dione are converted into active androgens and estrogens by steroidogenic enzymes, narnely 3mD, 17P-HSD, Sa-reductase, and aromatase in peripheral tissues. In order to investigate the medianism involved in systernic effeçtç caused by percutamous treatments of DHEA and dcdione described above, we have then measured the enzymatic activities in skin, prostate, and uterine. As iilustrated in Fig. 4A, 38- HSD acüvity was measured at the value of 0.15 f 0.01 pmol b4-dione formed/mg protein/min in the dorsal skin of intact male hamsters, 025 f 0.06 pmol (pq0.O versus intact control) was reached after 4-week castration while 0.04 pmol an 0.1 pmol d-dione formed were - obtained following DHEA and ~4-dion treatments, respectively. Seventeen b-HSD activity of dorsal skin in mal hamsters is shown in Fig. 48, 1.69 f 0.19 pmol testo formedbg protein/mi was observed in intact control while 2.48 f 0.25 pmol, 0.76 f 0.09 pmol, 0.79 f 0.a pmol were measured with control, percutaneous DHEA and ~4-dion treatments in ORCH hamsters. Parallel result was obtained with skin Sa reductase activity (Fig. 4C). Castration increased the Sa-reductase activity to 2.3 I0.20 pmol hm1.1 f 0.04 pmol DMfkd of intact control. inverse of thi effect, the treatments with DHEA and Acdione decreased the enzymatic activit to the value of 1.45 f 0.01 pmol and 0.45 f 0.03 pmol Dm formed, respective11 As shown in Fig. 4D, aromatase activity was measured by 30.97 f 0.93 ho1 test formed/mg protein/rnin in intact male hamster dorsal skin. In contrast wit other enzymes, ORCH decreased the aromatase activity to 11.63 f 1.7 fmol in th skin, and DHEA treatment caused a Merdecrease to 5.17 f 0.64 holwhile th treatment with A4-dione had no significant change in the parameter.

Comparable effects were observed in female hamsters, OVX inueased th activities of 3B-HSD, 17$-HSD, and 5a-reductase in the dorsal sh, and th administrations of DHEA and ~4-dionereversed the effects Vig. 5 A.B.C). 1 different from male hamster, OVX inaeased the rkin aromatase activity. In fac 6.36 f 0.55 fmoles testo formed/mg protein/min was measured in intact contr< while 14.25 f 0.97 fmoles was measured in OVX hamsters. Furthermore, 4.70 0.55 fmoles was obtained in DHEA treated group and 13.97 f 1.5 fmole wa obtained in dion one treated animals (Fig. 5D).

As you can see in Fig. 6A, 3B-HçD activity was measured by 0.1 f 0.01 pmc Ad-dione formed in prostate of intact male hamsters, while castration slightl increased the activity to 0.18 f 0.02 pmol. The treatment with DHEA furthe inaeased the prostatic 3p-HSD activity to 1.18 f 0.18 pmol whereas the treatmer with -one had rw, significant change. In contrast, orchiectomy inaeased th 17S-HSD and Sa-teductase activities whiie the reversal effects were obtaine with both DHEA and A4-dione treatment (Fig. 6B, C). Similar result wa oôtained in the fmu* ut- 3PHSû activity, 0.07k 0.008 pmols wis mearum in control animals, castration had no significant effect on this parameter whil the inaeases of 0.16 f 0.01 pmol and 0.25 f 0.02 pmol were adrieved by DHEA and ~4-dionetreatments (Fig. 7A), respectively. Fig. 78 shows utem 17B-HSD activity, 0.32 f 0.02 pmol and 1.3 f 0.13 pmol testo formed were measuréd in intact and castrated hamsters whüe 0.76 f 0.04 pmol and 0.96 f 0.M pmol were obtained in DHEA and ~4-dionetreated groups, respectively. In addition, 5a- seductase was inaeased to 10-fold after 4-week castration which was partiaiiy reversed by DHEA and ~cdionetreatments (Fig. 7C).

Table 1 summarizes the changes of serum steroids and LH and FSH concentrations aftei 4 week peicutaneous administrations of DHEA and 3- dione. in fact, in male hamster, serum LH and FSH concentrations were measured by 1.44 f 0.41 and 5-88 f 0.27 ng/ml in intact, respectively, which were increased to 4.29 f 0.52 and 42.62 f 159 ng/d foliowing orchiectomy. 3.59 f 0.33 and 31.92 f 5.53 were measured after DHEA treatment while 0.49 f 0.19 @ml of LH and 3.7 f 0.73 ng/d of EH weie obtained with A*-dione treatment. On the other hand, in female hamsters, ovariectomy increased LH level from 0.42 f 0.084 in intact to 4.74 f 0.56 ng/mi, but it had no significant changes with DHEA and ~4-dionetreatments at the doses used, however, elevated FSH were inhibited by 13.6% (p<0.05) and 31.2% (pc0.01) foiiowing DHEA and A4-dione trea tments, respectively. DISCUSSION

In the present study, we have demonstrateci the systemic effect of DHEA by comparing with A'-dione after topical administration in gonadectomized hamsters. The topical application of DHEA on the right flank organ, right ear and dorsal skin exerts a similar stimulatory effect on both Bank organs and increases prostatic and seminal vesicle weights in orchiectomized hamster. In ovariectomized female hamster, the treatment with DHEA percutaneously has a stimulatory effect on utem and vaginal weights while has no significant effect on the size of the flank organs. In agreement with this obseivation, we have reported previously that constant plasma concentrations of the. adrenal steroids DHEA and ~kiionemaintained within the range of those found in adult men have potent stimulatory effects on androgen-dependent gene expression, namely prostatic binding protein component 1 (PBP-Cl) and sperminebinding protein (SBP) in the rat ventral prostate as well as prostatic weight [6,28]. It should be mentioned that, in this study, we used percutaneous administration instead of common oral administcation, where DHEA has only been effective in large doses, apparently because much of the compounds is degraded in the liver before it reaches the bloodstream.

In this study, we have used flank organs of Golden Syrian hamsters as an androgen-dependent model. The flank organs are paired-pigmented spots on the backs of the animals. The flank organs of male hamsters are much larger and heavier than those of females. After castration of male hamsters, the size of the flank organ decreases and subsequent androgen administration restores their normal size. Similarly, the flank organs of females can be stimulated by androgens [29-31). Compared with the common models for androgen actions, such as prostate and seminal vesicles, the flank organs have the added advantages of visual assessment of hormone actions, also offers a comparable and practice model to investigate the androgenic action in female. The flank organs consist of large sebaceous glands, hair foIlicles, and demai pigment, which are androgen dependent tissues. The present data show that DHEA stimulates the prostate, seminal veside, and both sides of fiad organs in male hamsters. On the other hand, the stimulatory effect is only on the uterus and vaginai not on the size of fiank otgans in the fernale hamsters, this rnay indicate that DHEA can be transformeci mainly into androgens in orchiectomized male hamster, estrogens in ovariedomized female hamster in the peripheral targe organs at the dose used. We may alsa propose that, in female animals, DHEi can be conveted in a srnail quantity of indrogens which can stimulate the flanl organ but not enough to counterbalance the estrogenic inhibitory effect a DHEA. In contrast with DHEA, Ad-dione, on the other hand, has botl androgenic and estrogenic effects der being absorbed and converted in poten androgens and estrogens in the peripheral intracrine tissues.

Despite the abundance of DHEA and DHEA-S in serum, in fact, in humm the serum concentration of DHEA-S is 300 to 500 times higher than that O DHEA and 20 times higher than of any other steroid hormone, however, thei physiological role has so far remained undear. The previous studies strongl; suggest that the conversion from DHEA or A4dione into potent androgens an( estrogens occurs within the peripheral targe t tissues. Infusion of PH]-labelet DHEA-S or A4-dione in patients with benign prostatic hyperplasia befom prostatectomy resulted in high intraprostatic levels of radioactive DHT as well a the formation of various intermediates in the biosynthesis and metabolism O DHï [3]. In addition, the incubation of human ben@ hypertrophic tissue ii vitro with either [SHJDHEA or PHIDHEA-s led to the formation of DHï [3 Furthermore, in 1960, Cameron and his colleagues have reported th( transformation in vitro of [17a3HlDHE~to P~ltestosteroneby skin from me1 [2]. Recently, Kaufman and his colleagues have reported that 45-diol, A4-dione testo, DHT, androsterone, 5a-androstane-3a,17~olwere isolated following thi incubation of human genital skin with DHEA. The conversions were higher ii men than in women. In women, conversions of DHEA to AS-di01 and A4-dionc were highest, foliowed by conversions of DHEA to DHT and -5a-~4-dione whereas in men the transformation of Ad-dione to Sa-A*-dione was highest followed by AS-di01 and androsterone [32]. Thus, the data mentioned abovc support our finding that DHEA exerts androgenic effect in male and estrogenic effect in female gonadectomized hamsters. It is well known that AS-di01 has u fact been shown to exert direct estrogenic effects in bth- normal and malignan estrogenaensitive tissues at concentrations found in the &rculation of norma adult women (33.361. From 80 to 100% of AS-di01 in the circulation has beex found to derive from circulating DHE-AS and DHEA in postmenopausal anc women (341. Skin is the largest peripheral organ in human not only utilizing but als synthesizing the sex steroids. The skin presents al1 the enzymes to convei adrenal precursors into active androgens and estrogens [37-391. in the presei study, we have further demonstrated that the activities of 3P-MD, 17$-HSD, 5c reductase, and aromatase are presented in both male and female dorsal ski prostate, and uterus except the aromatase activity is very low in prostate an utew tissues (data not shown) with present methoci described in Materials an Methods above. Similar pattern was observed in both sexes, except aromatas activity in the skin. Gonadectomy increases the enzymatic activitiqs whil DHEA or Addione treatment reverses this effed for 3P-HSD, 17$-HSD, and 5a reductase. The inaeasing of steroidogenesis enzymatic activities followin gonadectomy may be due to the low levels of steroid substrates in peripheri tissues. The aromatase is the key enzyme for estrogen' formation. It i interesting to mention that aromatase activity is decreased in ORCH hamstg skin while DHEA adiministration had further reduction. In constrast, th aromatase activity is increased following OVX while DHEA treatment revers4 this effect. Although the mechanism is unknown, it may be involved in th explanation mentioned above, that DHEA exerts androgenic effect in ORCI male hamster and estrogenic effect in OVX female hamsters. On the other hanc the different effects among dorsal skin, prostate, and uterus on the 3P-HS1 activity may be due to different types of 3P-HSD presented in different peripheri tissues.

In summary, the present data dearly show that DHEA as well as A*-dion stimulated the size of the flank organ and the growth of prostate and semini vesicle in male and the growth of uterus and vagina in female gonadectomize~ hamsters after topical application for 4 weeks, thus eliminating the risk of a. exaggerated androgenic activity of DHEA adrninistered percutaneously fa systernic indications. REFERENCES

Beato, M.: Induction of transcription by steroid hormones. Biochim Biophys. A&. 910 (1987) 95-102.

Cameron, E.H., Baillie, AH., Grant, J.K., Milne, J.A. and Thomson, J. Transformation in vitro of [7a-3HJ dehydroepiandrosterone to [3H: testosterone by skin from men. In Prdings of the lOlst ZMeeting of fhi Soc* fir Endoctr'nology. London. J. Endoainol. (1966) Vol. 35,19 (abst.).

Harper, M., Pike, A., Peeling, W. and Griffiths, K.: Steroids of adrenal origu metabolized by human prostatic tissue both in vivo and in vitro. J Endmrinol. 60 (1974) 117-125.

Abul-Hajj, Y.J.: Metabolism of dehydroepiandrosterone by hormone dependent and hormone-independent human breast carcinoma. Steroids 24 (1975) 4û8-500.

Perel, E., Wilkins, D. and Kilhger, D.: The conversion of androstenedion4 to estrone, estradiol, and testosterone in breast tissue. J. Sferoid Biochem. 1: (1980) 89-94.

Labrie, C., Bélanger, A. and Labrie, F.: Androgenic activity oi dehydroepiandrosterone and androstenedione in the rat ventral prostate Endocrinology 123 (1988) l412-14f 7.

Manger, B., Bélanger, A., Labrie, F., Dupont, A., Cusan, L. and Monfette, G. Cornparison of residual C-19 steroids in plasma and prostatic tissue oi human, rat and guinea pig after castration: unique importance oi extratestidar androgens in men. 1. Steroid Bidchem. 32 (1989) 695-698.

Labrie, F.: Intraainology. Mol. W. Endom'nol: 78 (1991) C113C118.

Forest; M.: Physiological changes in circulafing androgens. Base1 Karge~ (1989) 104-129.

Orentreich, N., Brind, J.L, Rizer, RL. and Vogelman, J.H.: Age changes and sex differences in serum dehydroepiandrosterone sulfate concentratiom throughout adulthood. J. Clin. Endocrinol. Mefab. 59 (1984) 551455. 11. Migeon, C.J., Keller, A.R., Lawrence, B. and Shepart II, T.H. Dehydroepiandrosterone and anchosterone levels in human placenta Effect of age and sex: day-to-day and diunial variations. J.Clin. Endocriml Metab. 17 (1957) 1051-1M2.

12. Nordin, B.E.C., Robertson, A., Seamark, RF., Bridges, A., Philcox, J.C.,Need AG., Horowitz, M., Morris, H.A. and Deam, S.: The relation betwee~ calcium absorption, serum dehydmepiandrosterone, and vertebral minera density in postmenopausal women. J. clih. Endocrinol. Metab. 60 (1985) 651, 657.

13. Deutsch, S., Benjamin, F., Seltzer, V.. Tafreshi, M., Kocheril, G. and Frank A.: The correlation of serum estrogens and androgens with bone density U the late menopause. Int. J. Gynecol. Obstet. 25 (1987) 217-222.

14. Barrett-Connor, E., Khaw, KT. and Yen, S.S.C.: A prospective study oi dehydroepiandrosterone sulfate, mortality and cardiovascular disease. N Engl. J. Med. 315 (1986) 1519-1524.

15. Rose, D.P., Stauber, P., Thiel, A., Crowley, J.J. and Milbrath, J.R.: Plasmz dehydroepiandrosterone sulfate, androstenedione and corticol, and urinaq free cortisol excretion in breast cancer. Eur. J. Cancer 13 (1977) 43-47.

16. Schwartz, A.G.: Inhibition of spontaneous breast tancer formation In fernale C3H (Avy/a) mice by long-term treatment with dehydroepiandrosterone. Cancer Res. 39 (1979) 1129-1132.

17. Parker, L.N. and et al: Control of adrenal androgen seaetion. Endocr. Rm. 4 (1980) 392410.

18. Kent, S.: DHEA: "Miracle8' drug? Get.iatrics 37 (1982) 157-161.

19. Coleman, D.L., Schwizer, RW. and Leiter, E.H.: Effect of genetic background on the therapeutic effects of dehydroepiandrosterone (DHEA) in diabetes- obesity mutants in aged normal mice. Diabetes 33 (1983) 26-33.

20. Regelson, W.. Loria. R and Kalimi, M.: Hormonal intervention: "Buffei hormonesi' or "State dependency". Ann. New Yak Acad. Sei. 521 (1988) 260 273. Schwartz, A.G., Whitcomb, J.M., Nyce, J.W., Lewbart, M.L. and Pashko, L.L. Dehydroepiandrosterone and structural analogs: a new class of cancei chemopreventive agents. Adv. Cancer Res. 51 (1988) 3911424.

Thoman, M. and Weigle, W.: The cellular and sub&ellular bases O immunosenescence. Adv. Immununol. 46 (1989) 221-260.

Bradford, M.: A rapid and sensitive method for the quantitation o. microgram quantities of protein utilizing the principle of protein-dyi binding. Anal. Biochem. 72 (1976) 248-254.

Thériault, C. and Labrie, F.: Multiple steroid metabolic pathways in ZR-75-1 human breast cancer cel1s.J Steroid Bwchem, Molec. Biol. 38 (1991) 155-164.

Walsh, P., Madden, J., Harrod, M., Goldstein, J., MacDonald, P. and Wilson J.: New Engl. 1. Med. 291 (1974) 944-949.

Rodbard, D. and Lewald, J.E.: Computer analysis of radioligand assay anc radioimmunoassay data. In Second Karolinska Symposium on Researcl Methods in Reproductive Endocrinology. Acta Endocrinol (Copenh) (1970' Vol. 64,79-103.

Kramer, C.Y.: Extension of multiple range tests to group means with uniqui numbers of replications Biometrics 12 (1956) 307-310.

Labrie, C., Sirnard, J., Zhao, H.F.,Bélanger, A., Pelletier, G. and Labrie, F. Stimulation of androgen-dependent gene expression by the adrena precursors dehydroepiandrosterone and androstenedione in the rat ventra prostate. Endocrinology 124 (1989) 274502754.

Hamilton, J.B. and Montagna, W.: The sebaceous glands of the hamster Morphological effects of androgens on integumentary structure. Am. ] Anat. 86 (1950) 191-233. -

Takayasu, S. and Adachi, K.: Hormonal control of metabolism in hamstei costovertebral glands. J Invest Dmnatol55 (1970) 13-19.

Ebling, F.J.: Hormonal control and methods of measuring sebaceous gland activity. J Invest Dennafol62 (1974) 161-171. 32. Kaufman, F.: Dehydropiandrosterone and Dehydropiandrosterone sulfatc metabolism in human genital skin. Ferfil. StmW 54 (1990) 251-254.

33. Fouiin, R and Labrie, F.: Stimulation of celi proliferation and estrogenic reponse by adrenal~19-~~teroidsin the ZR-75-1 human breast cancer ce1 line. Cancer Res. 46 (1986) 4933-4937.

34. Adams, J.: Control of secretion and the function of C19-delta-Ssteroids oi the human adrenal gland. Mol. ce11 Endocrinul. 45 (1985) 1-17.

35. Simard, J. and Labrie, F.: Adrenal C19-5-ene steroids induce full estrogenic responses in rat pituitary gonadotrophs. J. Steroid Bwchem. 26 (1987) 539. 546.

36. Simard, J., Vincent, A., Duchesne, R. and Labrie, F.: F& oestrogenic activit) of ~~~-~~-adrenalsteroids in rat pituitary lactotrophs and somatotrophs Mol. Cell. Endocritfol. 55 (1988) 233-242.

37. Voigt, W., Fernandez, E. and Hsia, S.: Transformation of testosterone intc 1713-hy&oxy-5-androstan-310ne by microsomd prapations of human skin. 1, Biol. Chem. 245 (1970) 5594-5599.

38. Hay, J.B. and Hodgins, M.B.: Distribution of androgen metabolizing enzymes in isolated tissues of human forehead and axillary skin. J. Endocrinol. 79 (1978) 29-39.

39. Baillie, A., Calman, K. and Milne, J.: HistochemicaI distribution oi hydroxysteroid dehydrogenase in human skin. Br. Dmntol. 77 (1970) 610- 616. Figure 1. Effect of percutaneous aduninistration of DHEA or d'&one on the fladc organ areas in ORCH (panel A) or OVX @anel 8) hamsters. The animais received topical application of 5 mg DHEA or 2 mg ~4-dionetwice daily for 4 weeks in the right side flank organs, 2 cm in diameter dorsai skin, and right ears The both side of the control group and the left side of the treated animals receivd the vehicie only (50% ethanol and 50% propylene glycol vol/vol). Data are expressed as means f SEU ++p<0.01vs left side, and *P<0.01 vs right side of Gonadectomy controls.

Figure 2. Effect of percutaneouS adiministration of DHEA or ~4-dioneon the ventral (panel A), dorsal (panel 0) prostatic, and seminal vesicle (panel C) weights in male hamsters. The treatment is as deskribed in Fig- 1. Data are expressed as means f SEM. **pc0.01 vs ORCH control.

Figure 3. Effect of percutaneous adiministration of DHEA or Ar-diane on the uterus (panel A), and vaginal (panel B) weights in female hamsters. The treatment is as described in Fig. 1. Data are expressed as means f SEM. **p

Figure 4. Effect of percutaneous adiministration of DHEA or ~4-dioneon 38- HSD (panel A), I7B-HSD (panel B), Sa-reductase (panel C), and aromatase (panel D) in male hamster &in. The treatment is as described in Fig. 1. The 38-HSD, 17&HSD ,la-reductase , and aromatase were measured by the formation of A4- dione from DHEA, testo from ~d-dione,DHT from testo, and El from DHT, respectively. Incubation were perfonned at 31oC for 150 min with 0.5 pM labeled substrates and 1 mM correlated cofactors. Data are expressed as means f SEM of pmol product formed /mg protein/min. *P

Figure 5. Effect of percutaneous adiministration of DHEA or d'-&one on 38- HSD (panel A), 17B-HSD (panel B), 5a-reductase (panel C), and aromatase (panel D) in female hamster skin. The treatment is as described in Fig. 1 and the enzymatic assay is measured as described in Fig. 4. Data are expressed as means f SEM of pmol product fonned /mg protein/min. .*pc0.01 vs OVX control. Figure 6. Effect of percutaneous adiministration of DHEA or ~4-dioneor prostatic 3B-HSD (panel A), 17B-HSD (panel B), and Sa-reductase (panel C~ activities in male hamster prostates. The treatment is as desaibed in Fig. 1. Th 3P-HSD , 17mD, and 5aweductase activities were measured by the formatior of Ar-dione from DHEA, testo from A%iione, DHT from testo, respectively. Thc reactions were perforemed at 370C of 150 min: for 3P-HSD, 60 min for both 178* HSD and 5a-reductase activities with 0.5 pM labeled substrates and 1 mhi correlative cofactors. Data are expressed as means f SEM of pmol produc formed /mg protein/min. .*p<0.01 vs ORCH control.

Figure 7. Effect of percutaneous aâiministration of DHEA or Ad-dione on uterui 3P-HSD (panel A), 17p-HSD (panel B), and Sa-reductase (panel C) activities ir female hamster prostates. The treatrnent is as desaibed in Fig. 1. The 3P-HSD 17$-HSD ,and Sa-reductase activities were measured as described in Fig. 6. Dat; are expressed as means f SEM of pmol product formed /mg protein/min ++p

Effect of .percutaneous adirninistration of DHEA and ~Uioneon serurn LHand FSH concentrations in male and female hamsters.

FSH (ndml)

Female Male 1 Female

Intact 1.44 f 0.41 0.42 f.O.08

Gonadectomy 4.29 f 0.52 4.74 f 0.56

Gonad + DHEA 3.59 f 0.33, 6.52 f 0.55

Data are expressed as means f SEM (n=6). *, p < 0.05, p c 0.01. SURFACE AREA OF HAMSTER FLANK ORGAN (mm2) A N O P O O O O O

FIGURE 3

1 CONT 1 DHEA ~A~~DIONE~ CONT 1 DHEA 1~4-DIONE INTACT INTACT OVARIECTOMIZED FIGURE 4 FIGURE 5

INTACT -- OVARIECTOMPED FIGURE 6

CONT DHEA asDDHE INTACT ORCHIECTOMKED FIGURE 7

CONT DHEA JWONE INTACT THE CONTROL OF ANDROGEN ACTION IN PERIPHERAL TISSUES IN THE HAMSTER In the skin, the dermis is composed of large sebaceous glands and hair folücle! besides many 0th- components. They are highly androgen-dependent tissues Excess of DHT or increasing local sensitivity to androgens may be the majoi reason of some skin disorders, such as acne, hirsutism, and male patterr baldness. Thus blockage of Dmaction by either inhibithg the DHT formatior from T by Sa-reductase inhibitors, or blocking the peripheral androgen effec through competitive inhibition of the formation of the DHT-androgen receptoi complex by antiandrogens rnay be the logical sites to treat the above mentionei androgen associateci skin disorders. It has been reported that oral administratior of the antiandrogens, such as flutamide, cyproterone acetate, and spinolactom can telieve the syrnptoms, however, such drugs can only be used in womer using adequate contraception. Therefore, an ideal compound for topical use witk only local effects and without any systemic side effects is desired. in this chapter we have used the hamster flank organ and ear models in the saeening of s number of synthesized antiandrogenic compounds for the above mentionec purpose. in the first article, we have demonstrated both the local and systemic inhibitory effects of antiandrogen flutamide and Sa-reductase inhibitoi finasteride after percutaneous administration. Furthermore, we have alsc demonstrated that EM-250, one of 17s-alkynyl substituted T derivatives, is s pure topical antiandrogen (article two). Finally, we have examined the effects oi a series of 17B-(N-alkyl/aryformamido)- and 178-[(N-arkyl/aryl: alky/arylamido]-4-methyl-4-aza-3-oxo-5a-androstane-3-onesas Sa-reductas~ inhibitors in hamster flank organs. The results showed that EM-402 is a poteni topical Sa-reductâse inhibitor and is discussed in the article three. LOCAL AM) SYSTEMIC REDUCIION BY TOPICAL FiNASTERIDE OR EUTAMIDE ON HAMSTER FLANK ORGAN SIZE AND ENZYME ACTIVlTY

Cailin CHEN, Libertad A. PUY, Jacques SIMARD, Xun LI, Shankar M. SNGH and Femand LABRIE

MRC Group in Moleculnt End~crinology,CHUL Research Center and Lova1 University, Québec, GI V 4G2, Canada

J. Invest. Derma fol. 105 (1995)678-682. ABSTRACT

The hamster flank organ and the inner ear skin are widely used models O: the control of sebaceous gland activity by androgens and antiandrogens Finasteride, a Sa-reductase inhibitor was administered locally on the surface O. the right flank organ and right ear twice daily for 4 weeks. The treatment causec similar 12% to 30% reductions in the size of the sebaceous glands on both flad organs. Similar results were observed on the size of the sebaceous glands of botl ears. Moreaver, relative mRNA levels of the androgen-regulated FAR-17a gent measured by in situ hybridization as well as [3H]-thymidine incorporation anc Sa-reductase activity were similarly decreased in the two flank organs aftei topicai application. The pure antiandrogen flutamide, at the same doses, exertec a more potent effect on al1 the same parameters, and the effect was alsc comparable on the treated and untreated side of flank organs and ears Finasteride and fiutamide significantly decreased ventral and dorsal prostatic weights after topical application. The present data show that the topica. administration of finasteride, in analogy with flutamide, causes local inhibitior of sebaceous gland growth in both the costovertebral organs and eus. However as demonstrated by the similar inhibitory effect in the contralateral untreatec side and the reduced weight of the dorsal and ventral lobes of the prostate anc seminal vesicies, finasteride, as weli as flutamide both exert significant systemic effects. INTRODUCTION

It is well recognized that acne, seborrhea, hirsutism and androgenic dopecia are asçociated with excess androgens and that, moreover, the skin synthesizes androgens from inactive steroid precursors [l-61. In fact, inueased local biosynthesis of 5miihydrotestosterone (Dm),the most potent androgen from the weaker androgen testosterone, by the enzyme Sa-reductase has been suggested to be one of the mechanismi involved m. Because androgens play such an important role in the etiobgy of these common diseases, the discovery of locally effective antiandrogew or androgen antagonists would be of major interest. Previously, many compounds have ben investigated in laboratory experiments and in ciinical trials, namely spironolactone [BI, cyproterone acetate [9], 17a-methyl-B-nortestosterone[IO] and flutamide [Il]. The problem with these compounds, however, is their systemic action whereby these drugs might, among other effects, reduce libido and impair spermatogenesis in men and feminize male fetuses in pregnant women. It is thus important to make available a compound that could exert local activity in the skin and its appendages after topical application without causing the above-mentioned systemic side effects.

The flank organs in the hamster are oval, darkly pigmented structures located in the costovertebral region, that ~owistmainly of large sebaceous glands and hair follicles. The development, growth and secretion of these glands are highly regulated by androgens and this animal mode1 has been widely used to investigate the activity of antiandrogenic compounds [12-151. Similady, the skin of the inner ear lobe contains a dense layer of sebaceous glands that are also highly sensitive to androgens [16-181.

Finasteride (MK-906, Proscar, N-[l,l-dimethyl-ethy1]-3-oxo-4-aza-5a- androst-1-ene-17p-carboxamide)is a specific inhibitor of testosterone Sa- reductase without binding to the androgen receptor in the rat, dog, monkey and human [19,20]. In this study, finasteride was administered topically on the right flank organ and ear of the Golden Syrian hamster. Besides morphomettic and histologie evaluation, namely size of flank organs, size of sebaceous gIands and [OH]-thymidine incorporation in the sebaceous gland cells, we have also measured the relative levels of FAR-17a mRNA by in situ hybridization. This gene first isolateci and characterized by Seki et al. [21] is highly sensitive to androgens and is expressed at high level in the hamster Bank organ. Castration causes a marked deaease in the levels of FAR47a mRNA within a few days to undetectable levels wherease topical application of testosterone and dihydrotestosterone fully restores mRNA Ievels [22]. The dot-blot hybridization data obtained were, however, semiquantitative, and the measurement of FAR- 17a mRNA leveIs by in situ hybridization in hamster flank organ was not previously performed. We have thus compared the effect of finasteride on the above-mentioned parameters to the effect of the antiandrogen flutamide, which is already known to affect some of these parameters [13]. MATERIALS AND MEïHODS

Male adult Syrian hamsters (110-120 g) were purchased from Charles-Rive] Laboratories, (St-Constant, Québec) and housed five per cage in experiment : and one animal per cage in experiment 2 in a light (14h light/day) anc temperature (22 f l0C)-controlled environment. The animals were housed ii plastic boxes on sawdust and ad libitum access to hamster chow (Agway Countq Food hc., Syracuse, NY) and tap water.

Finasteride was synthesized in our medicinal chernistry division followiq the synthetic approach of Rasmusson [23]. The compound was 298% pure b] HPLC. Flutamide was kindly provided by Dr. R. Neri, Schering-Plougl Laboratories, (Kenilworth, Nj).

In experiment 1, the hamsters were randomly divided into seven groups: 1: intact controls; 2), 3), and 4) 30,100 and 300 pg of finasteride; 5), b), and 7) 30,100 and 300 pg of flutamide. At the beginning of the experiment, the ear was markec for identification purpose, and the area of the flank organ was shaved witl animal clippers. The appropriate doses of finasteride and flutamide wen dissolved in 50% ethanol-50% propylene glycol (vol/vol), 10 pl wai administered on the right side of the hamster flank organs as well as on the ears Meanwhile, the intact pupand the left side of the flank organs and ears of fh finasteride and flutamide-treated group received the vehide only. To examint the possibility of oral intake of finasteride due to the hamsters grooming, wt performed an experiment comparing the activity of the 100 pg dose of finasteridr in a group of animals in which the flank organ was covered with microparc surgical tape with another group in which the right flank organ was lefi uncovered. The length and width of the darkly pigmented oval spot outiining each flank organ were carefully measured with a vernier caliper (Fishei Scientific). Flan. organ area was then calculated according to the following formula: area of ellipse = .rc (L/2 x W/2) = 3.14 (Lx W)/4. The animals were killed by decapitation. The flank organs and ears were then removed quickly and fixed for histologicai examination and in situ hybridization or frozen in ice and stored at 40°C for enzymatic assays. The ventral and dorsal lobes of thi prostate as weii as the seminal vesicles were removed, dissected, freed from fa and connective tissue, and tapidly weighed. The ventral lobe of the prostate ant the testes were frozen and kept at -80°C

The pieces of hnk organs and ears were fixed in 4% paraformaldehyde fo 48 h and then sinseci in 15% suaose phospha. buffer for 24 h at 4' C. The flanl organs were then cut in the middle in the rostral to caudal direction while th ears were cut between the first and second cartilage ridges. The tissues were thei embedded in tissue-tek (Miles Inc., Diagnostic Division, Elkhart, USA) and CU with a cryostat in eight pm-thick sections before staining with hematoxylin eosin. The surface areas of the sebaceous glands of flank organs were drawn oi the photographs with a graphics tablet (Xewlett-Packard 9111A graphics tablet while the surface areas of the sebaceous glands of the ears were estimatec directly under optical miaoscopy using a grid. The results were calculated ii units and expressed in percentage while the control group was taken as 100%.

The 474 base pair cDNA insert used in this study was obtained b! amplifying part of the nucleotide sequence of the FAR-17a cDNA [21]. Thc reactions were performed with 1 pg of poly (A+) hamster flank organ mRNA a, template for reverse transcriptase (Pharmacia, 2500 U/ml) in a buffer containini MgCl2 (5 mM), KCl (50 mM), Tris-HC1 (10 mM), dNTPs (1 mM) and th( downstream primer (5'-CCACTGGGTA-TTATCATCTTCTTCTTG-3')at i concentration of 25 PM.The reactions were performed for 60 min at 37OC according to Perkin-Elmer's "Gene amplification" RNA polymerase chaii reaction kit (mine, CA). Further amplification of the cDNA fragment wai achieved in 1W @ of the same buffet except that the MgCl2 concentration wa: about 2 mM, using a step program (95OC, 1min; 60°C, 1min; 7PC, 1.5 min) afte a 10-min incubation at 100°C before adding the upstream prime: (S'GGAC~GTACITGACGCT'CCTT-3')at a concentration of 25 pM an( 2.5 U Taq polymerase for a total of 35 cycles. Mer verification of its size on 1%agarose non-denaturating gel, the PCI fragment was purified following the Gene clean II protocol (La Jolla, CA) an^ subcloned in the Bluescript plasmid (pBSKSII+; 2.85 Kb Stratagene hc., La Jolli CA). Nucleotide sequence of the arnplified cDNA fragment deterrnined by th modified dideoxy chah reaction method Sequenase kit (Pharmacia LK1 Biotechnology), showed that it was identical to the FAR-17a cDNA sequenc reported by Seki et al 1211, and corresponded to the nucleotides 276 to 749. Th 474 cDNA fragment was then cut from the vector using Hind Iïï and Xhc restriction enzymes, and the size of the isolated insert was deterrnined by 19 agarose gel electrophoresis and purified by Gene Clean II (La Jolla, CA). Th fragment was then radiolabelld with [&%] dCTP (~mersham,Oskvilie, Ont., : 3000 Ci/mmole) by the random primer method (241 to high specific activity (IO dpm/~rg)*

tn. situ. hv- 0. . Flank organs fixed for 1-2 h ui 4% (w/v) paraformaldehyde in 0.1 h phosphate buffer (pH 7.4) were rinsed in the same buffer for 10 min and then fo 30 min in 0.05 M phosphate buffer containing 15% suaose at 4OC. The tissue were then embedded and cut as described above for preparation of the sections ti be used for measurement of the size of the sebaceous glands. Frozen sections, 01 the other hand, were mounted on gelatin- and polylysine-coated slides and kep at 40°C until hybridization. On the day of hybridization, the samples werl acetylated with 0.25% acetic anhydride in 0.1 M triethanolamine (pH 8.0) for 11 min, before immersion in 2 x SSC (1 x SSC = 0.15 M NaCl/O.OlS M Na citrate, pI 7.0) for 10 min. Prehybridization was performed as described previously 1251 Briefly, the reaction was performed for 1-2 h in a buffet containing 50% (v/v formamide, 5 x SSPE (1 x SSPE = 0.18 M NaCl, 10 mM NaH2PO4 and 1 mh EDTA, pH 7.4), 2 x Denhardt's solution, 0.1% SDS, 200 &ml denatured salmol testis DNA, 200 pg/ml yeast transfer RNA/d and 200 &ml poly CA. 'Zhc slides were then hybridized for 16 h at 42°C in 0.2 ml of prehybridization buffe containing, in addition, 4% dextran sulfate, 10 mM dithiothreitol and 2 x 10 cpm heat-denatured [3SS]-labelled FAR-17a cDNA probe. The slides werc subsequently washed at room temperature in 2 x SSC for 2 h, 1 x SSC for 2 h ,O.! x SSC for 1 h, 0.5 x SSC for 1h at 42°C and 0.5 x SSC 1 h at room temperature an( dehydrated in a graded ethanol series and exposed to Kodak4 omat films for : days at -70°C. In order to determine the amount of nonspecific backgount hybridization, flank organ sections were treated with pancreatic nbonudease-1 (200 pg/ml) and ribonuclease Tl (150 U/ml) for 1 h at 37OC beforl prehybridization. The tissue specificity of the probe was asceitained b: hybridizing sections from hamster kidney, liver, testis and spleen with the FAR 17a cDNA probe. No specific hybridization could be obsetved. Densitometri measurement of autoradiographs of flank organs from all experimental group was performed wing a Bioirnage analysis systern (Millipore, USA) and the me& optical intensity value per mm2 of labeled areas for each section was thu caldated.

Ten pCi [methylJH)-thymidine (82 Ci/mmol, Amersham) was diluted il 100 pl 0.9% NaCl and locally injected subcutaneously into flank organs 2.51 before sacrifice. Each flank organ was then cut in the center and fixed in Bouin' fluid for 2 days. All tissues were dehydrated in a graded series of ethanol am embedded in paraffin. Six 7 p-thick consecutive sections were cut for ead specimen and mounted on glass slides before drying for 24h at 3TC. nie slide: were then deparaffinated with toluene and a series of ethanol bath. Afte drying, the sections were coated with a liquid photographic emulsion (Kodal NTB-2 Eastman Kodak, Rochester, N.Y.) and exposed .to au toradiography for : weeks. The sections were then processed and stained with hematoxylin+osin The total nurnber of labeled basal cells was counted under optical microseopy fo. each section and the means of si; consecutive sections were cdculated for ead flank organ. The results are expressed as a percentage change while the tota number of labeled cells in the control group is set at 100%.

Measurement of 3$-HSD, 17fbHçD and Sa-reductase activities wa! performed as described previously [26].

Statistical significance was measured according to the multiple-range test o. Duncan-Kramer [27. Data are expressed as means f SEM. RESULTS As iîlustrated in Fig. 1, topical administration of finasteride on the dg1 side flank organs and ears of ,intact hamsters at the doses of 30, 100 and 300 p twice daily for four weeks caused 4.3% (nonsignificant, N.S.), 19.5% (p<0.01) an 30.2% (pc0.01) decreaçes in the. surface areas of the right side flank organs whü fiutamide uused 6.5% (N.S.), 47.2% @ < 0.01) and 56.F (p c0.01), deaease, respectively. In fact, the surface area of coneol flank organs was measured i about 30 mmz, while the surface areas of the three finasteride-treated grour were measwed a 28.8.24.3, (pc0.01) and 21.0 (pcO.01) JI& respectively. It ca be seen in the same figure that treatment with finasteride induced 6.3% (N.S. 19.2% (p<0.01) and 29.1% (p<0.01) redukons in the surface areas of the left sid (untreated) flank organs while flutamide caused 7.9% (N.S), 45.9% (p < 0.01) an 55.1% (p < 0.01) reductions, respectively.

Figure 2 shows the changes in size of the underlying sebaceous glands c flank organs from animals who received inaeasing doses of finasteride c flutamide. Treatment with Finasteride with the 30 pg dose caused a 15.9' (pc0.05) reduction on the left side and a 12.4% (N.S.)reduction on the right sid compared with the intact control group. At the 100 pg dose, a 16.7% reductio was observed on flank organs of both sides (~~0.05)while a 29.9% .decreas (p<0.01) on the left side and a 29.4% decrease on the right side (p<0.01) wa observed at the dose of 300 pg. On the other hand, the 300 pg dose of flutamid caused a maximum decréase of 52.0% (p < 0.01) on the left side and a 52.5% (p 0.01) on the right side of flank organs.

As shown in Fig. 3, the relative mRNA levels of the androgen-regulate FAR-17a gene measured by in situ hybridization were inhibited by 2.7% (N.S. 15.5% (~~0.05)and 20.6% (pc0.01) in the left side Bank organs and by 1.1% (N.S. 15.1% (p<0.05), and 20.9% (pc0.01) in the tight side flank organs in a dos6 dependent manner after 4 weeks of treatment with finasteride. Flutamide, O the other hand, caused 14.4% (N.S.),24.9% (p < 0.01) and 34.0% (p < 0.01 inhibition in the left side flank organs while 13.6% (N.S.), 26.0% (p < 0.01), an 35.5% (p < 0.01) inhibitory effects were measured in the right side flk organ! Fig. 4 illustrates typical autoradiographs of flank organ sections from hamster treated with finasteride and flutarnide and hybridized with the .FA&17a cDNA probe, thus permitting direct vkdization of the results presented in Fig. 3.

Figure 5 illustrates the results of PH]-thymidine incorporation into sebaceous gland ceils in the flank organs. This parameter indicates the proliferation of sebocytes and is considered as being a highly sensitive parametex of response of the sebaceous glands. It can be seen that finasteride, at the dose 01 30 pg, already deaeased the total number .of labled cells by 24.4% on the left side (p

Figure 6 illustrates the light microscopie autoradiography of hamster flanb organ sections following [3H]-thymidine incorporation in the' typical groups illustrated in Fig. S. It can be seen that the most densely labeled cells are in thf basal layer of the sebaceous glands and that siiver grains are present exclusivelj in the nudei. The total number of labeled cells corresponds to approximately 20. 25% of total basal cells in the intact control group (data not shown).

It can be seen in Fig. 7A that finasteride exerts a potent inhibitory effect or Sa-reductase activity in both treated and untreated hamster flank organs, In fact even at the low dose of 30 pg, twice daily, the deaease is highly significant a 37.5% (pc0.01) on the left side and 54% on the right side (pc0.01). At the dose! of 100 pg and 300 pg, treatment with finasteride caused further deaeases by 42.2% and 51.6% on the left side and 57.8% and 61.3% on the right side, respectivelj (~4.01for al1 groups). Although treatment wifh flutarnide had no significan effect at the low dose, the 300 pg dose caused a 23.4% (p c 0.05) decrease in th left side and 42.2% (p < 0.01) deccease in Sa-reductase activity on the right side.

Figure 78 shows 17p-HSD activity in the flank organs after four weeks o. treatment with finasteride and flutamide. Finasteride cause'd 34.5%, 27.6% anc 33.3% decreases on the untreated left side (p < 0.01 for al1 groups) and 36.2%, 37.4% and 37.4% decreases on the treated right side (p < 0.01 for al1 pups) while flutamide showed some inhibition ody at the highest dose used, nameiy a 23.0% (p < 0.05) deaease on the left side and a 27.0% @ < 0.01) inhibition on the right side at the 300 pg dose. *

Results very similar to those described above on the size of the-sebaceous glands in the flank organs were observed on the size of the sebaceous glands of the ears. In fact, treatrnent with finasteride caused 19.7% (p < 0.05), 23.1% (p < 0.01) and 30.7% (p < 0.01) decreases in ear sebaceous gland area on the treated side and 16.6%, 22.2% and 26.4%. @ < 0.01) decreases on the contralateral side following treatment with the 30,100 and 300 pg doses of finasteride, respectively, flutamide, on the other hand, caused 30.4%, 39.0% and 67.8% reductions on the left side and 25.2%, 48.0%, 67.4% inhibition on the right side (p c 0.01 for all the groups) following treatment with the same doses (Fig. 8).

It is well recognized that prostatic and seminal vesicle weights are under stringent control by androgens and are the most widely used mode1 for assessing androgenic and antiandrogenic activity [28-30). As illustrated in Table 1, finasteride caused 11.9% (p < 0.05), 13.9% (p < 0.05) and. 27.5% (p < 0.01) inhibitions Ln ventral prostate weight, wherease flutamide caused a deaease of 15.7% (p c 0.05) at the largest dose used. On the other hand, finasteride caused 40.3%, 45.3% and 46.5% inhibitions @ < 0.01 for all doses) in dorsal prostate weight, and flutamide caused 17.5%, 19.4%, 25.2% reductions in the value of the same parameter. As shown in the same Table,- finasteride causes 32.7%, 36% and 37.7% inhibitions in seminal vesicle weight compared to the control group at the 30,100 and 300 pg doses, respectively, wherease 300 yg flutarnide caused a decrease of 19.4%.

Table 2 shows the testicular 3B-HSD, 17p-HSD and Sa-reductase activities after four weeks of treatment with inaeasing doses of finasteride and fiutamide. Finasteride showed no significant effect on these parameters except a 55.9% inerease of 5a-reductase activity at the dose of 300 pg, while flutamide caused a significant increase in the three enymatic activities, which is probably related to the accornpanying humase in sem LH and EH le& (data shown). In order to avoid the possibility of oral absorption of the drugs by thl hamsters' grooming, an experiment was performed where the sites O application were covered. Finasteride and flutamide decreased the surface arei of flank organs and underiying sebaceous glands as weil as sebaceous glands a ears on both the treated and untreated sides. In fact, the efficacy was similas ti that obtained in the above described data, where the same 100 pg, twice daily, wa used. Furthetmore, as mentioned above, these two compounds also signihcantl; inhibited the size of the ventral and dorsal prostates and seminal vesides afte topical application and inhibited Saoreductase activity in both sides of flad organs. AU the parameters measured show no significant difference betweén fi uncovered group and the covëred group for both finasteride and fiutamide. The present data clearly demonstrate that the topical administration of th Sa-reductase inhibitor finasteride significantly deaeases the size of the sebaceoi; glands in hamster flank organs, this effect king accompanied by deaeased PH thymidine incorporation, size of the ear sebaceous glands, FAR-17a mRN. levels and size of the flank organs. It is weii-know that hamster flank orgar consist of large sebaceous glands, hair foilicles and dermal pigment related to tk presence in the demis of dendritic cells containing dense structures resemblin stage IV ineIanosomes and that all these components are highly androgei sensitive [31,32].

It has been reported thaï the pigmentation of the flank o&ms and th growth of their underlying sebaceous glands may have different thresholds c response to androgenic or antiandrogenic stimuli [33]. It was thus felt moi reliable to measure the effect of treatment with potential Sa-reductase inhibit01 or antiandrogenic compounds in both the flank organs and ears. As can be see in the present study, comparable effeds were observed in the sebaceous glands c the flank organs and ears. An important observation, however, is that the effeci observed are almost identical in both'the treated and untreated sides of bot flank organs and ears. Such data clearly indicate that finasteride is absorbe through the systernic circulation and thus reaches the contralateral flank orga as weii as other tissues. As confirmed in the present study, similar results ai observed with the antiandrogen flutarnide [13].

The decrease in the size of flank organs and size of the underlyin sebaceous glands was accompanied by a decrease in PH]-thymidin incorporation, as well as FAR-17a mRNA levels in the sebaceous glands. 1 addition, treatment with finasteride was accompanied by a marked inhibition c Sa-reductase activity in both fia& organs at the lowest dose used, namely 30 pl It remains to be seen, however, if this decrease in Sa-reductase activity is due t inhibition of Sa-reductase gene expression, protein content and/or activity, a due to Finasteride remaining in the cels at time of sacrifice. .

The present study aiso show that changes in the levels of FAR-l7a ~RNA can be quantified by in situ hybridization. This technique does II& require th extraction of RNA from tissue, thus eliminating an important source of assay variation. Moreover, the observation of flank organ sections under the light microscope after hybridization can provide useful information on the cellular localization of the mRNA of interest and make possible a correlation between changes in mRNA content and gross histological aspect of the tissue. The availability of in situ hybridization of FAR-17a mRNA should'facilitate study of the factors involved in the control of androgen-regulated gene expression in this tissue and thus help to assess the importance of various steroids and other factors, in the androgenic regdation of flank organ activity.

Finasteride has been found, after oral administration, to cause a marked reduction in the intraprostatic concentration of Dmand a 2530% reduction in prostatic size in men [Ml.Comparable effects were observed in the present study after topical administration In fact, the weight of the ventral and dorsal lobes of prostate as well as that of the seminal vesicles was already significantly inhibited at the lowest dose of finasteride used. Such data, in addition to the action of Finasteride on the contralateral flank organ and eu, show the systemic action of the 5a-reductase inhibitor after local application on the skin.

Harris et al. [35] reported that finasteride has higher affinity for hurnan prostatic Sa-reductase than scalp 5a-reductase, the concentration of finasteride required to cause 50% inhibition in the prostatic homogenate being 4.2 nM while being 500 nM for the scalp homogenate. Recently, two types of Sa- reductase cDNAs have been isolated in the hurnan and rat, and the structure of the corresponding genes in the human has been elucidated [36-391. The somewhat higher sensitivity of the prostate and seminal vesicles than the sebaceous glands to the inhibitory action of finasteride observed in the hamster could be explained by the possible difference in sensitivity to the Saoreductase inhibitor of different tissue-specific isoenzymes. Elucidation of the structure of the variow types of Sa-reductase expressed in the sebaceous glands, prostate and seminal vesicles in the hamster and study of their enzymatic characteristics should provide answer to this question. In fact, there are many reports describing the presence of Sar-reductase isoenzymes in both the rat and human [40-42]. It çhould be mentioned that treatment with finasteride can reverse the balding process adenhance hair regrowth by topical application in the balding stump tail macaque due to its inhibition of 5a-reductase [43]. Compounds having systemic action can affect all androgen-sensitiv organs and functions such as, for example, male sexual differentiation, growf and function of the prostate and acceçsory sex organs in the male, as well as bon formation and resorption. It is expected that compunds having such potentii effects due to their systemic action have Limitations for the treatment c androgenlsençitive abnormalities such as acne, seborrhea, hirsutism an1 androgdc alopecia. The present data demonstrate that finasteride, at least i the hamster, has significant systemic effects after topical administration i: relatively small amounts on the skU1, a characteristic shared by the antiandroge: flutamide. AQ(N0WLEDGEMENTS

The authors thank Ms. France Lepire for her skillful technical assistance. REFERENCES

Milne JA: The metabolism of androgens by sebaceous glands. Br f Dermatol 81: (Suppl2): 2328,1969

Birgham KD, Shaw DA: The metabolism of testosterone by human male scalp skin. J Endocrino1 57A11-121,1973

Liang T, Rasmusson GH, Brooks JR: Biochemical and biological studies with 4aza-steroidôl Sa-reductase inhibitors. J Steroid Biochem 19: 385-390, 1983

Mauvais-Jarvis P, Bercovia JP,Gauthier F: In vivo studies on testosterone metabolism by skin of normal males and patients with the syndrome 01 testicular feminization. J ChEndocrinol Metab 29: 417421, 1969

Wilson JD,Walker JD: The conversion of testosterone to 5a-andsostan-17p 01-3-one (dihydrotestosterone) by skin slices of man. J Clin Invest 48: 371- 379,1969

Labrie F: Intraainology. Mol Cell Endocrinol78: C113-C118,1991

Kuttenn F, Mowszowicz 1, Wright F, Wright F, Baudot N, Jaffiol C, Robin M, Mauvais-Jarvis P: Male pseudohermaphroditism: a comparative s tudy of one case of 50-reductase deficiency with tluee complete forms 01 testicular feminization. J Clin Endocrinol Metab 49: 861465,1979

Goodfellow A, Alaghband-Zadeh J, Carter G, Cream JJ,Holland S, Scully J, Wise P: Oral spironolactone impsoves acne vulgaris and reduces sebm exaetion. Br J Dermatol111: 209-214,1984

Burton JL, Laschet U, Shuster S: Reduction of sebum excretion in man by the antiandrogen, cyproterone acetate. Br J Dermatol89: 487-490,1973

Zarate A, Mahesh VB, Greenblatt RB: Effect of an antiandrogen, 17a- methyl-Enortestosterone on acne and hirsutism. J Clin Endocrino1 Metab 26: 1394-1398,1966 Cusan L, Dupont A, Manger A, Tremblay RR, Manhes G, Labrie 1 Treatment of hirsutism with the pure antiandrogen flutamide. J Am Aca Dermatol23: 462-469,1990

Burdick KH, Hill R: The topical effect of the antiandrogen chlorrnadinon acetate and some of its chemical modifications on the hamstt costovertebral organ. Br J Dermat0182 (suppl6): 19-25,1970

Lutsky 8, Budak M, Kozial P, Monahon M, Neri R: The effect of nonsteroidal antiandrogen, flutamide, on sebaceous gland activity. J Inve! Dermatol64: 412-417,1975

Luderschrnidt C, Eierma~W, Jainny J, Bidlingmaier F, Ring J: 170 propylmesterolone (SH 434): an antiandrogenic sebosuppressive substanc not influencing circulating testosterone concentrations. Experimenti studies in Syrian hamsters. Ar& Pharmacol 328: 214-218,1984

Brooks JR, Primka RL, Berman C, Krupa DA, Reynolds GF, Rasmusso GR Topical antiandrogenic of a new 4-azasteroid in the hamster. Steroid 56: 42043,1991

Matias JR,0rent.reic.h N: The hamster ear sebaceous glands. I. Examinatio of the regionai variation by stripped skin planimetry. J Invest Dermatol 8: 43-46,1983

Hamilton JB, Montana W: The sebaceous glands of hamstei morphological effects of androgens on the tegumentary structure. Am Anat 86: 191-233,1950

Plewig G, Luderschmidt C: Hamster ear mode1 for sebaceous glands. Invest Dermatol6& 171-176,1977

Liang T, Heiss CE, Cheung AH, Reynolds GI, Rasmusson GH: 4-azasteroidi Sa-reductase inhibitors without affinity for the androgen receptor. J Bi( Chem 259: 734739,1984

Liang T, Cascien MA, Che- AH, Reynolds GF, Rasmusson GH: Specie differences in prostatic steroid Sa-reductase of rat, dog and huma1 Endocrinology 177: 571679,1985 21. Seki T, Ideta R, Shibuya M, Adachi K: Isolation and characterization of cDNA, an androgen-cegdated mRNA in the flank organ of hamsters. J Invest Dermatol96: 926-931,1991

22. Hisaoka H, Ideta R, Seki T, Adadii K: Androgen regdation of a specific gene in hamster fldorgans. Arch Dermatol Res 283: 269-273,1991

23. Rasmusson CH, Reynolds GF, Steinberg NG, Walton E, Patei GF, Liang T, Cascieri MA, Cheung AH, Brooks JR, Berman C: Azasteroids: structure- activity relationships for inhibition of Sa-reductase and of androgen receptot binding. J Med Chem 29: 2298-2315,1986

24. Feinberg AP, Vogelstein B: A technique for radiolabeling DNA restriction endonuclease fragments to high specific activity. Anal Biochem 132: 6-12, 1983

25. Pelletier G, Labrie C, Simard J, Duval M, Martinoli MG, Zhao HF, Labrie F: Effect of sex steroids on regdation of the levels of Cl peptide of rat prostatic steroid-binding protein mRNA evaluated by in situ hybridization. J Mol Endocrinol 1: 213-223,1988.

26. Maztel C, Meher MH, Gagné D; Simard J, Labrie F: Widespread tissue distribution of steroid sulfatase, 3B-hydroxysteroid dehydrogenase/~5-A4 isomerase (3B-HSD), 17p-HSD Sa-reductase and arornatase activities in the rhesus monkey. Mol Cell Endoainol104: 103-111,1994.

27. Kramer CY: Extension of multiple-range test to group means with unique numbers of replications. Biometrics 12: 307-310, 1956.

28. Labrie C, Simard J, Zhao HF, Belanger A, Pelletier G Labrie F: Stimulation of androgen-dependent gene expression by the adrenal precursors dehydroepiandrosterone and ancirosteneàione in the rat ventral prostate Endoaihology 124: 27452754,1988

29. Labrie C, Cusan L, Plante M, Lapointe S, Labrie F: Analysis of the androge~cactivity of synthetic "progestins" currently used for the treatment of prostate cancer. J Steroid Biochem 28:379-384,1989 Labtie C0 Trudel C, Li S, Martel C, Couet J0 Labrie F: Combination of a antiandrogen and a Sa-reductase inhibitor: a further step towards toti androgen blockade? Endocrimiogy 129566-568,1991

Hamilton JûOMontagna W: The sebaceous glands of the hamster. Morphological effects of androgens on intergumentary structure. Am Anat 86: 191-234,1950

Frost P, Giegel JL, Weinstein GD, Cornez EC: Biodynamic studies c hamster flank organ growth: hormonal influences. J ïnvest Dermatol r 159-167,1973

Goos GMAA, Writz P, Verrnokken AJM: An improved method fc evaluating antiandrogens. Arch Oermatol Res 273: 333-341,1982 hperato-Mffiinley J0 Shackleton C, Stoner E: Cornparison of plasma an1 urinary Cl9 and C21 k-reductase metabolites in subjects treated with th 5a-reductase defidency. Proc 71st AM Meet Endw Soc, p. j32, Abst. 163! 1989

Harris G, Azzolina 8, Baginsky W, Cimis G, Rasmusson GH, Tolma RI Raetz CRH, Ellsworth K: Identification and selective inhibition of a isoenzyme of steroidal Su-reductase in human scalp. Biochemistry 85 10787-10791,1992

Andersson S, Bishop RW, Russell DW: Expression, cloning and regulatio of steroid 5a-reductase, an enzyme essential for male sexual differentiatiox J Bi01 Chem 264: 16249-16255,1989

Andersson S, Russell DW Structural and biochemical properties of clone and expressed huaian and rat steroid Sa-reductase. Proc Nat1 Acad Sci USA 87: 3640-344,1990

Jenkins EP, Hsieh CL, Milatovich A, Normington K, Berman DM, Franck U, Russell DW: Characterization and chromosomal mapping of huma. steroid ,5a-reductase gene and pseudogene and mapping of the mous homologue. Genomics 11: 1102-1112, 1991 39. Labrie F , Sugimoto Y, Luu-The V, Simard J, Lachance Y, Batchvarov C Leblanc G, Dwocher F, Paquet N-Structure of human type lI 5œ-reductae gene EndOQinology 131:1571-1573 ,1992

40. Moore RJ, Wion JD:Steroid Sa-reductase in cultured human fibroblasts. Bi01 Chem 251: 5895-5900,1976

41. Martini L, Zoppi S, Motta M: Studies on the passible existence of two 5a reductases in the rat prostate. J Steroid Biochem 24: 17%182,1986

42. Hudson RW: Cornparison of nuclear Sa-reductase activities in the stroma and epithelial fractions of hurnan prostatic tissue. J Steroid Biochem 26 349-353,1987

43. Diani AR, Mulholland MJ, Shull KI., Kibicek MF, Johnson GA, Schostare: HJ,Brenden MN, Buhl AE: Hair growth effects of oral administration O Finasteride, a steroid Sa-reductase inhibitor, alone and in combinatioi with topical minoxidii in the balding stumptail macaque. J ChEndouino Metab 74: 345-350,1992 LEGENDS TO FIGURES

Figure 1. Effect of topicai appIication of finasteride or flutamide on the righl flank organ and ear on the swface &ea of both flad organs in intact male hamsters. The animals received topical administration of finasteride 01 flutarnide at the doses of 30 pg, 100 pg or 300 pg twice daily for 4 weelcs on the right side nank orgarrs and ears while both sides of the control group and lefi side of the treated animals received the vehide only (50% ethano1/50% propylene glycol (v/v). The results are presented as means f SEM (n=10) +p<0.05, ++p<0.01vs left side d intad control. Wp<0.05, ~~~0.01vs right side ol intact control.

Figure 2. Effect of topical application of finasteride or flutamide on the righi flank organ and ear on the sebaceous gland area of flank organs in intact rnde hamsters. ~reatmentis as described in Legend to figure 1. .The units of the control group are set ai 100%. The data are presented as means î SEM (n=8). +p<0.05, ++p<0.01vs left side of intact animal. *p<0.05, **p<0.01 vs right side 01 intact control.

Figure 3. Effect of topical application of finasteride or flutamide on the righi flank organ and ear on FAR-17a mRNA levels in intact male hamster fld organs. The treafment is as described in Legend to Figure 1. mRNA levels pei mm2 of labelled areas were measured by in situ hybridization using the [35S]- labeled hamster flanic organ FAR-17a cDNA Data are expressed as meam f SEM (n=3). +p<0.05, ++p<0.01vs left side of intact control. 'pc0.05, mp

Figure 4. Typical X-ray autoradiograph showing the in situ hybridization 01 flank organ sections using the [%SI-labeled hamster flank organ FAR-17a cDNA probe. Note that there is a decrease in size in a dose-dependent manne1 following treatment with finasteride and flutamide. Treatment was.as described in Legend to Fig. 1.

Figure 5. Effect of topical application of increasing doses of finasteride 01 flutamide on the right flank organ and ear on [WJ-thymidine incorporation in the sebaceous glands of the flank organs of intact hamsters. Thé total numbet 01 labeled cells was counted in six consecutive sections from each flank organ. Th total number of labeled cells in the control group is taken as 100%. The data ar presented as means f SEM (n=3). +p<0.05, ++p<0.01vs left side of control grou1 *p<0.05, **p

Figure 6. Histological sections of [3H]-thpidine incorporation in hamster flan organs. The experiment was as described in Legend to Fig 1. Note that th Iabeled cells are mainly located on the basal ce& of acini. Exposure the: weeks. Bar: 20 pM.

Figure 7. Effect of topical application of increasing doses of finasteride a flutamide on the right flank organ and ear on Sa-reductase activity (panel A and 17P-H!5D activity (panel B) in the flank organs of intact male hamster2 Treatment is as desaibed in Legend to Figure 1. The 5a-reductase activity wa measured by the formation of [W]-DHT from [14C]-testosterone by flank orgai homogenates while 178-HSD activity was measured by the formation of [l4C] testosterone from [14C]-A*-dioneand the results are expressed in nrnol produc formed/mg protein/min. Data are expressed as means f SEM (n=6). ++p<0.01v left side of intact control. **p<0.01 vs right side of intact control.

Figure 8. Effect of topical application of increasing doses of finasteride o flutamide on the sebaceous gland area of ears in intact male hamsters Treatment is as described in &end to Figure 1. The total units in the intac control are set at 100%. The data are presented as means f SEM (n=8). +p

Inhibition of prostate and seminal vesicle weight by topical finastende and flutamide applied on the right flank organ and eu.

Treatment

Control Finasteride (pg)

1 Ventral prostate weight (mg)

Dorsal prostate weight (mg)

Seminal vesicle 1 weight (mg)

Data are expressed as means f SEM (n=lO). g p < 0.05; **, p < 0.01. Table 2

Topicai fiutamide &d finasteride differentiaily stimulates testicular stemidogenic enzyme activity. The campaunds were applied on the right fimk organ and ear

Treatment

Finasteride (pg) Fiutamide (pg)

Data are expressed as means f SEM (n4). *: p c 0.01. FIGURE 1

FINASTERIDE

FIGURE 3

CONTROL FIGURE 4 [ HI-THYMIDINE LABELED CELLS ( % OF CONTROL ) A 4 h) 0i œ O O O g O O O O

17R-HSO ACTIVITY IN FLANK ORGAN sir -HEDUCTASE ACTIVITY IN FLANK ORGAN ( pMOL TEST0 FORMEDlmg prolilnlmln ) ( pMOL OHT FORMEDlrng protdnlmln ) 8e;b4 L: SC: FIGURE 8

1 CONTROL A PURE TOPICAL ANTIANDROGEN: EM-250

Caiiin CHEN, Claude TRUTEL, Yves MERAND, Jacques SIMARD and Femand LABRIE

MRC Group in Molecular Endocrinology, CWUL Research Center and Laval University, 2705 Laurier Boulevard, Québec, G1V 4G2, Canada EM-250 is the one of the 17p-alkynyl substituted T derivatives that inhibit! the stimulatory effect of dihydrotestosterone (Dm)on the growth of androgen sensitive Shionogi cells in vitro (Ki = 96 nM) at concentrations comparable t< those of hydroxy-flutamide (Ki = 64 nM), the active metabolite of flutamide ix vitro. Administration of EM-250 at doses of 30, 100, 300 pg twice daily for 4 weeks on the Syrian hamster right flank organs and right ears si@cantlj reduced the right flank organ area, the underlying sebaceous gland volume, the sebocyte proliferation, and the right ear sebaceous gland area in a dose dependent manner without affecting the size of the fiank organ on the lef contralateral control side, where the vehicle only was applied according to thc same schedule. in same expriment, EM-250 had no effect on hamster plasmi testosterone (testo) and Dmconcentrations or on prostatic and seminal vesich weights. Systemic treatment with EM-250 at the doses of 0.1, 0.3, 1.0, 3.0 rnj twice daily for 6 days in orchiectomized male rats implanted witl androstenedione (Ad-dione) had no effect on A4-dione-induced prostatic weight prostate binding protein mRNA levels, or serum lutenizing hormone concentration. Moreover, twice-daily treatment of ovariectomized rats for r days with EM-250 at doses up to 300 ~(8did not affect ute~eweight. Also, EM' 250 had no inhibitory effect on the activities of human type 1 and type II 3B-HSL and type 1 5a-reductase gene transfected cells or on 17P-HçD and aromatase o. human placenta1 and liver microsomal preparation at a concentration of 1 @ in vitro. At the doses used, EM-250 exerted a potent topical sebosuppressive effect without s ystemic androgenic or estrogenic effects or without inhibitors effect on Sa-reductase and aromatase activity. It suggests that EM-250 is a pure topical antiandrogen. Although several factors are likely to be involved in the pathogenesis c acne, the most important is related to exeess seburn production, which in turn i dependent on androgens [l-31. IL is well known that androgens increase the siz of the sebaceous glands and stimulate sebum production 14-61. Because acn occurs during puberty and has a comparable incidence in boys and girls [7),it i likely that androgens of adsenal orig" [a] are involved in its pathogenesis .

The predominant role of androgens is weU supported by the improvemem in acne and the marked deaease in sebum production observed after systemi treatment with the antiandrogens flutamide [9], spironolactone [IO, 11: cyproterone acetate [12,13], and 17u-methyl-Emrtestosterone [14]. The problea however, with al1 available antiandrogens is their systemic action, even whe~ applied topically on the skin [15]. Moreover, compounds with mixei androgenic-antiandroge~cactivity are not expected to have maximal efficac; [16-21). In fact, clinical studies on the treatment of acne with progestii derivatives that possess mixed androgenic and antiandrogenic activity, such a 17a-methyl-B-nortestosterone and cyproterone acetate, have shown equivoca results [14].

The disappearance of acne and hirsutism in women treated with th( antiandrogen flutarnide [9] clearly indicates that the la& of success of progesti derivatives is due to their intrinsic androgenic activity, their low potency, O their low degree of absorption, not to acne's lack of androgen dependence. T~I better results obtained with flutamide [9] compared with any other progestii derivative, espeaally cyproterone acetate, spironolactone, and l7u-methyl-B nortestosterone [a1 12-14, 221 support the advantages of using an antiandrogei with pure antagonistic activity such as flutamide [16-21) instead of CO~~OU~ with mixed androgenic-antiandrogenic activity such as the pmgestin derivative! available to date.

The search for topicaily appiied antiandrogens for the treatment of acne seborrhea, hirsutism, and scalp hair loss has so far been unsuccessful. Unti now, the efforts have led to compounds (such as cyproterone acetate) that an topically ineffective [23] or possesç systemic antiandrogenic effects that are unacceptable for men and unprotected fertile women [15,24].

We have synthesized a series of antiandrogens with potent local antiandrogenic activity when applied locally on the skin but that are devoid of significant systemic action. In the present study, we des& a new wmpound, EM-250, an antiandrogen that shows potent antiandrogenic activity when applied topicaiiy on the hamster's flank organ and ear, the most wide1y used models for the assay of compounds on sebaceous gland growth and activity [Il, 15, 251; in addition, it shows no systemic antiandrogenic and androgenic effects in the rat. MATERIALS AND METHODS

in vitro The xneasurement of the effects of increasing concentrations of EM-250 and hydroxy (OH)-flutamide on dihydrotestosterone (DHT)-stimulated ce11 proliferation in the androgen-sensitive SEM-107 clone of Shionogi mouse mammary carcinoma cells: Shionogi male mice bearing androgen-sensitive mammary tumors were originaily obtained from Dr. K. Matsumoto, Osaka, Japan, and Dr. Yvome Lefebvre, Ottawa, Canada. The SEM-107 clone was obtained as previously described [26]. Ceiis were plated in 24-well plastic culture plates (Falcon) at a density of 20,ûOO cells/well in MEM supplemented with 1% nonessential amino acids, antibiotics, and 2% dextran-coated charcoal-treated fatal bovine serum (271. Twenty-four hotus after plating, cells were exposed for 10 days to 10-5 to 10-10 moles of EM-250 and OH-fiutamide, respectively, in the presence of 0.3 nM DHT. Media were changed every 3rd day. At the end of the incubation, ce11 number was determined by measwing DNA content as previously described [28]. Data are expressed as the mean f SEM in triplicate dishes. The apparent dissociation Ki values of EM-250 and OH-fiutamide inhibitory action were calculated according to the following equation: Ki=ICso/(l+S/K) [29]. In this equation, S represents the concentration of DHT (0.3 nM), K is the apparent KD of DHT action on ceil proliferation in Shionogi cells (0.1 nM) and ICm is the concentration of the antiandrogen giving a 50% inhibition of DHT action on ce11 prolifera tion.

The measurement of the effects of EM-250 on 3P-HSD, 17$-HÇD, 5a- reductase, sulfatase and aromatase activities: human types 1 and II 3p-HSD and 5a-reductase activities were measured in transfected celIs [30-331;17p-HSD was measured by using human placenta cytosol as described previously 1341. Suifatase activity was assayed with homogenized JEG3 cells (purchased from A&can Type Culture Collection, Rockville, MD) by measuring the total PH]- DKEA formed from PH]-DHEAS, and aromatase activity was measured following the method described by Wing et al. using hurnan liver microsomal preparation [SI. Study in male hamsters Animals and treatments: Male adult Syrian hamsters (110-120 g) wea purchased from Charles River Laborabries (St-Constant, Quebec) and housec three or four per cage on sawdust in plastic boxes in a temperature (22 f 1°C) and light (14 h light/day, lights on at 6 a.m.)-controlled environment. Th animais were given ad libidum access to hamster chow (Agway Country Fooc Inc. Syracuse, NY) and tap water. The hamsters were randomly divided intc one controi group and three EM-250-treated groups that received doses of 3C 100,300 pg, respectively. The compound was dissolved in a solvent containhl 50% propylene glycol and 50% EtOH and then administered on the right flanl organs and ears twice daily for 4 weeks, the control pupand left sides of treatec groups had vehicle applied only. mank organ area was calculated according tc the following formula: area of ellipse = II (L/2 X W/2) = 3.14 (L X W)/4. Th length (L) and width (W) of the darkly pigmented oval spot of flank organ werl measured carefully using a vernier caliper (Fisher Scientific).

The animals were killed by decapitation. The blood was collected fo serum steroid concentration measurement, and the prostates and sernina vesicles were removed rapidly, disseded from connective tissue, and weighed.

easurement of sebaceous -glands The flank organs and ears were fixed in 4% paraformaldehyde for 48 h anc then rinsed in 15% sucrose phosphate buffer for 24 h at 4°C. The flank organ were then cut in the middle in the rostral to caudai direction, and the ears wen cut between the first and second cartilage ridges. The tissues were the] embedded in Tissue-Tek (Miles Xnc., Diagnostic Division, Elkhart, USA) and CU with a cryostat in 8 m-thick sections before staining with hematoxyiin-eosin The surface areas of the sebaceous glands of flank organs were drawn on th1 photographs with a graphics tablet (Hewlett-Packard 9111A graphics tablet) whih the surface areas of the sebaceous glands of the ears were estimated direct11 under optical rnicroscopy using a grid. The results were calculated in units ant expressed in percentages; the control group was taken as 100%. .. . PII1-th~l- Two and a haif hours before the animals were killed, 10 pCi [methylJH] thymidine (82 Ci/mmol, Arnersham) was diluted in 100 pl 0.9% NaCl anc localiy injected subcutaneously into the flank organs. Each flank organ was cut in the center and fixed in Bouin's fluid for 2 days. Ail tissues were dehydratecl in a graded series of ethanol and embedded in paraffin. Six consecutive sections (7 thick each) were cut for each specimen and mounted on glas slides before drying for 24 h at 37oC. The slides were then deparaffined with tolu& and a series of ethanol bath. After drying, the sections were coated with a liquid photographic emulsion (Kodak NTB-2 Eastman Kodak, Rochester, NY) and exposed to autoradiography for 3 weel

Study in male rats Animals and treatment: Male Sprague-Dawley rats (Crl=CD (SD) Br) weighing 85 to 100 g were used. Treatment with EM-250 (0.1, 0.3, 1.0, 3.0 mg, respectively) was initiated on the afternwn of the day preceding orchiectomy. The compound was administered twice daily for 6 days by subcutaneous injection in 5% ethanol-1.0% gelatin in 0.9 % NaCl (vehide) to eight animals per group. On the day of orchiectorny, hvo silastic implants of ~4-dionewere inserted subcutaneously in the dorsal area of each animal of the appropriate groups. The implants measured 0.3 cm in length, 0.155 cm inner diameter, and 0.313 cm outer diameter. The last injection of EM-250 was administered on the morning of day 6, two hours after injection, the rats were killed by decapitation, and the blood was collected for serum lutenizing hormone (LH) measurement. Ventral prostates were rapidly removed, dissected from adherent tissue, weighed, and fixed for in situ hybridization.

Measurement of prostate-binding proteinS3 (PBPC3) mRNA levels: ln situ hybridization of prostatic sections with PBPS3 cDNA probe was done as desaibed previously [371. In brief, the ventral prostates were fixed in 4% paraformaldehyde phosphate buffer for 48 h at 40C and subsequently waked in 0.1 M phosphate buffer containing 15% sucrose. Thereaher, the tissues wert rapidly frozen in isopentane-solid Ceand kept at -80C MuItiple (8 or 10) 10 )im-thick tissue sections from each pros tate were mounted on gelatin-coatec glass slides. Prehybridization buffer contained 50% formamide, 5 X SSPE (1 > SSFE = 0.18 M NaCl, 10 mM NaH2PO4, and 1 mM EDTA, pH 7.4), 0.1% sodiun dodecyl sulfate, 0.1% bovine serum albumin, 0.1% Ficoll, 0.1Y polyvinylpyrrolidone, 0.2 mg/d yeast tRNA, 0.2 mg/d denatured salm01 testis DNA, and 2 wg/ml poly(A). The siides were hybridized in thi prehybridization buffer containing, additionally, 4% dextran sulfate and saturint concentrations (1.0 to 1.5 X10' cpm/ml) of PBPÇ3 3%-labeled cDNA probe2 overnight at 3W. The sections were then washed in a series of SSC dehydratec and exposed for autoradiography. To determine the amount of nonspeàfic background hybridization, prostatic sections from each treatment group wen treated with pancreatic ribonuclease-A (20 &ml) for 45 min at 37°C beforc prehybridization. The following data were obtained: For each prostatic tissuc section, 20 randomly selected areas measuring 0.25 mm2 (excluding auna1 lumen) were analyzed using an Image Re-search Analysis System (Amersham, Arlingon Heights, IL), and the mean optical density value for each section waz calculated.

Measurement of serum LH concentrations: Serum LH was measured by double-antibody RIA using rat hormones (LH-1-6 for iodination, LH-RF-2 ar standard) and rabbit antisera anti-rLH-S-8. The products were generously supplied by the National Pituitary fmgrani (Baltimore, MD).

e rab Animals and tre atrnent: Female Sprague-Dawle y rats weighing 85-100 g were ob tained from Charles River Inc. (St-Constant, Québec, Canada). The animals were ovariectomized under general anesthesia via bilateral flank incisions and randomly assigned to a group. The performance of treatment with EM-250 was the same as described above for male rats. On the morning of the 6th day, the rats were killed and uieri removed and weighed.

Statistical significance was measured according to the multiple-range test of Duncan-Kramer [38]. Data are expressed as mean f SEM. RESULTS

Study in vitro As iiiustrated in Fig. 1, EM-250 inhibited the stimuiatory effect of DHT of the growth of the androgen-sensitive Shionogi ceiis in vifro at concentration! comparable to those of OH-flutamide, the active metabolite of flutamide. In fact the Ki value of EM-250 action was calculated at 96 nM; that of OH-flutamide i! 64 nM (nonsignificant, NS). Moreover, neither EM-250 nor OH-flutamide had i signtficant stimuiatory effect on cancer ceU growth, indicating an absence O: significant androgenic activity of the two compounds.

As shown in Table 1, we assayed the activities of major enzymes for steroic synthesis pathway namely, 38-HSD, 17p-HSD, Sa-reductase aromatase anc sulfatase. EM-250 inhibited these enzymes by O to 10% up to the highesa concentration used, which was 10-6 M of inhibitor; the only exception wai aromatase activity.

Studv in male hamsters As illustrated in Fig. 2, the surface area of the flank organ in control mal6 hamsters inaeased by 18 to 20 mm2 during the 4-week study. Administration oi the antiandrogen EM-250 at a dose of 30 kg twice daily for 4 weeks on the righi flank organ inhibited growth by 38.68% (p < 0.01) of the surface area withoui affecting the size of the flank organ on the left control side, where vehicle onlj was applied according to the same schedule. When the EM-250-dose wa! inaeased to 100 pg and 300 pg, a progressive inhibition, still limited to the righi side, was measured at 46.93% @ < 0.01) and 63.77 (p < 0.01), respectively. Th( local inhibitory effect of EM-250 is clearly illustrated in Fig. 3, which shows thai the inhibition described in Fig. 2 was limited to the right flank organ. As showr in Fig. 4, the sebaceous gland area in treated flank organs was also significantlq reduced by 20.01% (p c 0.05), 31.65% (p c 0.01), and 46.28% (p < 0.01) by all three dosage levels (30,100 and 300 pg twice daily), respectively. Fig. 5 illustrates thai tieatrnent with increasing doses of EM-250 progressively deaeased the size ol sebaceous glands. The inh'bitory effect is limited on the right side only.

As ülustrated in Fig. 6, the potent local activity of EM-250 was also reflected in the inhibition of Pa-thymidine incorporation measured in the ceiis of th€ sebaceous glands of the right flank organ, which was the site application of th antiandrogen EM-250. At the 30 pg dose administered twice daily, EM-251 inhibited Pa-thymidine incorporation by 60.4% (p 0.01), whereas inhibition of 63.6% (p < 0.01) and 65.7% (p c 0.01) were measured at the 100 pg and 300 pl doses. As a more visual illustration of the inhibitory effect of EM-250. Unde microscope, EM-250 progressively decreased the number of labeled cells and th number of silver grains of each celi, thus Uustrating decreased incorporation a radioactivity in the sebaceous glands of the flmk organs of hamsters injectei with pH]-thymidine.

The size of the sebaceous glands of the ears has been suggested as a gooc mode1 for the assay of antiandrogens [Il], therefore, we also examined the effec of topical application of EM-250 on the right ear of intact male hamsters. A illustrated in Fig. 7, an approximate 45.3% inhibition (p < 0.02) of ear sebaceou gland size was achieved with the 3bgdose of EM-250 and inhibitions of 51.59 (p c 0.01) and 82.1 (p c 0.01) were obtained with the 10Wg and 300 pg doses. Th marked inhibitory effects of EM-250 limited to the right ears are clearl: illustrated in Fig. 8, whi& shows the dramatic reduction in sebaceous gland sizl in the right ears, the sebaceous glands of the left ears remained normal. On th other hand, topical administration of EM-250 at doses of 30, 100, 300 pg, twicl daily for 4 weeks, did not inhibit the growth of prostates and seminal vesicle and had no effect on the serum testo and DHT concentrations (Data not shown),

dv in male rats The weight of the ventral prostate is well recognized as a specific anc sensitive indicator of androgenic statu in the rat. As illustrated in Fig. 9, th1 Wice daily subcutaneous injection of the antiandrogen EM-250 at doses rangin1 from 0.1 to 3.0 mg had no effect on androstenedione-stimulated ventral prostab weight. These data indicate the absence of systemic antiandrogenic activity o EM-250 up to highest dose used (3.0 mg).

The rat ventral prostate contains highly sensitive and specific androgen dependent proteins: (a) prostate binding protein (PBP) which has threc components; Cl .C2.C3, and @) spermine-binding proteins (PSP). Androgen: regulate these proteins' synthesis via changes in the concentrations of thi mRNAs encoding these proteins. As illustrated in Fig. IO, PBP€3 mRNA leveli deaeased to 6% of intact conkol in castrated rats using in situ hybridization. i physiologicai dose of adrenal precursor a4-dione completely reversed the effec of castration. Furthermore, injection of inaeasing doses of EM-250 had nr significant effect on this panmeter in castrated rats implanted with ~Ldione(O.: cm). Under the same experimental conditions, serum LH was not inhibited indicating the absence of androgenic or estrogenic activity of EM-250, at least a the doses used (Fig. 11).

The treatment of ovariectomized rats for 6 days with the antiandrogen EM 250 at doses up to 300 pg, twice daily, had no effect on uterine weight, indicatiq the lack .of an estrogenic effed of EM-250 (data not shown). DISCUSSION

It is well known that skin disorders such as acne, hirsutism, femalt androgenic alopecia, and male pattern baldness are considerd androgen-reiated problems. Most, if not dl, critical factors involved are probably reiated to hi@ overaii androgen production, increased availability of free andr~genbecause oj deficiency in sex hormone-binding globulin, and an amplineci target respom mediated either thtough Sa-reductase of testo or the capaaty of the intracdulai receptor to bind the hormone in target tissue and not to the. circulating steroid hormone level [3941]. It is reasonable that treatment with antiandrogen oi androgen antagonist can improve these conditions. The present data demonstrate that EM-250 inhibited the androgen-stimulated Shionogj carcinoma cells with Km comparable to antiandrogen flutamide but did no! have inhibitory effects on the enzymes of the steroid synthesis pathway in vitrc gene transfected cells, thus clearly indicating its property as a pure antiandrogen.

Antiandrogens that have been used in acne therapy exert their effect by inhibithg the binding of androgen to the androgen receptor. Cyproterone acetate and spironolactone are two such agents, both of which decrease sebuni excretion rates [42,43] and improve clinical acne (10, 42, 44, 451 when administered systematically. These agents have also been used in treating hirsutism [46]. Flutamide, a nonsteroidal antiandrogen that also inhibits androgen binding to androgen receptor, has recently been reported to improve seborrhea, acne, and hair loss in women who took the medication orally in combination with birth control pills [9].

To date, results with topical antiandrogen therapy for acne have &en disappointing. Topical cyproterone, spironolactone, and flutamide can produce inhibitory effects on sebaceous tissue in anirnals models (15,47491, but in doing, topical cyproterone and flutamide have also been shown to exert a systernic effect [15, 481. Topical spironolactone does not suppress sebum production in humans [50] Several uncontrolled studies have reported topical spironolactone to benefit clinical acne [SI, 521, but no controiied studies have confirmed these observations.

in our present study, topical administration of EM-250 on the right fhnk organs and right ears significantly decreased the pigmentation area of treated flank organ, the size of underlying sebaceous glands, and the size of sebaceou! glands of treated ears in a dose-dependent rnanner without affecting th untreated sides. The inhibitory effect of [3HJ-thymidine incorporation on th6 treated flank organ indicates that the mechanism of effects of EM-250 or sebaceous glands involved inhibition of cell proliferation. Furthesmore, EMJ 250 did not effect the serum testosterone and DHï levels or other androgenic parameters, namely, prostatic and seminal vesicle weights, following 4 weeks O: topical administration.

The absence of systernic effects after topical administration of EM-250 ma) be due to following reasons: the drug could be rapidly converted into iner metabolites locally; it could be bound locally to the site of application and unablt to circulate; it rnay not have been absorbed in amount suffiaent to product effects at distant sites; drug may have been metabolized to other compound! dwing the circulation; or the drug could be picked up by circulatting molecule! that render it systematically ineffective. The latter two reasons may explain ow results in the present study, in which even systemic administration of EM-25( did not cause any antiandrogenic or antigonads effects. in fact, injection intc castrated rats of this compound up to 3 mg twice daily for 6 daYs,supplementec with a4-dione, did not change the weight of the prostates, and did not affect th androgen-regulated PBP-C3 mRNA level and senun LH level.

In the sebaceous glands of both rodents and humans, testo is converted intc DHT by the enzyme Sa-reductase [39, 53-58]. It then reacts with a cytoplasmic androgen receptor that has a higher affinity for DHT than for testosterone [59, 60). This receptor has been identified in the sebaceous glands of the hamster [dl, 621 and in human skin [59, 60, 631. However, antiandrogenicity in the hamste~ would not necessarily imply equal efficacy in humans. Also, the hamster flank organs consist of three different androgen dependent components: thr sebaceous glands, hair follides, and dermal pigment cells, The antiandrogenic effect on one skin structure, such as the sebaceous gland, is not necessarily equivalent to the effect on another, such as the hait follicle. Thus, a topical antiandrogen that may be effective in acne therapy rnay not work in treatlig hirsutism or balding.

Nonetheless, the present study clearly demonstrates that EM-250is a poteni topical pure antiandrogen without systemic antiandrogenic, androgenic and estrogenic effects and should encourage additional animal studies and result i carefuiiy controiied clinicd trails applied to disorders of end organ response 1 androgen in human. Darley, C.R, Kirby, J.D., Besser, G.M., Munro, D.D., Edwatds, C.R.W. am Rees, L.H.: Circulating testosterone, sex hormone binding giobulin am prolactin in women with the late onset or persistent acne vulgaris. Br. Dermatol. 106 (1982) 517-522.

Freinkel, RK.: Medical intelligence. Current Concepts, pathogenesis a acne vulgaris. New Engl. J. Md. 280 (1969) 1161-1163.

Cunliffe, W.J. and Shuster, S.: The rate of sebum seaetion in man. Th Uncet 1 (1969) 685-687.

Ebling, F.J.: The effects of cyproterone acetate and oestradiol upoi testosterone stimulated sebaceous activity in the rat. Acta Endocrinol (Copenh) 72 (1973) 361-365.

Pochi, P.E. and Strauss, J.S.: Endocrinologie control of the development am activity of the human sebaceous gland. J. Invest. Dermatol. 62 (1974) 191 201.

Strauss, J.S., Pochi, P.E. and Downing, D.T.: The sebaceous glands: twenty five years of progress. J. Invest. Dennatol. 67 (1976) 90-97.

Bloch, B.: Metabolism, endocrine glands and skin diseases, with specia reference in acne vulgaris and xanthoma. Br. J. Dennatol. 43 (1931) 61.

Labrie, F.: Intracrinology. Mol. Cd.Endocrinol. 78 (1991) C113C118.

Cusan, L., Dupont, A., Bélanger, A., Tremblay, RR, Manhès, G. and Labrie F.: Treatment of hirsutism with the pure antiandrogen flutarnide. J. Am Acad. Dermatol. 23 (1990) 462-469.

GcmdfeUow, A., Alaghband-Zadeh, J., Carter, G., Crearn, J.J., Holland, S. J.Scully and Wise, P.: Oral spironolactone improves acne vulgaris an( reduces seburn exaetion. Br. 1. Dermafol. 111 (1984) 209-214. 11. Matias, J.R. and Orentreich, N.: The hamster ear sebaceous glands. 1. Examination of the regional variation bp stripped skin planimetry. J. Invest. Detnaafol.81 (1983) 43-46.

12. Wirikier, K. and Schaefer, H.: Das Verhalten der Talgsekretion wahrend der Behadung der Acne mit Cyproteronacetat und Athinylostradiol. Archiv fur Dermatologische Forschung 247 (5973) 259-264.

13. Burton, J.L., Laschet, U. and Shuster, S.: Reduction of sebum excietion in man by the antiandrogen, cyproterone acetate. Br. J. Detrnatol. 89 (1973) 487- 490.

14. Zarate, A., Mahesh, V.B. and Greenblatt, RB.: Effect of an antiandrogen 17a-methyl-B-nortestosterone,on acne and hirsutism. J. Clin. Endwiml. Metab. 26 (1966) 1394-1398.

15. Lutsky, B.N., Budak, M., Koziol, P., Monahan, M. and Neri, RO.: The effects of a nonsteroid antiandrogen, Flutamide, on sebaceous gland. 1. Invest. Dermatof. 64 (1975) 412-417.

16. Neri, R, Monahan, M.D., Meyer, J.G.,Afom, B.A. and Tabachnick, I.A.: Biological studies of an antiandrogen (SH-714). Eur. 1. Pharmacol. 1 (1967) 438444.

17. Neri, R, Horance, K., Koziol, P. and Cleave, S.V.: A biological profile of nonsteroidal antiandrogen, SCH13521 (4'-nitro-3'. trifluoromethylisobutyranilide). Endoctinology 91 (1972) 427-437.

18. Poyet, P. and Labrie, F.: Cornparison of the antiandrogenic/androgenic activities of flutamide, cyproterone acetate and megestrol acetate. Mol. Cell. Endocrinol. 42 (1985) 283-288.

19. Labrie, C., Cusan, L, Plante, M., Lapointe, S. and Labrie, F.: Analysis of the androgenic activity of synthetic "progestins" currently used for the treatment of prostate cancer. J. Stemid BiUcb. 28 (1987) 379384.

20. Luthy, L, Mgin, D. and Labrie, F.: Androgenic activity of synthetic progestins and spironolactone in andragen-sensitive mouse mammary carcinoma (Shionogi) cells in culture. J. Sferoid Biochem. 31 (1988) 845-852. Labrie, C., Simard, J., Zhao, H.F., Pelletier, G. and Labrie, F.: Synthetil progestins stimulate prostatic bhding protein messenger RNAs in the ra ventrai prostate. Mol. Cell. Endacrinol. 68 (1990) 169-179.

Pria, S.D.,Greenblatt, RB. and Mahesh, V.B.: An antiandrogen in acne anr idiopathic hirsutism. J. Invest. Dennatol. 52 (1969) 348-350.

Cunliffe, W.J., Shuster, S. and Cassels-Smith, A.J.: The effect of topica cyproterone acetate on sebum secretion in patients with acne. Britisi loumr of D-~OZO~~81 (1969) 2001201.

Cunliffe, W.J. and Simpson, N.B.: Homonal control of sebaceous gland' with special reference to antiandrogens topically applied. in Percutaneou' Absorption of Steroids (Edited by P. Mauvais-Jarvis). Acadernic press, Nev York (1980) 149-154.

Burdick, K.H. and Hill, R.: The topical effect of the antiandrogei chlormadinone acetate and some of its clinical modifications on thc hamster costovertebral organ. Br. J. Dermatol. 82 (Suppl. 6) (1970) 19-25.

Labrie, F. and Veilleux, R.: Maintenance of androgen responsiveness b! glucocorticoids in Shionogi mammary caranorna ceUs in culture. J. Natl Cancer Inst. 80 (1988) 966-970.

Labrie, F., Veilieux, R. and Fournier, A.: Low androgen leveb induce th development of androgen-hypersensitive cells clones in Shionogi mousc mamrnary carcinoma cells in culture. I. Natl. Cancer Inst. 80 (1988) 1138 1147,

Simard, J., Dauvois, S., Haagensen, D.E., Lévesque, C., Merand, Y. anc Labrie, F.: Regdation of progesterone-binding breast cyst protein GCDFP-2r secretion by estrageris and androgens in human breast cancer ceils: a new marker of steroid action in breast cancer. Endocrinology 126 (1990) 3223q 3231.

Cheng, Y.C. and Prusoff, W.H.: Relationship between the inhibitior constant (Ki) and the concentration of inhibitor which causes 50 per cm inhibition (ICSO) on an enzymatic reaction. Biochem. Phannocol. 22 (1973 3099-3108.

30. Rhéaume, E., Lachance, Y., Zhao, H.F., Breton, N., de Launoit, Y., Trudel C., Luu-The, V., Simard, J. and Labrie, F.: Structure and expression of a nev cDNA encoding the almost exclusive 3p - h y d r ox y s te r O i c dehydrogenase/dS-A* homerase in human adrenals and gonads. Mol Endoctinol. 5 (1991) 1147-1157.

31. Lachance, Y., Luu-The, V., Labrie, C,, Simard, J., Dumont, M., Launoit, Yd Guérin, S., Leblanc, G. and Labrie, F.: Characterization of human 3b hydroxysteroid dehydrogenase/A 5-A 4 isomerase gene and its expression U mamrnalian cells. 1. Biol. Chem. 265 (1990) 20469-20475.

32. Labrie, F., Sugirnoto, Y., Luu-The, V., Simard, J., Lachance, Y., Bachvarov D., Leblanc, G., hocher, F. and Paquet, N.: Structure of human type II Sa reductase. Endocrinohgy 131 (1992) 1571-1573.

33. Luu-The, V., Sugimoto, Y., Puy, L., Labrie, Y., Lopez, I., Singh, M. anc Labrie, F.: Characterization, expression and immunohistochemica localization of Sa-reductase in human skin. J, Invest. Demafol. 102 (1994 221-226.

34. Luu-The, V., Labrie, C., Zhao, H.F., Couet, J., Lachance, Y., Simard, J., Côte J., Leblanc, J., Lagacé, L., Bérubé, D., Gagné, R. and Labrie, F.: Purification cloning complementary DNA structure and predicted amino acid sequencc of human estradiol 17P-dehydrogenase. Ann. N.Y. Acad. Sci. 595 (1990) 40 52.

35. Wing, L.Y., Garrett, W.M. and Brodie, A.M.H.: Effect of aromatasc inhibitors, aminoglutethimide, and 4-hydroxyandrostenedione on cydic rats and rats with 7,12-dimethylbenz(a)anthracene-induced mammarj tumots. Cancer RLS. 45 (1985) 2425-2428.

36. üélanger, A., Caron, S. and Picard, V.: Simultaneous radioimmunoassay O progestins, androgens, and estrogens in adult rat testis. 1. Sferoid Biochrtn 13 (1980) 1s190. Labrie, C., Simard, J., Zhao, H.F.,Béranger, A, Pelletier, G. and Labrie, F.: Stimulation of androgen-dependent gene expression by the adrenal precursors dehydroepiandrosterone and androstenedione in the rat ventral prostate. Endocrinology 124 (1989) 2745-2754.

Kramer, C.Y.: Extension of multiple range tests to group means with unique numbers of replications. Biometrics 12 (1956) 307-310.

Sansone, G. and Reisner, R: Differential rates of conversion of testosterone to dihydrotestosterone in acne and in normal skin: A possible pathogenic factor in acne. 1. Invest. Dermatol. 56 (1971) 366-372.

Ehrmann, D.A. and Rosenfield, R.: Clinical review 10. An endocrinologie approach to the patient with hirsutism. J. Clin. Endocfinal. Metab. 71 (1990) 1-4.

Serafini, P. and Loba, R.A.: Increased Sa-reductase activity in idiopathic hirsutism. Ferfil. Sferil. 43 (1985) 74-78.

Miller, J.A., Wojnarowska, F.T. and Dowd, P.M.:Antiandrogen treatment in women with acne: a controiied tract. Br. J. Demtol. 114 (1986) 705-716.

Keahey, N., Martinez, L., Blasco, K.J. and McGinley Leyden, J.J.: Suppression of sebum exaetion fdiowing treatment of acne and hirsutism with spironolactone. J. Invest. Dennatol. 80 (1983) 359 (Abst.).

Muhlerna~,M.F., Carter, G.D., Gream, J.J. and Wise, P.: OraI spironolactone: an effective treatment for acne vulgaris in women. Br. 1. D-tol. 115 (1986) 227-232.

Greenwood, R, Bruminitt, L., Burke, B. and Cunliffe, W.J.: Acne: double- bihd clinical and laboratory trail of tetracydine, oestrogen, cyproterone acetate, and combined treatment. Br. Med. 1.291 (1985) 1231-1235.

46. Lunde, 0. and Djseiand, 'O.: A comparative study of aldactone and dione in the treatment of hirsutism. 1. Steroid Biochem. 28 (1987) 161-165. Ebiing, F.J., Randall, V.A. and Skinner, J.: Local suppression of sebum secretion in rats by topical cyproterone acetate in ethanol. j. Inwst. Demtatol. 77 (1981) 458463.

Weissma~,A., Bowden, J., Frank, B.L., Horwitz, S.N. and Frost, P.: Antiandrogenic effects of topically applied spironolactone on the hamster fiank organ. Arch. Dermafol. 121 (1985) 57-62.

Luderschmidt, C., Bidlingmaier, F. and Plewig, G.: Inhibition of sebaceous gland activity of spironolactone in syrian hamster. 1. Invest. Dett~tol.78 (1982) 253-255.

Walton, S., Cuniiffe, W.J., Lookingbill, O.P. and Keezkes, K.: Lack of effect of topical spironolactone on sebum exaetion. Br. 1. Dermafol. 114 (1986) 261-269.

Nielsen, P.G.: Treatment of female acne vulgaris with a cream containing the antiandrogen camenone. Demtologica 166 (1983) 275-276.

Messina, M., Manieri, C., Rizzi, G., Gentile, L. and Milani, P.: Treating acne with antiandrogens: the confirmation of the validity of a percutaneous treatment with spironolactone. Curr. Ther. Res. 38 (1985) 264-282.

Takayasu, S. and Adachi, K.: The conversions of testosterone to 17P- hydroxy-Sa-androstan-3sne(dihydrotestosterone) by human hair foiiicles. J. Clin. EndocTinol. Mefab. 34 (1972) 1098-1101.

Price, V.H.: Testosterone metabolism in the skin: A review of its function in androgenic alopeda, acne vulgaris, and idiopathic hirsutism including recent studies with antiandrogens. Arch. Dermafol. 111 (1975) 14%-1502.

Thomas, J.P. and Oake, R.J.: Androgen metabolism in the skin of hirsute women. 1. Clin. Endmrinol. Metab. 38 (1974) 19-22.

Schweikert, H.U. and Wilson, J.D.:Regdation of human hair growth by steroid hormones: 1. Tstosterone metabolism in isolated hairs. 1. Clin. Endominol. Metab. 38 (1974) 811-819. 57. Takayasu, S., Wakioioto, H., Itami, S. and et al: Activity of testosterone Sa reductase in various tissues of human skin. J. Invest. Dennofol. 74 (1980 187-191.

58. Gomez, E.C1 Lleweliyn, A. and Frost, P.: Metabolism of testosterone414( by hamster skin and flank organ. 1. In-t. Rmtol. 63 (1974) 383-387.

59. Svensson, J. and Snochowski, M.: Androgen receptor leveis in preputia skin from boys with hypaspadias. I. Clin. Enducrinol. Mefab. 49 (1979) 34 345.

60. Mowszowicz, I., Riahi, M. and Wright, F.: Androgen receptors in humai skin cytosol. J. Clin. Endocrinol. Metab. 52 (1981) 338-344,

61. Adachi, K.: Receptor proteins for androgen in hamster sebaceous gland2 Journal of Investigative Dermatology 62 (1974) 217-223.

62. Adachi, K. and Kano, M.: The role of receptor proteins in controllin, androgen actions in the sebaceous glands of hamsters. Steroids 19 (1972 567-574.

63. Fichman, L.F., Nyberg, L.M.,Bujnovsky, P. and et al: The ontogeny of th androgen receptor in human foreskin. J. Clin. Endocrinol. Metab. 52 (1981 919423. LEGENDS TO FIGURES

Figure 1. Effect of inaeasing concentration of EM-250 and OH-Flutamide oi dihydrotesteroneatimulated cell proiiferation in the androgen-sensitive SEM 107 done of Shionogi mouse rnammary carcinoma cells. The data are expresse4 as means f SEM in triplkate dishes.

Figure 2. Effed of bpicai administration for 4 week of EM-250 on the right flad organs and ears on the size of flank organs in intact hamsters. Both flank mgan of control animals as well as the le& side fiank organ of anirnals of the othe groups received the vehide only (5096 ethanol and 50% propylene glycol) wM the right side flank organ received the indicôted doses of EM-250. Data ari expressed as means f SEM. **p<0.01, intact animals vs al1 the other groups.

Figure 3. Photograph of the flank organ area of hamsters treated as desuibed ii Fig. 2 with increasing doses of the antiandrogen EM-250 applied topically on th right fiank organs and ears.

Figure 4. Effect of topical administration for 4 week of EM-250 on the right flanl organs and ears on the size of underlying sebaceous glands in intact hamste flank organ. The treatment is as desaibed in Fig. 2. Data are expressed as mean f SEM. **p<0.01, intact animais vs ail the other groups.

Figure 5. Photograph of the underlying sebaceous glands of the hamster flanl organs treated as described in Fig. 2 with increasing doses of the antiandrogei EM-250 applied topically on the right flank organs and ears. The photos wer( directly took from slides uing a camera. L: lefi ear. R: right ear.

Figure 6. Effect of topical administration for 4 week of EM-250 on PH] thymidine incorporation in flank organs in intact hamsters. The treatment is a described in Fig. 2. Data are expressed as means I SEM whde the units of confira gtoup were set at IWO. *P<0.01, intact animals vs all the other groups.

Figure 7. Effect of topical administration for 4 weeù of EM-2% on the right fiani organs and eus on the size of ear sebaceous glands in intact hamsters. 'RN treatment is as described in Fig. 2. Data are expressed as means f SEM while thi units of control group were set at 100%.. **p<0.01, intact animais vs al1 th other groups.

Figure 8. Light microscopie photograph of the effect of EM-250 on the sebaceou glands of the ears in male hamsters treated as described in Fig. 2. (~200).L: Id ear. R: right eu.

Figure 9. Effect of twice daiïy treatment for 6 days with increasing doses of EM 250 on ventral prostate weight in the'orchiectomized male rats simultaneousl treated with an impht (0.3 cm length) of andmtenedione (~4-dione).Data ar expressed as means f SEM.

Figure 10. Effect of twice daiiy treatment for 6 days with increasing doses of EM 250 on ventral prostatic binding protein (PBPwC3) in the orchiectomized mal rats simultaneously treated with an implant (0.3 an length) of androstenedion (~4-&one).Data anexpressed as means f SEM.

Figure 11. Effect of twice daily treatment for 6 days with increasing doses of EM 250 on serum LH concentration in the orchiectomized male rats simultaneousl treated with an implant (0.3 cm length) of androstenedione (~4-dione).Data ar expressed as means f SEM. Table 1

In vitm inhibitory effecb of EM-250 on the 3&HSD, 17MD, Sa-teductase, aromatise, sulfatase activities

17B-HSD Sa-Reduc Aroma- Sulfatase Type1 Type Il tase tase

( % inhibition) I I 1 I t FIGURE 1

SHlONOGl MOUSE MAMMARY CARCINOMA CELLS

5- r cn P 4- -u Z u¶ 3- x

IDu Contrai O OH-FLU + 0.3nM DHT (Kk64.15 nM) EM-250+ 0.3 nM DHT (KI = 95 8 nM) O- 1 I 1 I 1 1

ANTIANDROGEN CONC. (Log M)

FIGURE 3

LW RIGHT FIGURE 4 FIGURE 5

INTACT CONTROL 1 FIGURE 7

CONTROL 1 30 1 FIGURE 8 FIGURE 9

INTACT - FIGURE 10

IN1ACT ; INTACT CONT CONT 1 0.1 1 0.3 1 1.O 1 3.0 QRCH EM-250 (mg, twice daily) ORCH t 4-DIONE Activity of 17 8-(N-AlkyUary1formunido)-and 17~4(N-AlkyUuyl) ;ilkyVarylamido]4methy14-~a-~x015a-~1dmstan-~nesas Sa-Reductase Inhibitor in ~amsterHu* Ogan and Eu

CaihCHEN, Xun LI, Shankar M. SlNGH and Femand LABRIE

MRC Group in Molecuiar Endoctinology, CHUL Research Center and Laval University, Québec, Gl V 4G2, Canada Skin disorders, such as am, female hirsutism and male pattern balcine! are associated with local androgenic activity. As the most potent androge dihydrotestosterone (DHT) is fomed fmm testosterone (testo) by the catalysis ( Sa-reductase, thus, it is clear that the inhibition of Sa-reductase is one of tk logical sites to block the androgenic actions in peripherai tissues in the treatme~ of such androgen dependent diseases. In present study, we have investigate the inhibitory effect of a series of 17 B-(N-atkyl/aiylformamido)- and 17&[(P Allcyl/~yl)alky/arylamido]4methyl4a2â-3~~0-5a-androstan-3-0ne derivatives as Sa-reductase inhibitors, following their topical administration a the size of flank organs, the size of underlying sebaceous glands and t) enzymatic activity of Sa-reductase in the flank organs and ears of Golden Syria hamster. The prelirninary study show that 17B-(N-amy1formamido)-4-methyl-! aza3a-androstan-3-one (EM-401), 1713~(N-hexyUormamido)-4-methyl-4-azadc androstan-3-one (EM-402), and 17f?-(N-heptylformamido)-4-rnethyl-4-aza-S~ androstan-3-one (EM-540) are the three most potent compounds studiec Further conduction has demonstrated that EM-402 deaeases the size of the ri@ flank organs (treated side) by 22.17%, 31.14%, and 32.14% (pc0.01 for all) aftf topical application of inaeasing doses of 30, 100, and 300 pg, respectively, twic daily for 4 weeks. EM-402 also reduces the underlying sebaceous gland size c right flank organs to 38.4%, 42.049, and 59.2% of intact control with increasin doses, respectively. Siffülar result was observed on the size of sebaceous gland of the ears. In addition, we have observed a dose dependent inhibition of 50 reductase of 46.94% to 79.59% (p<0.01) on the right flank organs and a inhibition of 46.25% to 80.27% (p<0.01) on the right ears using topical EM-40: EM-402 has no sigxuficant effects on the left contralateral side of flank organs c ears in al1 of the parameters studied. In addition, EM-402 has no effects on th prostatic and seminal vesicle weights while EM-401 and EM-540 show mm systemic effects. Al1 the experiments clearly illustrate that EM-402 exerts a poter topical antiandrogenic effect without any systemic action in the hamster at th doses used. The interaction between androgenic hormones and the skin has been i fascinating area of research that has tantalized endocrinologists anc dermatologists for several decades [1,2]. It is well known now that acne hirsutism, and male pattern baldness are al1 androgen-related disorders. Thc patients suffering of acne, may have an average more sebum than norma subjects. Abnormally high levels of sebum production can result from higl overall androgen production or increased availability of free androgen due tc the defiaency of sex hormone-binding globuün (SHBG). Hyperandmgenic statc can also result from an amplified target response mediated either through Sa reduction of testosterone (testo) or the cagacity of the intraceilular receptor tc bind the hormone.

Testo is a main androgen from testicular origin and is converted intc dihydrotestosterone (DHT)in the target ce11 cytoplasm by enzyme of Sa< reductase in peripheral tissues [3]. Dmappears to be a more potent androgex than testo. Thus, antiandrogenic activity may occur thsough severa mechanisms: the conversion of testo ta DHT may be blodced by Sa-reductasi inhibitors; testo or DHT may be prevented from binding by cornpetition for th cytosol or nuclear andsogen receptor by a substance that has no androgenic activity, i.e., an antiandrogen [4-61. Up ta date, a series of steroid or non-steroic compounds have been tested in animals or humans [7-151. Systemicallj administered antiandrogens, such as spironolactone ,cyproterone acetate , 17a methyl-8-nortestosterone, and flutamide improve acne, hirsutism, and femalc androgenic alopecia [7, 9, 10, 161, however, these applications are limited tc women. These antiandrogens can not be used with male to trea' hyperandrogenic disorders because of the potential inhibitory effects in th gonadal system. It therefore, would be of great therapeutic value to deveio~ drugs that have local antiandrogenic effect without causing the abovi mentioned systemic side effects in the male. The easy accessibility of the Syriar hamster flank organ and ear have made them good animal rnodels for the studj of such substances (17-19).

Since the enzyme Sa-reductase plays an important role in the loca: androgen formation in peripheral tissues, research has ben focused on this are4 in the past decade. Two types of human Sa-reductase, chronologicaUy identifie as type 1 120,211 and type II 122,231 isoenzymes have been isolateci fsom huma. prostatic cDNA libraries, and the structures of the two isoenzyme have bee. elucidated. The type 1 isoenzyme is predominately expressed in the ski. [20,21,24] while type II is responsible for male pseudohermaphroâitism and i the main type expressed in the human prostate. Previously, we have reportel that a series of 17 8-(N-alkyl/arylformamido)- and 17L3-[(N alkyl/aryl)alkyl/aryI~do]-4-methyl-4-a had hig. potent inhibitory effect on hurnan type 1 Sa-reductase and less potency O, human type II Sa-reductase in gene transfected cells in vitro 1251. In preser study, we have tested the antiandrogenic effects of these compounds on Syria. hamster flank organ and ear after 4 week topical administration. MATERIALS AND METHODS

Male Syrian hamsters approximately 8- to 10-week and weight 110-120 were purchased from Charles River Laboratoties, (St Constant, Québec) and kel 4-5 per cage in a light (14h light/day) and temperature (22 f 1°C)-controlle environment. The animals were houçed in plastic boxes on sawdust an received hamster chow (Agway Country Food Inc., Syracuse, N'Y) and tap wati ad libitum. The hamster was randomly divided to groups of 4-12 animals. At t.k beginning of the expetiment, the ear was marked for identification purpose an the area of the flank organ was shaved with a electronic anian'al clippers. Tl appropriate doses of certain compounds were dissolved in 50% ethanold04 propylene glycol (v/v) and administered in 10 pl on the right side of the hamstc flank organs as well as ears. Meanwhile, the intact group and the left side of tl flank organs and ears of the treated group received the vehide only. The thrg experiments were perforrned as following:

Experiment 1. Intact hamsters were treated with EM-423,347,401,402,42: 435,336,337,436,424,486 at the dose of 50 pg, twice daily.

Expriment 2. intact hamsters were treated with EM-410,541,497, 494,49; 498,503,580,568,606,567,682 at the dose of 100 pg, twice daily.

Expriment 3. intact hamsters were treated with EW01,402, and 540 at tk increasing doses of 30,100,300 pg twice daily, respectively.

The treatment last twice daily for 28 consecutive days. The length an width of the darkly pigmented oval spot outlining each flank organ wei carefuily measured with a vernire caliper (Fisher Scientific). Flank organ are was then calculated according to the following formula: area of ellipse = z(L/2 W/2) = 3.14 (L x W)/4. The animals were killed by decapitation on the ne] morning of last dose. The flank organs and ears were then removed quickly an fixed for histological examination or frozen in dry ice, and stored at -80°C fc enzymatic assays. The ventral prostate as weii as the semi~lvesicles wei removed, dissecteci, freed from fat and connedive tissue and rapidly weighted. The flank organs and ears were fixed in 4% paraformaldehyde for 48 h anc then rinsed in 15% suaose phosphate buffer for 24 h at 4' C. The flank organ; were then cut in the middle in the mshal to caudal direction while the ears weri cut between the first and second cartilage ridges. The tissues were thel embedded in tissue-tek (Miles hc., Diagnostic Division, Elkhatrt, USA) and CU with a cryostat in eight m-thick sections before staining with hematoxylin eosin. The surface areas of the sebaceous glands of flank organs were estirnatet with cornputer-assisted Image-Pro Plus program (Media Cybernetiics Silerspring, MD, U.S.A.) while the surface areas of the sebaceous glands of thi ears were estimated directly under optical microscopy using a grid. The result, were calculated in wiits and expressed in percentage while the control grou] was taken as 100%.

The enzymatic assay was performed as described previously [26]. In briel 100 fl aliquots of the 1000 x g supernatant of hamster Bank organs and ears weri incubated for 180 min. at 37OC in a total volume of 0.5 ml phosphate buffer (12.i mM KH2PO4, 1 mM EDTA, pH 7.5) containing 0.5 [4-l4C]-testosterone (S.A. 51.4 mCi/mmol) and NADPH was used as a cofactot. Labeled radioactivity wai purchased from New England Nuclear/Dupont (Markham, Canada) anc purified by thin-layer chromatography WC)before use. The enzymatic reactioni were stopped by chilling the incubation mixture in an ice-water slurry anc adding 3 ml of diethyl ether. The components were then mixed and frozen in i dry icesthanol bath. The organic phase was kept while the remaining frozex aqueous fraction was re-extracted once with ether. The organic phase was thex pooled and evaporated to dtyness under a nitrogen stream. Al1 component! were then separated on TLC (60-F29silica gel, E. Merdc, Darmstadt, F.R.G.) usiq the solvent system containing 4:l (v/v) to1uene:acetone before autoradiographj of the plates for 48h. The metabolites revealed by autoradiography werc identified by cornparison with standard steroid characterized by HPLC. The TLC areas corresponding testo and DHT including its further metabolites androstane-3a,l7w01 'and androstane3~,17&diolwere scraped and transferrec into vials containing 0.5 ml ethanol and 10 ml scintillation liquid. Th radioactivity was then measured in a spectrophotometer. Results are expresse1 as meaw f SEM in pmoles or nmdes product formed/mg/protein/min.

Statistical sikiificance was measured according to the multiple-range test c Duncan-Ktamer 127. Data are expressed as means f SEM. RESULTS

Experiment 1. A series d noveI 17 8-(N-alkyl/arylformamido)-and 17& [(Nulkyl/a~l)rlkyl/a~lamido]-4-methyI-4-~-3-0x~5a-androstan-~es(Fig. 1) were tested in hamster flank organ. The treatments were administered on the nght flank organs and right ears at the dose of 50 pg twice daily for 4 weeks while the control and the left side of treated groups were received vehide only. As can be Ken in table 1, decreases of 26.45% @<0.01), 26.39% (p

Experiment 2. A series of 1?~-[(N-alkyl/aryl)alkyl/ary1itrnido]4-methyl4- aza-3-0x0-Sa-androstan-hnes (Fig. 1) were also tested in hamster flank organs. As illustrated in table 2, the topical administration of EM-486, EM-568, EM-606, and EM-567 caused 20.52% (p<0.05), 30.02% (p<0.01), 29.02% (p<0.01), and 27.60% (p<0.05) reductions on the left side and 25.56% ,24.60%, 29.31%, and 28.59% (p<0.01 for all) reductions on the right side flank organ surface area compared with the intact control groupi The prostatic weight was decreased by 17.49% (p~0.05)~27.27% @c0.01), and 7.42% (NS) following the treatment with EM-568, EM-606, and EM-567, respectively while EM-486 did not affect this parameter. Moreover, 4 week treatment with EM-56û and EM-606 at the dose of 100 pg twice daiiy, caused 22.67% and 33.25% (pc0.01 for ail) inhibition on the seminal vesicle weight while EM- and EM-567 had no sigriihcant effect. On the other hand, EM-682 did not si@cantly affect the fiank organ surface area while the ventral prostate and seminal veside weights were reduced by 22.57% and 22.43% (p<0.01), respectively with topid treatment at the 100 pg twice daily dose. As can be seen in same table, the treatment with EM424, EM-497, EM494, EM-498, EM-503, and EM-SSO had no effect on either the surface area of both flank organ the prostate weights, or the se- veside weights while EM-493 had sma effect on the right side of flank orgm (not signifiant).

Experirnent 3. In order to investigate the pote- of best cornpounds fi topical use, we further tested the effects of EM401, EM-402, and EM-540 at tl

inaeasing doses of 30, 100, and 300 pg on the size of flank organ, the size ( sebaceous glands of fiank organs and ears, the Su-reductase activity of flar organ and ear as well as the weights of prostate and seminal veside. P Uustrated in Fig. 2, the surface areas of flank organs were measured as 32.68 1.0 and 33.38 f 1.22 mm2 on the left and right side fiank organ in intact anima respectively. Measurements of 28.11 f 1.10 and 29.25 f 1.72 d,26.04 f 1.39 an 25.03 f 1.52 md,and 25.26 f 2.0 and 21.86 f 0.66 k2were observed for bol flank organs with EM-401 treatment at the doses of 30,100, and 300 pg twice dail

for four week, respectively. On the other fiand, treatment of increasing doses ( EM-402 deaeased the right flank organ size to the values of 25,98 i1.16, 22.78 1.34, and 22.65 f 1.87 mm2 while the size of ieft fiank organs remained the rang of intact control. Similar resdts were observed in the groups treated with Eh 540, in fact, 26.60 f 1.16,25.29 f1.63, and 20.67 f1.02 mm2 were measured in tk right treated flank organ following increasing doses of 30, 100, and 300 pg respectively, while the changes of left side did not reach significance compare with intact control.

Fig. 3 illustrates the changes of the size of underlying sebaceous glands i hamster flank organs which being considered as a more precise parameter tha the changes of the size of flank organ itseif. The treatment with EM-401 reduce the size of the sebaceous glands by 10.70% (NS), 33.33% (p~0.01)~and 36.62' (pc0.01) on the left and 5.60% (NS), 58.0% (p<0.01) and 76.0% (pc0.01) on tk right flank organ. In contrast, the treatment with EM-402 and EM-540 affecte the right treated side only. EM-402 at the dose of 30 pg decreased the rigl sebaceous gland size by 38.4% (pc0.01) while decreases of 42.0% and 59.2' (pc0.01) were observed with the treatment of 100 and 300 pg twice dail! respectively. Moreover, the inhibitoty effect reached to 44.056 (pc0.01) with th smallest dose of 30 pg wMe the iarger doses of 100 and 300 pg resulted in furthr deaeases of 46.4% and 54.8% @<0.01), respectively. It is important to examine the effect of the three compounds in differen androgen dependent tissues. We then further observed the inhibitory effects oi the size of sebaceous glands of ears, since the inner side of hamster ear contain large sebaceous glands which is strictly regulated by androgen. As can be seen ii Fig. 4, Similar result was obtained as the size of sebaceous glands of flank orgaxu in fact, the topicd treatment with EM401 at the dose of 30 pg deaeased the siz of ear sebaceous glands to 57.61 f 6.88% (p4.01) of control while the left side hac no effect. Further deaeases were observed with increasing doses on both sides a ears. 91.4 f 4.35% and 73.14 f 6.23% Qx0.01) of control were measured on the lei side while 36.52 f 3.10% and 31.88 f 3.07% (p4.01)of control were measured oi the right side following 100 and 300 pg twice daily treatment, respectively. EM 402 had a inhibitory effect of 47.97 f 3.61%, 38.94 f 4.79%, and 48.m 5.149 (pc0.01) of control while decreases of 73.8 f 5.92%, 54.09 f 4.68%, and 43.04 : 2.24% were measured with EM-540 on the right side ear sebaceous glands at th doses of 30,100 and 300 pg, respectively. None of the three compounds had effec on the left side ears.

As we have reported previously, the series of 17 P-(N Alkyl/arylformamido)- and 17g-[(N-alkyl/aryl)akyl/arylamido]~a 3-0x0-5a-androstan-3sneshad a very high potent inhibitory effect on humai type 1 Sa-reductase in transfected DU-145 cells [25]. 'We then measured th1 inhibitory effect of EM-401, EM-402, and EM-540 on Sa-reductase activity O hamster flank organs as well as ears. As illustrated by Fig. SA, Sa-reductasc activities of 1.45 f 0.13 and 1.47 f 0.09 pmol DHT formed were measured in boil side flank organs of intact animals. While 1.58 f 0.13 end 0.63 f 0.07 pmols, 1.40 : 0.11 and 0.38 f 0.02 pmols, or 0.82 f 0.1 and 0.22 f 0.03 pmols were obtained 01 the left and right side flank organs after EM-401 treatment at the doses of 30,lW and 300 pg, respectively. The treatment of EM-402 decreased the Sa-reductasi activity to 0.78 f 0.06,0.30 f 0.02, and 0.33 f 0.09 pmols on right side of the flanl organs while left ride remained unchanged. Different with the data describec above with the size of the flank organ and the size of sebaceous gland of flanl organs and ears, EMW had a inhibitory effect on the treated side following 31 pg treatment and on the both side of flank organ after 100 and 300 (ig doses. Ii fa&, 1.59 f0.12 pmol and 1.07 f 0.06 pmol were measured at the dose of 30 pg 1.06 f 0.11 pmd and 0.51 f 0.02 prnol, and 0.94 f 0.09 pmol and 0.52 f 0.09 pmo were obtained at the increasing doses of 100 and 300 pg, respectively. Simila: result was observed in ear Sa-reductase activity. except that EM-540 had a inhibitory effect on the nght side eus only (Fig. SB).

Fig. 6 shows that the ventral prostate and seminal vesicle weights had nc signincant changes foilowing 4 week topical treatment either &th EM-401, EM* 402, or EM-540 at the doses used. DISCUSSION

nie present data demonstrated that some of 17 13-(N-alkyl/arylformarnido: and 17i3-((N-alkyl/aryl)alkyl/~ylamido]~ ones have potent inhibitory effect on the hamster flank organs and ears. Amon them, EM-402 (17B~N-hexylformamido)-4-methyl.9-aza-5) i the most potent in this series for topid effects. The topid administration c EM-402reduceâ the size of flank organs, the size of underlying sebaceous glaa of flank organs, the size of ear sebaceous glands, and inhibited the Sa-reductas activity in Rank organ and ear tissues in the treated side without affecting th contralateral untreated flank organs and ears as well as prostate and semh vesicle growth following the treatment with inaeasing doses of 30, 100 and 30 pg, twice daily for 4 week. in contrast, EM-401 (17G-(N-amylformamido)-4 methy1-4-aza-Sa-androstan-3-one)had lower ICso value (0.91 f 0.24 nM) tha tht of EM-402 (7.25 f 0.82 nM) in human type 1 Sa-reductase gene transfectel ceIl in vitro 1251. The present study also revealed that EM-401 is a more poter inhibitor than EM-402 in Syrian hamster. However, unfortunatel y, its inhibitor effects were observed in both treated and untreated sides after topici administration in a dose dependent manner, which indicates EM-401 exeri systemic effect after topical application in the hamster. Moreover, EM-540 (174 (N-heptylformamido)-4-methyl-4-aza-5a-androst~-3-0ne),on the other hanc reduced the size of flank organ, the sizes of sebaceous glands in flank organs an1 ears in unilateral treated side only, but its inhibitory effect to Sa-reductas activity were observed in both flank organ tissues, which may also hinder il further application as a good topical drug candidate.

AU hamsters, male and female, have a asymmetrical pair of pigmente flank organs (also called costvertebral glands) on their bado that are made up c three androgen-dependent structure: pigment cell, hair follicles, and sebaceou glands [17]. Systemic and local effects of topical antiandrogens can be bot studied by using this model. The presence of paired organs provides a built-i: contro for the study of topical applications to one side. Although it had bee: assumed that the thtee target tissues-pigmentation, hair, and sebaceous gland were equally responsive to androgen, recent studies have shown that each ma: indeed be under separate androgen control (18, 281. Thus, a compound tha inhibits sebaceous glands may mt have the same effect on hair. Earlier studies a the pigmented spot overlying the flank organ as a rneasure of androge response, but we now know that this color change does not necessarily refle( sebaceaus gland or hair foilicle response. It is important to observe the effeet o both tissues and to estimate the inhibitory potency of compounds. Indeet present date demonstrate that EM402 deaeased the size of pigmentation area b 22.17% to 32.14% and the size of underlying sebaceous glands by 42.0% to 59.2' in the Bght of flank organs after increasing dose treatment of 30,100, and 300~ twice daily. It thus suggests that the observation of inhibition of the sebaceou gland development by inhibitors is more specific and precise than the change î the pigmentation spot. It is also interesting to mention that the deaeases c 51.23% to 61.06% were obtained on ear sebaceous gland area following dos dependent treatment. Although no further reduction was observed at the large! dose of 300 pg twice daily treatment, this result may be caused due to th saturation of local androgen inhibition by this compound.

As Sa-reductase inhibitors do not block the binding of testo to its receptor these agents have not been reported to cause sexual problem in human male [2! 301. Previously, Rittmaster and his colleagues have reported that a Sa-reductas inhibitor, N.N-diethyl4-methyl-3-0~014-aza-5a-androstane-17~-carboxamide(4 MA), prevented baldness when topically administered to the stumptail macaqu (311. Oral treatment with N-[l,l~ethyl-ethyI-3-0~0-4-aza-5a-and 17&carboxamide (MK-906,Finasteride), another Sa-reductase inhibitor, teversel the balding process and enhanced the hais re-growth by topical minoXidi1 in th male balding stumptail macaque 1321. However, both compounds exert thei systemic effects by inhibiting prostate growth even after topical application [32 361. It should also be indicated that despite the high efficiency of Sa-reductase a an inhibitor of DHT formation, it seems logical that these compounds shoull not be administered alone in the treatment of androgen-sensitive diseases, du to the well recognized secondary increase in intratissular testo concentratio which is Uely to partially overcome the blockade achieved by the Sa-reductas inhibitor [33,35,36]. In fact, testo itself binds to the androgen receptor with a E value close to that of DHT and eventuaiiy fully activates the androgen recepta (371. The best achievement of inhibition of androgen action is the combinatio: of a Sa-reductase inhibitor and an antiandrogen. An ideal topical compound should exert local antiandrogenic effect on$ Although, the oral administration of systematicaily active antiandrogens such a cyproterone acetate or flutamide couid conceivably be an effective means c treatment, it is not a good choice in practice because of the possibiiity that thes dnigs might reduce libido and impair spermatogenesis in men and feminiz male fetus in pregnant women. The present data clearly demonstrate that Eh 402 is a potent Sa-reductase inhibitor exerting local antiandrogenic effect, whic should encourage the study of detailed mechanism and consider as the suitabl candidate for topical clinical trail. REFERENCES

Pochi, P.E. and Straus, J.S.: Sebum production, casual sebum levels, titratable acidity of sebum, and urinary fractional, 17-ketosteroid excretion in males with acne. 1. Inwsf. Dm~fol.43 (1964) 383-388.

Cunliffe, W.J.and Shuster, S.: Pathogenesis of acne. Uncet i (1969) 68547.

Labsie, F.: kitracrinology. Mol. Cell. Endocrirwl. 78 (1991) C113€118.

Voigt, W. and Hsia, S.L.: The antiandmgen action of 4-andrasten-%one 17a- carboxyI acid and its methyi ester in hamster fiad organ. Endocrinalogy 92 (1973) 1216-1222.

Neumann, F. and Steinback, H.: Androgens and Antiandrogens. Li Berlin Heidelberg (Edited by O. Eichler, A. Farh, H. Herken and A.D. Weleh). Springer-Verlag, New York (1991) 236.

Plewig, G. and Luderschmidt, C.: Effect of antiandrogen on sebaceous glands and acne vulgaris. Animal experiments and studies in acne patients. In Androgenizafion in women (Edited by J. Hammerstein and U. Lachnia- Fixon). Excerpta Medica, Amsterdam Oxford (1991) 96-102.

Lookingbill, D.P., Abrams, B.B., Ellis, C.N.,Jegasothy, B.V., Lucky, A.W., Ortiz-Ferrer, L.C., Savin, RC., Shupack, J.L., Stiller, M.J., Zone, J., Landis, J.R., Ramaswamy, R., Cherill, R.J. and Pochi, P.E.: Inocoterone and Acne. Arch. Dem~fol.128 (1992) 1197-1200.

Weissmam, A., Bowden, J., Frank, B.L., Horwitz, S.N. and Frost, P.: Antiandrogenic effects of topically applied spironolactone on the hamster fiank organ. Arch. Dennafol. 121 (1985) 57-62.

Cusan, L., Dupont, A., Gomez, J.L., Tremblay, R.R. and Labrie, F.: Cornparison of flutamide and spironolactone in the treatment of hirsutism: a randomized controlled trial. Fnfil. Std. 61 (1994) 281-287.

Cusan, L., Dupont, A., Cossette, M. and Labrie, F.: Flutarnide in the tzeatment of female androgenic dopecia. Cm.J. Dermafol. 5 (1993) 421427. Lutsky, B.N., Budak, M., Koziol, P., Monahan, M. and Neri, RO.: l'he effeci of a nhteroid antiandrogen, Flutamide, on sebaceous gland. J. Inan Dennrifol.64 (1975)41247.

La Vecchia, C., Franceschi, S., Decarli, A., Gaiius, G. and Tognoni, G.: Ris factors for endometrial cancer at different ages. [. Nafl. Catrcet Inst. 73 (198i 667-671.

Luderschmidt, C., Eiermann, W., Jawny, J., Bidiingmaier, F. and Ring, ] 17a-propylrnesterolone (SH-434): an antiandrogenic sebosuppressiv substance not influencing circulating testosterone concentration! Experimental studies in Syrian hamsters. Nauynyn Schmiedebergs Arcl Pharmacol. 328 (1984) 214-218.

Brooks, J.R., Primka, R., Berman, C., Krupa, D.A., Reynolds, G.F. an Rasmusson, G.H.:Topical anti-androgeniuty of a new 4-azasteroid in th hamster. Steroids 56 (1991) 428-433.

Bouton, M.M., Lecaque, D., Secchi, J. and Tournemine, C.: Effect of a ne1 topically active antiandrogen (RU38882) on the rat sebaceous glanc cornparison with cyproterone acetate. J. Invest. Demtol. 86 (1986) 163-167.

Zarate, A., Mahesh, V.B. and Greenblatt, R.B.: Effect of an antiandroge 17a-methy1-B-nortestosterone,on acne and hirsutism. J. Clin. Endonino Mefab. 26 (1966) 1394-1398.

Hamilton, J.B. and Montagna, W.: The sebaceous glands of the hamste; Morphologicd effects of androgens on integumentary structure. Am. , Anaf. 86 (1950) 191-233.

Vermorken, A., Goos, C. and Wirtz, P.: Evaluation of the hamster flan organ test for the screening of anti-andmgen Br. 1. Dmnntol. 106 (1982) 95 101.

Plewik G. and Luderschmidt, C.: Hamster ear mode1 for sebaceous glands. Invcst Dematof 68 (1977) 171-176. 20. Andersson, S. and Russel, D.W.: Structural and biochemical properties a cloneci and expressed human and rat steroid Sa-reductases. Proc. Nati Ad.Sn'. U.S.A. 87 (1990) 3640-3644.

21. Harth, G., Azzoh, B., Baginsky, W., Gimis, G., Rasmusson, G., Tolmar R, Raetz, L. and Ellsworth, K: Indentificaton and selective inhibiton of ai isozyme of steroid Sa-reductase in human scalp. Proc. Natl. Acad. Sci U.S.A. 89 (1992) 10787-10791.

22. Andersson, S., Bergman, D.M., Jenkins, E.P. and .Russel, D.W.: Deletion a steroid Su-seductase 2 gene in male pseudophermaphroditism. Nature 351 (1991)159-161.

23. Labrie, F., Sugimoto, Y., Luu-The, V., Simard, J., Lachance, Y., Bachvarov D., Leblanc, G., Durder, F. and Paquet, N.: Structure of human type II Sa reductase. Endocrinology 131 (1992) 1571-1573.

24. Luu-The, V., Sugimoto, Y., Puy, L., Labrie, Y., Lopez, L, Singh, M. anc tabrie, F.: Characterization, expression and immunohistochemica localization of Sa-reductase in human skin. 1. Invest. Dennatoi. 102 (1994 221-226.

25. Li, X., Singh, S. and Librie, F.: Synthesis ans in vitro activity of 1713-(N alkul/arylformamido)-and 178-[(N-alkyl/aryl/arylamido]-4-methyl-4 methyi-4-aza-3-oxo-Sa-androstan-3-onesas inhibitors of human Sa reductases an dantagonists of the androgen receptor. 1. Medicinal. Chem. 31 (1995) 1158-1173.

26. Martel, C., Melner, M.H.,Gagné, D., Simard, J. and Labrie, F.: Widespreac tissue distribution of steroid suifatase, 3B - h y d r O x y s t erO i c de hy drogenase /A=-b4 iromerase (3p-HSD), 17p-HSD 5a-reductase an( riromatase activities in the rhesus monkey. Mol. Cd. Endocnnol. 104 (1994 103-111.

27. Kramer, C.Y.: Extension of multiple range tests to group means witl unique numbers of repüeations. Biomefrics 12 (1956) 307-310. Vermorken, A., Goos, C. and Roelofs, H.: The antiandrogenic effect c progesterone on the hamster flank organ Br. J. Demafol. 102 (1980) 4554

Rittmaster, R.: Topical antiandrogens in the treatment of male pattez baldness. Clin. Dennafol. 6 (1988) 122-128.

Stoner, E.: The dinical development of a Sa-redudase inhibitor, finsiteridl 1. Sferoid Biachem. Mol. Biol. 37 (1990) 375-378.

Rittmaster, R, Vno, H., Povar, M., Mellin, T. and Loriaw, D.: The effects î N,N-diethyl-4-rnethyl-3-0~0-4-aza-5a-and~,a Sc reductase inhibitor and antidrogen, on the development of baldness in tk stumptail macaque. J. Clin. Endocrinof. MetPb. 65 (1987) 188-193.

Diani, A., Mulholland, M., Shull, K., Kubicek, M., Johnson, G., Schostare: H., Brunden, M. and Buhl, A.: Hair growth effects of oral administration c finasteride, a steroid Sa-reductase inhibitor, alone and in combination wit topical minoxidil in the balding stumptail macaque. 1. Clin. Endocrino Metab. 74 (1992) 345-350.

Chen, C., Puy, L., Simard, J., Li, X., Singh, S. and Librie, F.: Local an systemic reduction by topical finasteride or flutamide on hamster flan organ size and enzyme activity. J. Invest. Dennafol. 105 (1995) 678-682.

Imperato-Mcginley, J., Sanchez, R and Spencer, J.: Cornparison of the effeci of the effects of the Sa-reductassse inhibitor finasteride and th antiandrogen flutamide on prostate and genital differentiation: dosc response studies. Endocrinol. 131 (1992) 1141-1156.

Labrie, C., Trudel, C., Li, S., Martel, C., Couet, J. and tabrie, F.: Combinatio of an antiandrogen and a Sa-reductase inhibitor: A further step toward total androgen blockade? Endocrinology 129 (1991) 566-568.

Martel, C., Trudel, C., Couet, J., Labrie, C., Bélanger, A. and Labrie, F Blockade of androstenedione-induced stimulation of androgen-sensitiv parameters in the rat prostate by combination of mutamide and 4-MA. Mo îell. Enducrinol. 91 (1993) 4349. 37. Grino, P., Griffin, J. and Wilson, J.: Testosterone at high concentratioxu interacts with the human androgen receptor similarly ta dihydratestosterone. Endacrino1 126 (1990) 1165-1 172. LEGENDS TO FIGURES

Figure 1. Structure of novel compounds acting as Sa-reductase inhibitors. 1713 (N-alkylformanido)-4-methyl4-aza-Sa-androstan->nes:EM-423. EM-347, Eh4 401, EM-402, EM-540, EM-541, EM424, EM-435, EM-336,EM-337, and EM-43 l7B-[N-alkyl /aryl)atkyl /ar ylamido] 4methyi4aza-5a-mdrostan-36nes:EM-424 EM-497,EM494, EM-493,~~-486, EM498, EM-503,EMSBO, EM-586,W06, Eh4 567, and EM-682.

Figure 2. Effect of treatment with EM-401 or EM-402 or EM-540 on the size a flank organs. The right side flank organ and ear were treated with inaeasinl doses of 30, 100, and 300 pg for 4 weeks while the control and the left side o treated groups received vehicle only. + pc0.05, ++pcO.Ol vs left side of control *pc0.05, **p<0.01 vs right side of control.

Figure 3. Effect of treatment with EM-401 or EM-402 or EM-540 administered a described in legend to Fig. 2 on the size of underlying sebaceous glands of flan! organs. ++pc0.01 vs left side of control. **p

Figure 4. Effect of treatment with EM-401 or EM-402 or EM-540 administered a described in legend to Fig. 2 on the size of ear sebaceous glands. The data ar~ calculated in arbitrary units and expressed in percentage by seAg control grou] as 100%. ++p<0.01 vs left side of control. **p<0.01vs right side of control.

Figure 5. Effect of treatment with EM-401 or EM-402 or EM-540 administered a described in legend to Fig. 2 on Sa-reductase activity in flank organ (panel A and ear (panel 8). ++p

Figure 6. Effect of treatment with EM-401 or W02or EM-540 administered ai described in legend to Fig. 2 on prostatic (panel A) and seminal vesicle weigh (panel 8). Table 1.

Inhibi tory effects of 17B~N-ai~lformanido~methyl4aza-Saandmstan--ne series on the size of fiank organ, ventta1 prostatic, and seminai vcsicle weights. The compounds wen appIied on the tight flank organ

-- - -

Flank orgui surface area Ventral prostatic Seminai vesidi

(mm21 weight .(mg) weight (mg)

Left side right side kControl Data are expressed as means f SEM (n4). *, p < 0.05, ", p.< 0.01 vs intact controls. Table 2

Inhibitory effects of 17~-[~-dky~a~l)dky~~1arnidol4meth~14~-5a- androstan-3sne series on the size of fluùc organ, ventraï prostatic, and seminal vesicle weights. The compounds were applied on the right flank organ

- - EM Flank organ surface area Ventral prostatic Seminal vesid (100pg, BID weight (mg) weight (mg

Left side Right side

33.15 f 1.97 Data are expressed as means f SEM (n=4). *, p c 0.05, **, p < 0.01 vs intact controls. FIGURE 1 FIGUE 2 FIGURE 3 SEBACEOUS GLAND AREA OF €ARS (ARBITRARY UNITS) EAR FtANK ORGAN Sa-REDUCTASE ACTlVlTY Sa-REDUCTASE ACTlVlTY (pmol DHT formed I mg protein 1 min) (pmol DHT formed I mg protein 1 min) P -. A P io P A A N O K O in 0 0 ul O ti O I I 1 1 1 1 1 I r I I SEMINAL VESlCtE WEIGHT (mg) VENTRAL PROSTATIC WEIGHT (mg)

lu P Q, (3D O O O O O J

I D THE CONTROL OF ANDROGEN ACTION IN SHIONOGI TUMORS Of al1 hormone-sensitive cancers, prostate cancer is the most sensitive tc hormonai manipulation, and endocrine therapy is by far the most frequentlj used treatment for this disease. Although, a recent report has demonstrated th benefits of neoadjuvant combination therapy administered before radica prostatectomy or radiotherapy at early stages of the disease, the question of wher the combination therapy shouid be given is still in controversy. In this chapter we have tried b merthese two questions using androgen-sensitive Shionog carcinoma SC115 as a del.In the first article we have demonstrated that th Iarge Shionogi tumors lost their responsiveness to androgen deprivation. and u the second article, we have shown the additive inhibitory effects of th combination treatment with antiandrogen flutamide and Saoreductase inhibitoi finasteride on the Shionogi tumor and prostate growth. The results thus sugges, that the combination therapy is the most efficient therapy available foi advanced prostate cancer and should be given at the earliest stage. LARGE SHIONOGI TUMORS LOOSE THEIR RESPONSrVENESS TU ANDROGEN DEPRïVATiON

Cailin CHEN, Richard POULIN, Femand LABRIE

MRC Group in Molccular Endocrinology, CHUL Research Cenfer and Laval University, 2705 Laurier Boulevard, Québec, GIV 4G2, Canada

Stemid Biochpm Mol Bi01 In press Cancer of the prostate is the most frequent cancer and the second leadinl cause of cancer death in men in North America. The growth of Shionog carcinoma-115 (SC-115) ce& is highly sensitive to androgens, and this ceil line i! a well known and successfd mode1 of prostate cancer. The transplantabli Shionogi carcinoma tumor was used to assess the influence of tumor size 01 the response to androgen deprivation. Two weeks after subcutaneou: inoculation of tumor fragments in Shionogi mice, six groups of animals bearinl SC-115 tumors rrnging from 0.1 to 1.8 cm in diameter were treated witl flutamide (1 mg, twice daily). The castrated rnice received an androstenedionc (A*-dione) implant to mimic the human situation, where the adrenals producc precursor steroids transformed into androgens in peripheral intracrine tissues After 16 days, treatment with flutamide inhibited tumor growth by 32 to 57% i~ the four groups of mice having tumars ranging from 0.1 to 1.0 un in diameter a day 0, whereas no signihcant inhibitory effect was observed in larger tumors The same treatment, however, caused potent inhibitory effects on othei androgen-sensitive parameters, namely prostatic and seminal vesicle weigh and kidney or nithine decarboxylase (ODC) activity, the effect on thesc parameters being similar in all groups of animals, irrespective of tumor size Furthermore, when those larger tumors unresponsive to antiandrogenic treatment were cut into small fragments and inoculated into new groups O; mice, the same treatment with flutamide efficiently inhibited tumor growth treatment being started at tumor sizes of 0.1 to 0.3 cm in diameter. The presen data dearly demonstrate that small himors are highly sensitive to androgen de privation, while loss of response develops with increasing tumor size, thu! indicating that, for optimal efficacy, andtogen blodcade should be given at tht early stages of prostate cancer. Prostate cancer is the most frequent cancer and the second leading cause 01 cancer death in men in North America. in fact, it is predicted that 240,000 new cases of prostate cancer will be diagnosed and 44,000 men will die from this disease in the United States alone in 1995 111. A major factor involved in the high death rate from prostate cancer is that in the majority of cases, prostate cancer is first discovered only when the disease has already spread outside the prostate, usualiy in the bones. At this very advanced stage, no curative therapy is available and despite the recent improvements of endocrine therapy, including prolongation of life, a cure is no longer possible [2-81. Although tancez is believed to be monoclonal at origin 191, it is now weU recognized, starting with the evidence obtained in colon cancer (101 and retinoblastoma 111, 123 thai evolution of the cancer and cell division are accompanied by the accumulation of gene abnormalities with progressive increase in malignancy and abnormal ce11 replation, a process called clona1 evolution. In addition, gowth factors secreted by neighboring nomial or cancer ceils can act in a paraaine fashion and stimulate proliferation of cancer cells [13,14].

Androgen-dependent mouse mammary carcinoma (SC115) was established in 1964 by Minesita and Yamaguchi [15,16]. It has ken widely used as a mode1 of androgen-sensitive tumor under both in vitro [17-201 and in vivo 121-231 conditions. In the present study, we have compared the response to androgen deprivation according to the size of the Shionogi tumors in mice implanted with a silastic implant which releases a physiological dose of androstenedione (~4-dione)to more closely mimic the human biology situation. MATERUUS AND METHODS

Male DD/S Shionogi mice (weight, 30-35 g) raised in our laboratory wer used at 60 days of age and howd S per cage in a tempera- (23 f 1°C)- PN light (14h iight/day, lights on at 0600)- conttrolied environment. The mice wer given ad libitum access to rodent chow and tap water. This study was done U compliance with the Canadian Councii on Animal Care (CCAC) regulation am the local conunittee of laboratory animal protection. Shionogi male mic bearing androgen-sensitive mammary tumors were originally provided by DI K. Matsumoto (Osaka, Japan). Tumor fragments were inoculatei subcutaneously (s.c.) into the dorsal area of Shionogi mice as described [21,24: To obtain tumors of different sizes at the time of starting treatment, th inoculations were processed at various times.

At a minimum of 2 weeks after inoculation, the mice bearing Shionog tumors with diameters of 0.1 to 1.8 cm were orchiectomized, received ai androstenedione (ad-dione) implant and were randomly divided into th following groupr according to tumor si& (20 to 25 mi& per group): diameter a 0.1 to 0.30 cm, 0.3 to 0.5 an, 0.5 to 0.75.cm, 0.75 to 1.0 cm, 1.0 to 1.4 an0and 1.4 tc 1.8 cm. Animals were treated with the vehicle alone or with 1 mg of flutamid dissolved in 5% ethanol and administered in 0.2 ml 1%(w/v) gelatin-û.9% NaC administered s.c., twice daily. Castration was petformed under genera anesthesia (Avertin) via the scrotal route on day 1. The implants (1 an long were inserted s.c. in the dorsal area of the mice at the the of orchiectomj Twnor growth in each animal was esknated by measuring every 2 days th tumor average area (product of the two longest perpendicular diameter measured using a Vernier caliper, Fisher Scientific). Ihe mice were Lilled < decapitation on the 16th morning of treatment The last dose of flutamide wa administered and the last mersurement of tumor size was perforrned on th morning of sacrifice. The prostate, seminal vesicles, and lcidneys were ripi. rernoved, dissected, and weighed. The kidneys were frozen in liquid nitrogel and stored at -80°C for measurement of ornithine decatboxybse (ODC) aetivity. in the second part of the experiment, fragments of large tumors tha continued to grow under flutamide treatment, namely those having 1.0 to 1.1 cm and 1.4 to 1.8 cm in diameter at the start of the experiment were minced intr srnall fragments [21,24] and implanted into new groups of mice. Then, 2 week later, amimals bearing tumors having 0.1 to 0.3 cm in diameter were selected am randomly divided into control and flutamide-treated groups. Treatment wa the sarne as describeci above for the first part of the experiment.

Matenals Flutamide was generously supplied by Schering Plough Corporatioi (Kenilworth, NJ)while ~4-dionewas purchased from Steraloids Inc. (Wiltor NH). The size of a4-dione silastic implants (Dow Corning) was 0.078 inne diameter, and 0.125 outer diameter.

Frozen kidneys were placed in 2 ml of ice-cold buffer (50 mM Tris-HC1, pI 7.5; 100 $v¶ pyridoxal phosphate; 100 ph4 of EDTA; 5 mM DL-dithiothreitol) an1 homogenized for 20 sec with a polytron tissue homogenizer set at a speed of ; After centrifugation of the homogenate at 12,000 x g for 20 min at 4"C, th supernatant was used to measure of ODC activity and protein content. Thex 200 4 aliquots of the homogenate and 50 of the reaction mixture (250 mM c Tris-HC1; pH 7.5; 200 piid pyridoxal phosphate; 12.5 mM DL-dithiothreitol; 10 mM L-ornithine hydrochloride) containing 0.25 pCi of L[1-14CJornithin (Amersham, 58 ci/mol; 0.1 pCi/pl) were added to the test tubes and mixed. Th test tubes were closed with a rubber septum holding a propylene via1 (Konte Glass Company, Vineland, NJ)containing 200 of methylbenzethoniu~ hydroxide (Signa, St. buis, MO). The incubation was camed out for 30 min 2 37OC in a water bath under constant shaking. The reaction was stopped b adding 0.3 ml of 6N H2SO4, and the test tubes were incubated for an addition; 30 min to ensure complete coilection of WO2. The ODC: activity was calculate from the amount of radioactivity present in the propylene vials and measutel by liquid scintillation spectrometry. Protein content was measured according t the method of Bradford [25]. Results are expressed as mol 14CO2 formed/3 min/mg protein). Statisties Statistical signifieance was measured according to the multiple-range test 01 Duncan-Ki:amer 1261. Data are expressed as means f SEM. RESULTS

Figure 1 shows the effect of twice daüy injection of 1 mg flutarnide for 1C days on the average area of tumors having different sizes at start of treatment Significant deaeases of 46.6 f 8.0% (pcO.01) and 56.9 f 5.9% (p<0.01) wen measured in the 0.1 to 0.3 cm and 0.3 to 0.5 cm initial diameter groups, respectiveiy. The magnitude of the inhibitory effect deaeased with increasing tumor size with inhibitions of 33.2 f 2.2% (pe0.01) and 31.6 f 3.0% (pc0.05) ir the tumors measuring 0.5 to 0.75 cm and 0.75 to 1.0 cm at start of treatment respectively. The deaease in average tumor area in the larger tumor gmups, namely those having 1.0 to 1.4 cm and 1.4 to 1.8 an in diameter, at the start 01 treatment, did not reach the level of significance.

As an example, Fig. 2 shows the time-course of the effect of androger deprivation on tumor growth during the 16-day treatment period for tumors having diameters of 0.1 to 03 cm (Fig. 2A) and 1.4 to 1.8 cm (Fig. 2B) at start ol treatment. In the 0.1 to 0.3 ,cmgroup, the first significant inhibitory effeci (42.9 f 8.6% deaease, pe0.01) was observed on day 6 of treatment and continued up to day 16, the last time interval studied. On the other hand, as shown in Fig' 28, average tumor area inaeased steadily in bath conttol and flutamide-treated animals and only small and nonsignificant changes could be seen during the 16 day observation perid-

It is weil known that prostatic weight is a sensitive parameter of androgen action As illustrated in Fig. 3,lMay treatment with fiutamide caused a marked inhibition of prostatic weight ranging from 52.6 f 4.3 to 70.5 f 6.8% in the various groups and no significant difference of effect was found between the groups. Similarly, another androgen-sensitive parameter, seminal vesicle weight, decreased by 47.9 f 3.6 to 56.7 f 1.7% after 16 days of treatment (data noi shown). As mentioned above for prostatic weight, no significant difference was found between animals bearing small and large tumors.

Mouse kidney ODC activity is a highly sensitive parameter of androgen action which demases dramaticaiiy in a few hours after ostration and is rapidly stimulated by testosterone and dihyd.mtesbstenaie 127.28). We have shown in a preiiminary experiment, that the d-dione implant used (1-cm long) completely reverses the effect of castration on this parameter. Twice daily treatment with 1mg of flutamide for 16 days decreased kidney ODC activity t about 1%of control in aii groups (Fig. 4).

Most importantly, it can be seen in Fig. 5 that Oie small tumors grown froi ftagments obtained from large unresponsive tumors (1.0 to 1.8 cm in diamete Fig. l), regained full responsiveness to the androgen deprivation achieved wit flutamide. In fact, 2 weeks after reimplantation, tumors of 0.1 to 0.3 cm i diameter were divided into control and flutamide-treated groups. As we iilustrated in Fig. 5, the average tumor area was aiready deaeased by 40.9 f 9.4' on day 4 (pg0.01) of flutamide treatment. A comparable 60% to 65% inhibitor effect lasted until &y 16, the last the interval studied. DISCUSSXON

The present data clearly demonstrate the aitical effect of tumot size on thc inhibitory response of tumor g~owthto androgen deprivation In fact, tumor! having a largest diameter lower than 1 cm regessecl rapidly following androger blockade, whereas larger tumors showed little or no response to the sam treatment with the pure antiandrogen flutamide. Even more significant war the finding that when fragments of large tumors unresponsive to androgei deprivation were reimplanted and were submitted to the same antiandrogenic treatment before they reached the critical size, the full inhibitory effect O, androgen deprivation was regained. Such data strongly suggest that the size O the tumors rather than the genetic or the phenotype of the tumor cells i! involved. In agreement with this data, it has previously been reported that aftei initiation of growth by androgen, the subsequent fate of SC115 tumor followiq androgen removal dependent on the size of ihe himor: small, medium, an( large hunors showed complete, temporary, and no regression, respectively [29] On the other hand, the same antiandrogenic treatment inhibited prostate ant seminal veside weight and dramaticaily deaeased mouse kidney ODî activity to the same degree in al1 groups of animals.

Afthough tumors are believed to be monoclonal at origin, clearly most, i not ali, advanced tumors represent a mixed population of cells with a widi range of phenotypes with different morphological properties, chromosoma composition, biochemical markers, metastatic behavior, and sensitivities O subpopulations to treatment modalities [9]. in fact, Our previous in vitro datl obtained with the Shionogi tumor ce11 line dearly showed that a wide range o sensitivities were present in the clones derived from a single tumor and thai moreover, heterogeneity rapidly develops during tissue culture of cloned cell 116,171. In fact, the 3 clones obtained from a single Shionogi mouse marnmar] -or showed marked heterogeneity of sensitivity to DHT action. The origina tumor had a Km value of 0.9 nM for DHT, while Km values of 0.024 nM, 0.11 nM, and 0.30 nM were measured in those three different clones. in fact, therl was a 1,250-fold difference in Se~iti~itYta DHT between the three done denved hma single himot (17). Such donal wolution permits cancer ceiis ti accumulate successive gene rbmrmaiitiis while becoming more aggressive am with inaeaed metastatic potential(10,13,14]. It is therefore reasonable to expec that delaying treatment of prostate cancer permits an increase in the heterogeneity of sensitivity of tumors to androgens, thus decreasing the efficacy of response to combination therapy [7].

Contrary to the thesapeutic approadi suggested by the VACURG (Veterans Administration Cwperative Urological Research Group) study in 1967 (301 favoring delayed hormonal therapy versus early treatment in patients diagnosed as having metastatic prostate cancer, the present data demonstrate the dramatic benefit of treatment beginning when tumors are small, thus strongly suggesting early combination treatment. In agreement with these experimental data, Eisenberger et al. [31] have observed that the benefits of combination therapy (Flutamide + Lupron) in stage D2 patients having minimal disease translated into a prolongation of 19 rnonths of Me while the benefit was only 5 months for those having a more severe disease. Similar data have been obtained in EORTC (Eutopean Organization for Research and Tseatment of Cancer) Study 30853 (L. Denis, personal communication) and we have observed that administration of combination therapy to stage D2 prostate cancer patients having 1 to 5 bone metastases adds a minimum of 4.4 years of good quality of life compared with patients whose disease is more advanced [32).

Although long-term progression of the tumors leads to clonal evolution and to the appearance and accumulation of gene abnormalities [IO-121,the present data show that the size of tumors can also be a crucial factor to take into account. A possible explanation for the present findings could be the paracrine secretion of stimulatory growth factors [13, 141 or the limited diffusion of flutamide in the larger tumors. This fast possibility would however require thaf ~4-dionewould diffuse more easily than the antiandrogen Independently ffom the mechanisrns involved, the present observations provide further support ta the rapidIy recognized importance of early treatment of prostate cancer, a proposa1 in agreement with convincing clinical data [L. Denis personna1 commwiication and 7, 25, 261 but not yet genesalized in the medical practice. This approach of early treatment is so aitical that the patients with "minimal disease" in the NCI (National Cancer Institute) study who received combination therapy remained free of disease progression for an average of 49 months whüe 50% of the patients in the group who received castration id placebo had already died at 42 months or 7 months before the patients in the Fiutamide- treated group had progressed, thus illustrating the particularly high sensitivitJ to androgen deprivation at this moderately advanced stage of the disease. Thc high sensitivity of early prostate cancer is ahclearly demonstrated by ax increase in organ cbnfined disease from 49.3 to 78.3% foUowing only 3 monthi of combination therapy compared to no endocrine therapy before radica prostatectomy [33]. REFERENCES

1. Wingo, P-A, Tong, T. and Bolden, S.: Canca statirtics, 1995. CA Cuncer Clin. 45 (1995) 8-30.

2 Labrie, F., Dupont, A., Bélanger, A., Cusan, L., Lacourcière, Y., Mpnfette, ( Laberge, J.G., Emond, J., Fazekas, A.T.A., Raynaud, J.F. and Husson, J-b New hormonal therapy improstatic carcinonu: combirkd treatment with . LHRH agonist and an antiandrogeh Uin boat Med S (1982) 267-275.

3. Labrie, F., Dupont, A. and Bélanger, A.: ~orn~lekanârogen blockade for t treatment of prostate cancer. In lmporfant Adtiances in ûncology (Edited 1 V.T. de Vita, S. Hellman and S.A. Rosenberg). J.B.- Lippincott, Philadelpl (1985)'193-217.

4. Crawford, D., Eisenberger, MA., McLeod, DG., Spaulding, J.T.,Benson, 1 Dorr, F.A., Blumenstein, D.A., Davis, M.A. and Goodman, P.J.: A 'controlk trial of leuprolide with and without flutamide in prostatic carcinoma. Nt Engl. J. Med. 321 (1989) 419-424.

5. Denis, L., Cameiro de Moua, J.L., Bono, A., Sylvester, R., Wheeton, 1 Newling, D. and Pauno, M.D.: Goseselin acetate and flutamide vs bilatei orchiectomy: a phase Jïï EORTC trial (30853). Urology 42 (1993) 119-129.

6. Janknegt, R.A., Abbou, C.C., Bartoletti,.R, Bernstein-Hahn, L., Bracken, 1 Brisset, J.M.,Silva, F.C.d., Knonagel, H. and Vemer, P.: Orchiectomy ii~ Anandron (Nilutamide) or placebo as treatment of metastatic prostal cancer in a multinational double-blind randomized trial. J. Urology 14 (1993)77-83,

7. tabrie, F.: Endocrine therapy of prostate cancer: optimal form and timing. Clin. Endhnol. Mefab. 80 (1995) 1066-1071.

8. Labrie, F., Cusan, L., Gomez, J.L., Diamond, P. and Candas, B.: Combinatit of screening and preoperative endocrine therapy: the potential for a important deaease in prostate cancer mortality. J. Clin. Enducrillol. Mefa a0 (1995)2002-2013. 9. Dexter, D.L. and Calabresi, P.: Intraneoplastic diversity. Bioch. Biophyi Acta 695 (1982) 97-112.

10. Fearon, E.R and VogeIstein, B.: .Aenetic mode1 for cdotectal tumorigenesi CeU 61 (1990) 759-767

11. Nowell, P.C.: The clonal evolution of tumor ce11 populations. Science 19 (1976) 23-28.

12. Saable, HJ., Sapienza, C. and Cavenee, W.K.: Genetic and epigenetic loss of heterozygosity in cancer predisposition and progression. Mv. Cuncer Rei 54 (1990) 25-62.

13. Artega, C. L. and .Osborne, C. K.: Growth factors as mediators c estrogen/antiestrogen action in human breast cancer cells In Regulator mechanisms in breast cancer (Edited by M.E. Lippman and RB. Dickson Kluwer Academic Press, Boston (1991) p289.

14. Macïndoe, J.: The hydrolysis of estrone suifate and dehydroepiandrosteron sulfate by MCF-7 human breast cancer cells. Endocrino1123 (1988) 1281-1287,

15. Minesita, T. and Yamaguchi, K.: An androgen-dependent tumor derivel from a hormone-independent tumor of a female mouse. Steroids 4 (1964 815-830.

16. Minesita, T. and Yamaguchi, K.: An androgen-dependent mouse mammar tumor. Crnrcer Res. 25 (1965) 1168-1175.

17. Labrie, F. and VeilIewc, R: A wide range of sensitivities to androgen develops in cloned Shionogi mouse mammary tumor ceus. The Prostate ; (1986) 293-300.

18. Labrie, F. and Veilleux, R.: Maintenance of androgen responsiveness b: glucocorticoids in Shionogi mammary car cinoma cells in culture. J. Nat, Cancer Inst. 80 (1988) 966-970.

19. Labrie, F., Veilleux, R. and Fournier, A.: Glucocorticoids stimulate th' growth of mouse mammary carcinoma Shionogi cells in culture. Mol. Ccli Enddnol. Sû (1988) 207-211. 20. Labrie, F., Veilleux, R. and Fournier, A.: Low androgen levels induce th6 development of androgen-hypersensitive cells clones in Shionogi mous6 mammary caranorna cells in culture. J. Nafl. Cancer Insf. 80 (1988) 1138. 1147.

21. Plante, M., Lapointe, S. and Labrie, F.: Stimulatory effect of synthetic progestins currently used for the treahnent of prostate cancer on growth oj the androgen-sensitive Shionogi tumor in mice. J. Steroid Biockm. 31 (1988) 61-64.

22. Suzuki, T., Horibe, I., Uchida, N., Ezumi, K., Uchida, K., Tadeda, k., Tanaka, A., Nishizawa, Y. and Matsumoto, K.: Effects of antiandrogens on pwth01 androgen-dependent mouse mammary tumor (Shionogi Carcinoma 115) ir vivo and in vitro. J. Steroid Biochem. Molec. Biol. 37 (1990) 559-567.

23. Noguchi, S., Nishizawa, Y., Motomura, K., Inaji, H., Imaoka, S., Koyama, H, and Matsumoto, K.: Inhibitory effect of a somatostatin analogue (SMS 201- 995) on the growth of androgen-dependent mouse mammary tumoi (Shionogi Carcinoma 115). Ipn 1. Cancer Res. 84 (1993) 656-663.

24. Luthy, I., Bégin, D. and Labrie, F.: Androgenic activity of synthetic progesh and spironolactone in androgen-sensitive rnouse mammary carcinoma (Shionogi) cells in culture. J. Steroid Biochem. 31 (1988) 845-852.

25. Bradford, M.: A rapid and sensitive method for the quantitation ol microgram quantities of protein utilizing the principle of protein-dyc binding. Anal. Biochem. 72 (1976) 248-254.

26. Kramer, C.Y.: Extension of multiple range tests to group means with uniqu~ numbers of replicatiors. Biomet rics l2 (1956) 307-310.

27. Bullock, L.P.: Androgen and progestin stimulation of ornithine decarboxylase activity in the mouse kidney. EndocrinoZogy 112 (1983) 1903- 1909.

28. Pajunen, A.E.I., Isomaa, V.V., Janne, O.A. and Bardin, C.W.:Andmgenic regulation of ornithine decarboxylase activity in mouse kidney and its relationship to changes in cytosol and nuckar androgen receptor concentrations. J. Biol. Chem. 257 (1982) 8190.8198.

29. Kitamura, Y., Okamoto, S., Uchida, N., Yamaguchi, K. and Marsumoto, K.: Effect of androgen depletion on growth and androgen dependency of Shionogi carcinoma 115. Cancer Res. 38 (1978) 47114716.

30. Veterans Administration Cooperative Urological Research Group: Treatment and survival of patients with cancer of the prostate. Surg. Gynecol. Obstet. 124 (1967) 1011-1017

31. Esenberger, M.A., Crawford, D.E.,Wolf, M., Blumerstein, B., McLeod, D.G., Benson, R., Don, F.A., Benson, M. and Spaulding, J.T.:Prognostic factors in stage D2 prostate cancer - important implications for future trials - Results of a cooperative intergroup study. Semin. Oncol. 21 (1994) 613419.

32. Labrie, F., Dupont, A., Cusan, L., Gomez, J.L. and Diamond, P.: Major advantages of "early" administration of endocrine combination therapy in advanced prostate cancer. Clin. Invest. Med. 16 (1993) 493498.

33. Labrie, F., Cusan, L., Gomez, J.L., Diamond, P., Subutu, R., Lemay, M., Têtu, B., Fradet, Y. and Candas, B.: Downstaging of early stage prostate cancer before radical prostatectomy: the first randomized trial of neoadjuvant combination therapy with flutamide and a luteinizing hormone-releasing hormone agonist. Urology 44 (1994) 29-37. LEGENDS TO FIGURES

Figure 1. Effect of tumor size on the inhibitory effect of flutamide on Shionog iumor growth Flutamide was injected at the dose of 1 mg, twice daily, for 1C days in DD/S mice implanted with androstenedione (1 cm long implant). 'Ru size of the tumors in the different groups ranged from 0.1 to 1.8 cm in diametel at day 0. Data are expressed as means f S.E.M. (n=25-30). *p<0.05; -p<0.01, Flutamide treatment vs control.

Figure 2. Time-course of the inhibitory effect of flutamide on the growth 01 Shionogi tumors in DD/S mice bearing tumors having initial diameters of 0.1 tc 0.3 cm (2A) and 1.4 to 1.8 cm (2B). Data are expressed as means f S.E.M. (n=25 30). *p<0.05; **p<0.01, Flutamide treatment vs control.

Figure 3. Effect of flutamide treatment on prostatic weight in mice bearing Shionogi tumors of the indicated sizes at the start of treatment. Data arc expressed as means f S.E.M. (n=25-30). **p<0.01, Fhtamide trea tment vr control.

Figure 4. Effed of 16-day treatment with flutamide on the kidney ODC activity in DD/S mice bearing Shionogi tumors 'of the indicated sizes at the start 01 treatment. Data are expressed as means f S.E.M. (n=6). '~0.05; **p<0.01, Flutamide treatment vs control.

Figure 5. Effect of 16-day treatment with flutamide on the growth of Shionogj tumors in DD/S mice inoculated with fragments of the nonresponsive tumors shown in Fig. 1. The treatment was started at a tumor size of 0.1 to 0.3 cm in diameter. Data are expressed as means f S.E.M. (n=25). '~0.05; *+p<0.01, Flutamide treatment vs control. FIGURE 1

0 CONTROL FLVTAMIDE, (1 mg. TWICE DAILY)

1 n.. l

0.1-0.3 f 0.3-0.5 1.0-1.4 1.4-1.0 TUMOR SR€ (LARGES1 DIAMETER ON DAY O, cm ) FIGURE 2

0.1-0.3 cm

0 COMROL FWAMIDE, ( lmg. TWCE DAlLY) T

OAY S OF 1REATMENT FIGURE 3

n CONTRIOL &I FLWAMIDE, (1 mg, TWlCE DAILY)

TUMOR SlZE (LARGEST DIAMETER ON DAY O. cm ) FIGURE 4

0 CONTROL FLUTAMIDE, (lmg, B.1.D.)

TUMOR SITE ( LARGES1 DIAMETER ON DAY O, cm ) I FIGURE 5

CONTROL FLLUTAMIDE, (lmg, WlCE DAltY)

DAYS OF TREATMENT ADDïïiVE IN VLVO GROWTH INHIBITORY EFFECTS OF FLUTAMIDE AND FINASTERIDE ON ANDROGEN- SENSITIVE SHIONOGI 115 CARCiNOMA

Cailin Chen, Xun Li, Shankar Singh, Alain Bélanger, and Femand tabrie

Medical Research Council Group in Molecular Endocrinology, CHUL Research Center and Laval University, Quibec, GIV 4G2, Canada ABSTRACT

The antiandrogen flutamide and the Sa-reductase inhibitor finasteride were administered twice da* at the dose of 1 mg aIone or in combination in mice bearing androgen-sensitive Shionogi tumors, a model predictive of androgen-sensitive prostate cancer in men. Ïntact or orchiectomized animais were supplemented with an implant of androstenedione in order to rnimic the contribution of androgens of adrenaî origin in men. Treatment for 20 days with flutamide alone caused 28% and 36% deaeases in tumor size and prostatic weight, respectively, whereas finasteride alone caused 18% and 16% inhibitions of the same parameters. On the other hand, combination of the two compounds inhibited by 43% and 58% the value of the same parameters. In orchiectomized animals also implanted with androstenedione, flutamide alone caused 35% and 39% decreases in tumor growth and prostatic weight, respectiveIy, whereas finasteride alone caused respective 27% and 15% inhibitions, and combination treatment caused 53% and 61% inhibitions in tumor growth and prostatic weight, respectively. The present data show that the inhibitory effects of flutamide and finasteride on Shionogi tumor and prostate growth are nearly additive or additive, respectively, and suggest that such a combination of the two compounds could provide the basis for further improvement of the endocrine therapy of prostate cancer and induce the maximal degree of apoptosis or cancer ce11 death. The limited efficacy of the Sa-reductase inhibitor alone resulting from the increased intratamoral testosterone concentrations can be regained by the antiandrogen, thus permitting complementary action of the two compounds. The present data also show that the maximal inhibitory effects aze achieved with the triple combination of castration, a pure antiandrogen, and a Sa-reductase inhibitor. Prostate cancer is the most frequent cancer in men, with an estimatec incidence of 240,000 new cases in the United States alone in 1995 while 44,001 men are expected to die from this disease during the same period (11. The bes known characteristic of prostate cancer is its high rensitivity to androge] deprivation. In fact, since the observation of Huggins and Hodges [2], blockadl of testicular androgens has been the standard treatment for advanced diseas6 Recently, more complete androgen blockade by combination therapy has led ti the first demonstration of improved survival in advanced prostate cancer [3-7 and prornising results have been obtained uçing the same combination therap: before radical prostatectomy [8,9].

An important consideration in the endoaine therapy of prostate cancer i that both the testes and the adrenals provide almost equal amounts of th androgens responsible for stimulation of cancer ceii growth (4, 10). Testosteron from testicular origin can be easily eliminated by orchiectomy (21 or by medica castration with a luteinizing hormone-releasing hormone agonist [Il], wherea androgens of adrenal origin can best be neutralized by blocking the androgei receptor with a pure antiandrogen (3, 4, 121. Unfortunately, al1 the presen* available antiandrogens have a relatively low affinity for the androgen recepto [13], thus leaving significant concentrations of intracellular Sa dihydrotestosterone (DHT)free to continue to interact with the androgei receptor, especially in androgen-hypersensitive tumors 114, 151 where lou androgen levels are sufficient to stimulate cancer growth. The importan clinical advantages demonstrated by the therapy combining a pure antiandrogei and castration in both advanced (3-7land localized [8,9] prostate cancer raises th hope that a more complete androgen blockade could further improve th therapy of this disease by further induction of apoptosis or cancer ce11 deatk rl6,m

Testosterone is converted by Sa-reductase into DHT which interacts mor effiaently with the prostatic androgen receptor (18-211; therefore, in order ti achieve a more complete androgen blockade, a logical site of therapeuti intervention is this last step in Dmformation using an inhibitot of Sa reductase to deaease intrapmtatic Dmlevels, thus faciiitating the neutriùiziq action of the antiandrogen in support of this approach, we previously showed that the inhibitory effects of the antiandrogen flutamide and the Sa-reductase inhibitor &MA are additive on androstenedione (~4dione)-stimulatedprostatic growth and on prostatic binding protein mRNA levels in the castrated rat [22,23]. However, the potential benefits of such combination therapy had not been investigated in a tumor model.

The androgen-sensitive mouse mammary Shionogi carcinoma has been widely used as model of prostate cancer under both in vitro [14,15,24-261 and in vivo conditions [27, 281. In the present study, we used the same rnodel ta investigate the effects of finasteride and flutamide, alone or in combination, on Shionogi tumor and prostate growth in intact and castrated animals thai received ~4-dioneimplants to mimic the human situation of a continuous secretion of precursors of androgens by the adrenals [29,30] We also measured semm and tumor steroid levels in order to better understand the mechanisms involved. MATERIALS AND METHODS

Mamials Flutamide was supplied by Schering Plough Corporation (Kenilworth, NJ while finasteride was synthesized in our medicinal chemistry division, and A*. dione was purchased from Steraloids hc. (Wilton, N.R.).The size of the Silasth implants (Dow Corning) was 0.078 inches (inside diameter) and 0.125 inch (outei diameter).

Following methanol and diethyl ether extraction for tissue samples anc diethyl ether extraction for serum samples followed by chromatography on LE% 20 colurnns, steroid concentrations were measured by radioimunoassays using antisera developed and characterized in ou laboratory, as described in detaï previously (29, 31-34]. Al1 the samples were chromatographed anc radioimrnunoassayed simultaneously.

Male DD/S mice weighing 30 to 35 g bred in oui laboratory, were used at : months of age and housed five per cage in a temperature (23 f I0C)-and lighi (14 h light/day, lights on at 6:00 a.m.)-controlled environment. The mice werc given ad libitum access to rodent chow and tap water. Shionogi male micc bearing androgen-sensitive mammary tumors were originally obtained througk the courtesy of Dr. K. Matsumoto (Osaka, Japan). For turnor transplintation, turnor-bearing mice were anesthesized by isofiurane, the tumors were ther quickly removed, dissected free from adhering fat, and co~ectivetissue. Afte~ rinsing twice in 0.9% saline, the tumors were minced in srna11 fragments and suspended in 0.9 NaCl (1 g tumor/20 ml saline) containing penicillin (200 U/d] and streptomycin (0.1 rng/ml). Then, the tumor fragments were injected s.c in the dorsal area of Shionogi mice. I Approximately two weeks after inoculation, the Shionogi mice bearinl tumors having a diameter of 0.1 to 0.3 an were randomly divided into thi following groups (20 mice per group): (1) intact controls; (2) intact androstenedione (Akhone) implant; (3) intact + ~4 dione+finasteride, 1 mg; (4 intact + ~d-dione+ flutamide, 1 mg; (5) intact + 64-dione + finasteride . flutamide; (6) castration + ~4-dione;(7) castration + ~4-dione+ finasteride, 1 mg (8) castration + dkiione + flutamide, 1mg; (9) castration + 64-d.ione + finasteridc + flutamide. Orchiectomy was performed via the scrotal route under genera anesthesia (Averti.). The 1-cm-long implants containing ad-dione won inserted S.C. into the dorsal area at time of orchiectomy while treatment witi flutamide and finasteride was initiated on the afternoon of the day O orchiectomy. The compounds originally dissolved in 5% ethanol were injectec s.c. twice daily in 1%(w/v) gelatin-û.9% NaCl, Tumor growth in each anima was estimated by measuring every 2 days their average area (product of the twc longest perpendicular diameters using a vernier caliper (Fisher Scientific). Thi mice were killed by decapitation on the 20th moming of treatment. The last dos was administered and the last measurement of himor size was performed 01 the morning of sacrifice. Trunk blood was pooled and collected foi measurement of plasma steroid concentrations. The prostate, liver, and kidnel were immediately dissected and weighed while the tumors were carefulll dissected free from adhering fat and co~ectivetissue. The prostate and tumon were then frozen in liquid nitrogen and stored at -80°C until assayed for 5aa reductase activity whiie tissues were stored at -20°C for steroid measurement.

Frozen tissues were homogenized with a Polytron in phosphate btiffer (2( mM KH2PO4, 0.25 M suaose, 1 mM EDTA, pH 7.5) containing proteast inhibitois (1 mM phenylrnethyisulfonyl fluoride and 5 (rg/ml each of peptatir A, antipain and leupeptin) and cehtrifuged for 30 min at 1000 x g to remove cd debris. Then, 100 pl of the supernatant were incubated for 120 min for tumoi 6amples uid 80 min for prostate samples at 37°C in a total volume of 0.5 ml phosphate buffer (12.5 mM mPO4, 1rnM EDTA, pH 7.5) containing 0.5 pkl [14C]-testosteroneand 1 mM NADPH. Labeled testosterone (51 mCi/rnmol) was purchased from New England Nuclear/Dupont (Markham, Canada) and purified by thin-layer chmnatography WC)More use. The enzymatic teactions were stopped by chilling the incubation mixture in an ice-water slurry and adding 3 ml of diethyl ether. The components were then mixed and frozen in a dry ice-ethanol bath. The organic phase was kept while the frozen aqueous fraction was reextracted once with ether. The organic phases were then poded and evaporated to dryness under a nitrogen stream. AU components were then separated by TLC using the soivent system containhg 4:l (v/v) taluene:acetone before autoradiography of the plates (60 Fmsilica gel, E. Merck, Darmstadt, Germany) for 48 h The metabolites revealed by autoradiography were identifiecl by cornparison with standard steroids characterized by HPLC as desaibed previously (351. The TLC areas corresponding to testosterone, Dm,androstane- 3a,17fi-diol, and androstane-3&17p-diol, were saaped and transferred into vials containing 0.5 ml of ethanol and 10 ml of scintillation liquid before measurement of radioactivity. Protein content was determined by the metfiod of Bradford using bovine serum albumin as standard 1361. Results are expressed in pmol or fmol Dmforrned/mg protein/min (DHT formation includes DHT itself as well as additional metabolites, namely, androstane-3a,l7$-di01 and androstane-3P,17&diol).

Statistical significance was measured according to the multiple-range test of Duncan-Kramer [371. Data are expressed as means f SEM. RESULTS

in intact mice implanted with ~4-dione,finasteride done adrninistered for 20 days, at the twice daily dose of 1mg, cada nonsignificant 18.2% inhibition of tumor size (Fig. 1). At the same dose, flutamide decreased tumor size by 27.8% (pc0.05). When the two compounds were administered simultaneously, however, the inhibitory effect was inaeased to 42.8% (pe0.01 versus control, pc0.05 versus flutamide, pe0.01 versus finasteride). In castrated mice supplemented with ~4-dione,the Sa-reductaso inhibitor alone caused a 27.0% (pc0.05) inhibition of tumor size while flutamide alone caused a 35.0% (p < 0.01) inhibition. Combination of the two compounds, on the other hand, caused a further decrease to a value of 53.1% (pc0.01 versus control, p<0.05 versus flutamide, and p < 0.01 versw finasteride).

As shown in Fig. 2A, the serum A4-dione concentration is very low at 0.29 f 0.04 nM in intact mice. The insertion of one 1 cm-long implant of ~4-dioneled to serum ~4-dionelevels in both intact and castrated DD/S mice bearing Shionogi tumors which are analogous to those seen in castrated men [33,38,39]. Treatment with finasteride or flutamide alone or in combination had no signihcant effect on serum ad-dione levels at a dosage of 1 mg twice daily for 20 days. On the other hand, as shown in Fig. 28, treatment of intact mice with finasteride alone or combined with flutamide increased the d4-dione concentration in the tumors by 52.2% and 79.7%, respectively. in fact, the tumor 44-dione concentration increased from 11.5 f 1.9 pmoles in intact mice implanted with ~4-dioneto 17.5 f 1.3 pmoles and 20.7 f 2.6 pmoles (p c 0.01) in mice treated with finasteride alone and in combination with flutamide, respectively. The hunor d-dione concentration was measured at 21.1 f 1.4 pmoles in castrated mice supplemented with ~4-dione,whereas increases to 28.0 f 3.2 pmoles (p < 0.05) and 32.6 f 3.0 pmoles @ < 0.01) were observed following 20 days of treatment with finasteride alone or combined with flutamide, respectively. By itself, flutamide had no detectable effect on tumor A'-dione levels in either intact or castrated animals.

Figure 3A shows that serum testosterone levels of 18.1 f 1.8 and 16.1 f 1.7 nm, respectively, were seen in control intact and castrated animals supplemented with an Ad-dione implant. Foilowing treatment with flutamide done, increases of 133.3% @ < 0.01) and 23.3% (p < 0.05) in semm testosteron' concentrations were obsewed in intact and castrated animals, respectively, wu uicreases of 204.9% @ < 0.01) and 38.0% @ < 0.01) occurred in the same groups a animals who received combination treatment. Finasteride alone had a significant effect on serum testosterone levels. Tumor testosterone levels, O: the other hand, increased hm5.20 f 0.5 in control intact anirnals supplementec with dione one to 11.6 f 0.5 pmoles (p<0.01) and 19.1 f 1.5 pmoles (p<0.01) i intact animals treated with finasteride and flutarnide, respectively (Fig. 38: Treatrnent with both the Sor-reductase inhibitor and the antiandrogen caused slightly furthex haease in tumor testosterone to 21.6 f 0.9 pmoles (pq0.01 v control and fmasteride gmups), on the other hand, tumor testosterone levels di1 not change sigrufiuntly in cashated animals treated with flutamide done wu they increased from 4.83 t 0.5 pmoles/g tissue to 9.24 24 0.74 pmoles (p<0.01) mi 8.77 f 1.0 pmoles (pe0.01) in anirnals treated with finasteride alone and ii combination with flutamide, respectively.

Serum DHT levels increased from 1.97 f 0.3 nM to 3.24 f 0.3 nM (p < 0.01 in intact animals following 20 days of treatment with flutamide alone while value of 2.62 f 0.2 nM (p c 0.01) was found following combination treatrnen with finasteride (Fig. 4A). SemDKT levels were not influenced significantl by treatment with flutamide done in castrated animals supplemented with b4 dione. While the 20.0% decrease in serum DHT induced by finasteride in intac animals was not significant, a 40.0% (pe0.05) inhibition of serum DHT' wa observed in castrated mice treated with finasteride alone. On the other hanc treatment with finasteride caused a 38.2% (pcO.05) decrease in tumor DH' concentration in intact mire whereas flutamide alone or in combination witl hasteride increased tumor DHT levels by 98.2% (p < 0.01) and 37.3% (p < 0.05: respectively (Fig. 4B). On the other hand, in castrated anirnals supplementet with Addone, treatment with finasteride caused a 36.9% @ < 0.05) deaease ii tumor DHT; in contrast, treatment with flutamide alone inaeased DHT level by 65.0% (p c 0.01) and combination treatment increased tumor DHT levels b: 59.2% @<0.01).

As shown in Fig. 5, treatment with finasteride and flutamide alone causec respective 16.1% ( p < 0.05) and 36.3% (p < 0.01) inhibitions of ventral prostab weight in intact arllmals mpplernented with A*-dione. Combination of the twi compounds, however, led to additive inhibitory effects rneasured at 58.1% (p 4 0.01 versus control, flutamide or finasteride alone). On the other hand, 15.1% (1 < 0.05) and 39.3% (p < 0.01) inhibitions were observed in castrated animal: supplemented with ~4-dionefollowing finasteride and flutamide treatment respectively, whiie a further decrease by 60.6% (p c 0.01 versus flutamide 01 finasteride atone) was observed on ventral prostate weight after treatment witl combination of the Su-reductase inhibitor and antiandrogen.

Figure 6A shows that finasteride alone, flutamide alone, and combinatioi of the two compounds deaeased tumor Sa-reductase by respectively 50.5% 31.8%. and 48.4% (p < 0.01 for ail groups) in intact animals bearing a A'-dion4 implant; decreases of 46.5% (p < 0.01), 18.0% (N.S.),and 37.7% (pc0.01) werc observed in the corresponding groups of castrated animals. On the other hand finasteride caused 44.0% (p < 0.01) and 42.3% (p < 0.01) inhibitions of prostatic Sa-reductase activity in intact and castrated animals, respectively, whereas nc significant effect was obtained in the other groups (Fig. 68). No significant effec on iiver or kidney weight was observed after 20 days of treatment witi finasteride or flutamide alone or in combination (data not shown). DISCUSSION

The present data clearly demonstrate that the inhibitory effects of the antiandrogen flutamide and the 5a-reductase inhibitor finasteride are nearly additive or additive on Shionogi tumor and on ventral prostate gmwth, respectively. The interest of the present study is strengthened by the use of d4- dione implants, which permit to achieve constant circulating levels of the steroid, rimiiar to those found in adult men, thus more do& mimicking the human situation, where the adrenals seaete a continuous supply of precursot steroids which are transforrned into the androgens testosterone and DHT in a large series of peripheral intracrine tissues, including the prostate [4,29,30]. In fact, in 65-year-old men, the androgens of adrenal origin are responsible, on average, for 40% of total androgens, contrary to the previous belief that only 5 ta 10%of androgens were of extratesticular origin (10).

in fact, possibly the most important discovery in the field of hormone- sensitive cancers in recent years is that humans are unique among species in having adrenals that secrete large amounts of the inactive precursor steroids dehydroepiandrosterone (Dm)and its sulfate (DHEA-S), which are converted into androgens and estrogens in peripheral tissues, the rate of transformation depending upon the level of expression of the various steroidogenic and steroid metabolizing enzymes in those tissues [IO, 301. The secretion of DHEA and DHEA-S by the adrenals inaeases during the adrenarche in childhd at the age of 6 to 8 years, and maximal values of circulating DHEA-S are found at the age of 20 to 30 years while a progressive decreases occurs thereafter [40,4l]. In fact, plasma DHEA-S levels in adult men are 100 to 500-fold higher than those of testosterone [4], thus providing the high level of substrate required for conversion into active sex steroids in peripheral tissues.

Although castration (orchiectomy or LHRH agonists) causes a 90 to 95% reduction in serum testosterone concentration, a much smaller effect is seen on the only meaningful parameter of androgenic action, namely, the intraptostatic concentration of DHT. Aftet blockade of the testes by medical or surgicd castration, the intraprostatic concentration of DHT remains at about 4û96 of that xneasured in intact men [4, 421. ïhe importance of extratesticular &gens is also be weU illustrated by the finding that in adult men, 40 to 50% of androgen metabolites remain in the circulation after castration 131,433.

We have previously used the Shionogi twnor ce11 line to study the heterogeneity of androgen sensitivity in different dones derived from a single tumor [14] and the changes in androgen sensitivity which develop during long terni culture of cloned cells in vitro (15). The data then obtained dearly showed that a wide range of hormone sensitivities were present in thtee clones derived from a single original tumor [14]. In fact, whereas the Km value of the original tumor for DHT action was measured at 0.9 nM, the Km values for DM stimulation in clones A, 8, and C were calculated at 0.024, 0.15, and 30 nM, respectively, thus showing a 12501foId difference in sensitivity to DHï between clones A and C obtained from a single tumor. Although the origin of tumors h believed to be monoclonal, clearly most, if not dl, advanced tumors are composed of mixed populations of cells with a wide range of phenotypes. Fol such androgen-hypersensitive tumors or these tumors responsive to very low levels of androgens, a more complete blockade of Dmis thus needed in order ta decrease the concentration of the steroid below the threshold of responsiveness of the cancer cell. niese hypersensitive tumors are those which require the more complete blockade of androgens which can only be achieved with the combination of flutamide and finasteride while the growth of other less androgen-sensitive tumors would be inhibited by a less complete andsogen blockade.

The study in castrated mice mimics the human situation where the testicles are removed surgically or blodced by an LHRH agonist and where the contribution of the adrenals corresponds to the ad-dione released by the androstenedione implant. In fact, the Shionogi tumors (Labrie et al., unpublished data) in analogy with the human prostate [IO] possess al1 the enzymes required to transfonn DHEA-S, DHEA or dione one into DHT.

The absence of significant change in either serum androstenedione or testosterone levels in intact or castrated animals treated with finasteride while the intratumoral concentration of the two steroids was elevated dearly shows the importance of the intracellular tegulation of steroid Ievels independently hmdetectable changes in circulating levels of the rame stemids. The serum and tumor testosterone concentrations. in intact mic supplemented with ~4-dionetreated with flutamide were high compared wit: the levels observed in castrated rnice similarly supplemented with 64-dione m treated with the antianârogen. In intact animals, flutamide is weii hwntl increase serum luteinizing hormone and testosterone levels, this effect bein, secondary to the neutralization by flutamide of the inhibitory feedback action c androgens at the hypothalamic and pituitary levels [44]. This limitation i eiiminated by surgical or medical LHRH agonist castration which, in any casr shouid always be part of the endocrine therapy of prostate cancer [4,8,45,46].

The present data show that the Sa-reductase inhibitor finastende caused 50% inhibition of tumor Sa-reductase activity, but had no significant inhibitor effect on tumor growth, this effect of finasteride king accompanied by increasei tumor testosterone concentration in both intact and castrated mice. Simila results were observed by Brooks et al. [47l who found that finasteride (MK-90t Proscar) failed to influence the growth of the R-3327 Dunning rat prostati carcinoma in testosterone propionate-treated castrates. Both studies strongl: indicate that the Shionogi tumor as well as the R-3327 tumor can responc directly to the increased intraceiiular levels of testosterone.

Despite the well recognized efficacy of finasteride and other Sa-reductas inhibitors to inhibit DHT formation, it seems logical that these compound should not be administered alone in the treatment of androgen-sensitiv diseases because of the well-recognized secondary increase in prostati testosterone concentrations [48], thus compensating for the finasteride-inducec decrease in intracellular DHT levels, which is likely to partially or totall: overcome the blockade of DHT formation achieved by the Sa-reductasi inhibitor. In fact, testosterone itself binds to the androgen receptor with ai affinity 25% to 35% that of DHT itself and evenhially fully activates th androgen recep tor [49].

As illustrated in the present study, the beneficial effects of the Sa-reductas inhibitor become significant following the addition of the *antiandrogei flutamide that ensures that no adverse androgenic effects result from the Sa reductase inhibitor-induced inaease in local testosterone concentration. T~I increased efficacy of finasteride on tumor growth and prostatic weight observa upon addition of the Sa-reductase inhibitor to the antiandrogen despite thi increase in intracellular testosterone can be explained by the lower affinity testosterone for the androgen receptor compared with Dm1491.

in sumrnary, the present data dearly show that combined treatment wi the antiandrogen flutamide and the Saoreductase inhibitor finasteride Ci inhibit in an additive manner prostate weight and in a near-additive mann Shionogi tumor growth in both intact and castrated mice. These findin confirrn the speaficity of the complementary inhibitory actions of the two dru, and suggest that the combination of both compounds may prove beneficial the treatment of diseases where optimal inhibition of DHT formation and actk is desired. Furthermore, the present data show that the maximal inhibito: effects are obtained with the triple combination of castration, a -pu antiandrogen, and a 5a-reductase inhibitor. REFERENCES

1. Wingo, P.A., Tong, T. and Bolden, S.: Cancer statistics, 1995. CA Cancer J8 Clin. 4s (1995) &30.

2. Huggins, C. and Hodges, C.V.: Studies of prostatic cancer. 1. Effect ot castration, estrogen and androgen injections on serum phosphatases ir metastatic carcinoma of the prostate. Cancer Res. 1(1941) 293-307.

3. Labrie, F., Dupont, A., Bélanger, A., Cusan, L., Lacouta&e, Y., Monfette, G., Laberge, J.G.,Emond, J., Fazekas, A.T.A., Raynaud, J.P. and Husson, J.M.: New hormonal therapy in prostatic carchorna: combined treatment witk an LHRH agonist and an antiandrogen. Clin Intlest Med 5 (1982) 267-275.

4. Labrie, F., Dupont, A. and Bélanger, A.: Complete androgen blockade foi the treatment of prostate cancer. In Important Advances in Oncologs (Edited by V.T. de Vita, S. HeUman and S.A. Rosenberg). J.B. Lippincott Phiiadelphia (1985) 193-217.

5. Crawford, D., Eisenberger, M.A., McLeod, D.G., Spaulding, J.T., Benson, R, Dorr, F.A., Blumenstein, D.A., Davis, M.A. and Goodman, P.J.: A controlled trial of leuprolide with and without flutamide in prostatic carcinoma. New Engl. J. Med. 321 (1989) 419-424.

6. Denis, L., Carneiro de Mowa, J.L., Bono, A., Sylvester, R, Wheeton, R Newling, D. and Pauno, M.D.:Goserelin acetate and flutamide vs bilatera orchiectomy: a phase Ri EORTC trial (30853). UroZogy 42 (1993) 119-129.

7. Janknegt, RA., Abbou, CC., Bartoletti, R, Bernstein-Hahn, L., Bracken, B., Brisset, J.M., Silva, F.C.d., Knonagel, H. and Venner, P.: Ordiiectomy ani Anandron (Nilutamide) or placebo as treatment of metastatic prostatic cancer in a multinational double-blind randomized trial. J. Urology 145 (1993) 77-83.

8. Labrie, F., Dupont, A., Cusan, Lm,Gomez, J.L., Diamond, P., Koutsilieris, M., Suburu, R, Fradet, Y., Lernay, M., Têtu, B., Emond, J. and Candas, B. Downstaging of locrlized prostate cancer by neoadjuvant therapy wit) flutamide and lupron: the first controiled and randomized trial. Clin. Invest. Med. 16 (1993) 511-521.

Labrie, F., Cusan, L., Gomez, J.L., Diamond, P., Suburu, R, Lemay, M., Tetu, B., Fradet, Y. and Candas, B.: Down-staging of early stage prostate cancer before radical prostatectomy: the fïrst randomized trial of neoadjuvant combination therapy with Flutarnide and a luteinizing hommereleasing hormone agonist. Urology 44 (1994) 29-37.

Labrie, F., Manger, A., Dupont, A., Luu-The, V., Simard, J. and Labrie, C.: Science behind total androgen blockade: from gene to combination therapy. Clin. Inwst. Med. 16 (1993) 487-504.

Labrie, F., Bélanger, A., Cusan, L., Séguin, C., Pelletier, G., Kelly, P.A., Reeves, J.J., Lefebvre, F.A., Lemay, A. and Ranaud, J.P.:Antifertility effects of LHRH agonists in the male. J. Androl. 1(1980) 209-228.

Labrie, F.: Mechanism of action and pure antiandrogenic properties of flutamide. Cancer 72 (1993) 381603827.

Simard, J., Luthy, I., Guay, J., Bélanger, A. and Labrie, F.: Characteristics of interaction of the antiandrogen flutamide with the androgen receptor in various targe t tissues. Mol. Cell. Endocrinol. 44 (1986) 261-270.

Labrie, F. and Veilleux, R: A wide range of sensitivities to androgens develops in doned Shionogi mouse mammary tumor ceiis. The Prostate 8 (1986) 293-300.

Labrie, F., Veilleux, R and Fournier, A.: Low androgen levels induce the development of androgen-hypersensitive cells clones in Shionogi mouse mammary carcinoma cells in culture. 1. Natl. Cancer Inst. 80 (1988) 1138- 1147.

Vaw, D.L., Haecker, G. snd Strass-, A.: An evolutionary perspective on apoptusis. Ceil 76 (1994) 777-779.

Kerr, J.F., Wyllie, A.H. and Currie, A.R: Apoptosis: a basic biological phenornenon with wide-ranging implications in tissue kinetics. Br. J. Can~n26 (1972) 239-257. Andersson, S., Bishop, kW. and Russell, D.W.: Expression cloning anc regdation of steroid Sa-reductase, an enzyme essential for male sexui differentiation J. Biol. Ch.264 (1989) 16249-16255.

Andersson, S., Bergman, D.M., Jenkins, E.P. and Russel, D.W.: Deletion oi steroid SU-reductase 2 gene in male pseudophermaphroditisn Nature 354 (1991) 159-161.

Labrie, F., Sugimoto, Y., Luu-The, V., Simard, J., Lachance, Y., Bachvarov D., Leblanc, G., Dumcher, F. and Paquet, N.: Structure of human type II Sm reductase. Endocrinology 131 (1992) 1571-1573.

Anderson, K.M. and Liao, S.: Selective retention of dihydrotestosterone bj prostatic nuclei. Nature 219 (1968) 277-279.

Labrie, C., Tnidel, C., Li, S., Martel, C., Couet, J. and Labrie, F.: Combinatior of an antiandrogen and a Sa-reductase inhibitor: A further step toward: total androgen blockade? Endocrinology 129 (1991) 566568.

Martel, C., Trudel, C., Couet, J., Labrie, C., Bélanger, A. and Labrie, F. Blockade of androstenedione-induced stimulation of androgen-sensitivi parameters in the rat prostate by combination of flutamide and W.Mol Cell. Endocrinol. 91 (1993) 43-49.

Minesita, T. and Yamaguchi, K.: An androgen-dependent mousi mammary tumor. Gzncer Res. 25 (1965) 1168-1175.

Labrie, F. and VeUux, R.: Maintenance of androgen responsiveness bj glucocorticoids in Shionogi mammary carcinoma cells in culture. J. Nat2 Cancer Insf. 80 (1988) 966-970.

Desmond Jr, W.J., Wolbers, S.J. and Sato, G.: Cloned mouse mammary cel: lines requiring androgens for growth in culture. Ce11 8 (1976) 79-86.

Plante, M., Lapointe, S. and Labrie, F.: Stimulatory effect of synthetic progestins currently used for the treatment of prostate cancer on pwthoi the androgen-sensitive Shionogi tumor in mice. J. Stemid Biochm. 3 1 (1988) 614. Bégin, D., Luthy, LA. and Labrie, F.: Adrenal premsor Cl9 steroids are potent stimulators of growth of androgen-sensitive mouse mammaq carcinoma Shionogi ceUs in vitro. Mol. W. Enhrinol. 58 (1988) 213-219.

Labrie, C., Bélanger, A. and Labrie, F.: Androgenic activity oi dehydroepiandrosterone and androstenedione in the rat ventral prostate EndoMnoIogy 123 (1988) 1412-1417.

Labrie, F.: Intracrinology. Mol. Cell. Endom'nol. 78 (1991) C113Cll8.

Bélanger, A., Brochu, M. and Cliche, J.: Levels of plasma steroic glucuronides in intact and castrated men with prostatic cancer. 1. Clin Endacrinal. Metab. 62 (1986) 812-815.

Bélanger, A., Caron, S. and Picard, V.: Shultaneous radioimmunoassay oi progestins, androgens, and esbogens in adult rat testis. J. Steroid BiochPm 13 (1980) 185-190.

Bélanger, B., Bélanger, A., Labrie, F., Dupont, A., Cusan, L. and Monfette G.: Comparison of residual C-19 steroids in plasma and prostatic tissue 01 human, rat and guinea pig after castration: unique importance oi extratesticular androgens in men. J. Steroid Biochm. 32 (1989) 695498. .

Bélanger, A., Couture, J., Caron, S. and Roy, R.: Determination of non conjugated and conjugated steroid level in plasma and prostate aftei separation of C-18 columns. In Steroid Formation, Degradaiion and Actiofi in Peripheral, Normal, and Neoplastic Tissue (Edited by H. Bradlow, L, Castagnetta, S. d'Aquino and F. Labrie). Ann. N.Y. Acad. Sci., (1990) 251- 259.

Thériault, C. and Labrie, F.: Multiple steroid rnetabolic pathways in ZR-75-1 human breast cancer cells. 1. Steroid Biuchem. Molec. Biol. 38 (1991) 155- 164.

Bradford, M.: A rapid and sensitive method for the quantitation oi rniaogram quantities of protein utilizing the prinaple of protein-dye binding. Anal. Biuch. 72 (1976) 248-254. Kramer, C.Y.: Extension of multiple range tests to group means with unique numbess of replications. Biomettics 12 (1956) 307-310.

Brochu, M., Bélanger, A-, Dupont, A., Cusan, L. and Labrie, F.: Effects of Flutamide and aminoglutethimide on plasma Sa-reduced steroid glucuronide concentrations in castrated patients with cancer of the prostate. 1. Stemid Biochenr. 28 (1987) 619622.

Brochu, M. and Bélanger, A.: Inaease in plasma steroid glucuronide levels in men fiom infancy to adulthood. 1. Clin. Endoctinol. 64 (1987) 1283-1287.

Bélanger, A., Candas, B., Dupont, A., Cusan, L., Diamond, P., Gomez, J.L. and Labrie, F.: Changes in serum concentrations of conjugated and unconjugated steroids in 40-yearold. J. Clin. Endocrhol. Metab. 79 (1994) 1086-1090.

Orentreich, N., BM~,J.L., Riter, RL. and Vogelman, J.H.:Age changes and sex differences in serum dehydroepiandrosterone sulfate concentrations throughout adulthood. J. Clin. Endoctinol. Meiab. 59 (1984) 552-555.

Gelles, J., Albert, J.D., Nachtsheom, D.A. and Loza, D.C.: Cornparison of prostatic cancer tissue dihydrotestosterone levels at the time of relapse foilowing orchiectomy or estrogen therapy. 1. Urol. 132 (1984) 693-696.

Moghissi, E., Ablan, F. and Horton, R.: Origin of plasma androstanediol glucuronide in men. 1. Clin. Endocrinol. Metab. 59 (1984) 417-421.

Marchetti, B. and Labrie, F.: Characteristics of Flutamide action on prostatic and testicular functions in the rat. 1. Steroid Biochem. 29 (1988) 691-698.

Labrie, F.: Endocrine therapy of prostate cancer: optimal form and timing. f. Clin. Endocrinol. Metab. 80 (1995) 10664071.

Labzie, F., Cusan, L., Gome& J.L.,Diamond, P. and Candas, B.: Combination of screening and preoperative endocrine therapy: the potential for an important deaease in prostate cancer mortality. 1. Clin. Endocrinol. Metub. (1995) in press. 47. Brooks, J.R, Berman, C., Nguyen, H., Prahalada, -S., Primka, EL. Rasmusson, G.H. and Slater, E.E.: Effect of castration, DES, flutamide, am the 5a-reductase inhibitor, MK-906, on the growth of the Dunning ra prostatic carcinoma, R-3327. The Prostate 18 (1991) 215-227.

48. Brooks, J.R, Baptista, E.M., Berman, C., Ham, E.A., Hichens, M., Johnson D.B., Primka, RL., Rasmusson, G.H., Reynolds, G.F., Mf,S.M. anc

&th, G.E.: Response of rat ventral prostate to a new and novel Sa1 reductase inhibitor. Endocrinology 109 (1981) û301836.

49. Asselin, J., tabrie, F., Gourdeau, Y., BoMe, C. and Raynaud, JP.:Binding o. [3H]rnethyltrienolone (R1881) in rat prostate and hurnan benign prostath hrpertrophy (BPH). Steroids 28 (1976) 449-459. Figure 1, Effect of 20-day treatment with finasteride (1 mg, twice daiiy) anc Flutamide (1 mg, twice daily) alone or in combination on growth of Shionog turnors in intact or castrated mice supplemented with an androstenedionc implant. The results shown were obtained on the Iast day of treatment, and thc product of the two longest perpendicular diameters is expressed as' mean f SEM *pe0.05, UpcO.Ol experimental vs intact or castrated control implanted with d43 dione, respectively

Figure 2. Effect of 20-day treatment with finasteride (1 mg, twice daily) anc Flutamide (1 mg, twice daiiy) alone or in combination on plasma (A) and tumoi (B) androstenedione (A(-dione) concentrations in intact or castrated animal! supplemented with an ~4-dioneimplant. Data are expressed as mean f SEM *p<0.05, **p

Figure 3. Effect of 20-day treatment with finasteride (1 mg, twice daily) anc Fiutamide (1 mg, twice daily) done or in combination on plasma (A) and tumoi (B) testosterone concentrations in intact or castrated animals supplemented wiü an androstenedione (ddione) implant. Data are expressed as mean ISEM

*pc0.05, **p<0.01experimental vs intact or castrated control implanted with ~44 dione, respectively.

Figure 4. Effect of 20-day treatment with finasteride (1 mg, twice daily) anc Flutamide (1 mg, twice daily) alone or in combination on plasma (A) and tumoi (B) DHT concentrations in intact or castrated animals supplemented with ar androstenedione (~4-dione)implant. Data are expressed as mean f SEM *peO.OS, **p

Figure 5. Effect of 2û-day treatment with finasteride and Fiutamide alone or ir combination on prostatic weight in intact or castiated mice bearing ar androstenedione (b4-dione) implant. Data are expressed as means f SPA *p<0.05, wpcO.Ol, experimental vs intact or castrated cantrol implanted with al- dione, respectively. Figure 6. Effect of 2May treatment with finasteride and Flutamide alone or ù combination on tumor (A) and prostatic (8) 5a-reductase activities in intact a castrated mice bearing an androstenedione (~Cdione)implants. Enzymati activity was measured by the formation of [14C]-DHT and its metabolites hoa P4C]-testosterone using NADPH as cofactor by hunor or prostate hornogenate! The results are expressed as fmol or pmol of product formed/mg protein/mir Data are expressed as mean f SEM. PcO.05, -p

4.5

4 A N s3.1 C W! 5a 824 J : 8 C L! P

O.! FIGURE 2 FIGURE 3 FIGURE 4 FIGURE 5

CHAPTERV

CONCLUSION The present thesis is composed of three major parts: the tissue distribution oi the enzymatic activities of steroidogenic enzymes; the importance of adrenk precursors in the formation of active androgens in the peripheral tissues; anc the developmnt of novel approaches to effiaently block the formation and th@ action of arrdmgeris.

First, we have studied the tissue distribution of the enzymatic activities oi steroidogenic enzymes required for the formation of the active androgens and estrogens in both male and female hamsters. The activities of steroid sulfatase 313-HSD, 17%HSD, Sa-reductase, and aromatase were measured in al1 the tissue! examined, namely adrenal, testis, ovary, kidney, liver, ventral and dorsal prostate, seminal veside, uterus, vagina, ventral and dorsal skin, ear, and flank organ. The results have demonstrated that, as with humans, the conversion 01 adrenal precursors into active androgens and/or estrogens can be wried out no1 only in the classical endocrine tissues such as adrenal, ovary, and testes, but alsc in the peripheral intracrine tissues such as skin and prostate in the hamster Xowever, hamsters like other experimental animals such as rats, mice, and guinea pigs, do not seaete large amounts of adrenal precursors. Also, it i: believed that the formation of androgens and estrogens takes place exclusivelq in the gonads in these animals.

In addition, we have studied the effed of the androgen precursors DHEA and ~4. dione by comparing them with active androgens T and DHT on androgen sensitive parameters in the castrated hamster. The steroids were released fiom silastic implants inserted in the hamster dorsal skin for 4 weeks. Data have shown that the serum androgen concentrations decreased to undetectable leveh after orchiectomy (ORCH), or they maintained the normal or highei concentration levels with simultaneous implantation of either DHEA, A~- &one, T, or DHT. Moreover, T and DHï stimulated the size of the flank organs, the size of underlying sebaceous glands, and the weight of prostate, as well as did adrenal precursors DHEA and d'diane. However, among these compounds, DHEA had the least potency. The pardlel data were observeci in the followiq study in which DHEA and ~4-dionewere administered on the surface area ol gonadectomized male and female hamster skin. After absorption, the compounds were converted into active androgens and estrogens in th peripheral tissues, where the stimulation of the flank organ size and prostai growth in the male and uterus growth in the female were observed. To searc for the mechanism involved, we have measured the steroidogenic enzymat activities in the skin, prostate, and uterus. The results have shown that ORCI increased steroidogenic enzymatic activities, namely 3/3-HSD, 17p-HSD, and 54 reductase in male hamster dorsal skin and prostate. On the other hanî administration of DHEA and AI-dione had an inhibitory effect except for 3&HS which was inaeased following DHEA treatment in the prostate. Moreove ORCH deaeased the aromatase activity in hamster dorsal skin and a furth decrease was obsemed following DHEA treatment. Li conhast, ~Cdionehad n significant effect on this parameter. Similar results were obtained in the ski and uterus in female hamsters except for the following two observations: th aromatase activity was increased following OVX; and DHEA reversed the effec of OVX in the skin. These data illustrate the importance of extragonadi peripheral intracrine formation of active steroids and suggest that the castrate hamster supplemented with adrenal precursor steroids is a good mode1 th; closely represents the human situation where adrenal steroids play a important role in androgen formation and action in peripheral tissues.

In addition to the above studies, we have investigated the inhibitory effects c Sa-reductase inhibitor finasteride and pure antiandrogen flutamide in th hamster. The two cornpounds were dissolved in EtOH/ propylene glycc (50%/50%) and administered localiy on the surface of the right fiank organs an ears of the treated group twice daily for 4 weeks, while the left flank organs an' eus, as well as the control group received the vehicle anly. In bath the rigk and left sides of the flank organs this treatment resulted in: similar reduction on the size of the sebaceous glands; mRNA levels of the androgen-regdatel FAR-17a gene; as well as PH]-thymidine incorporation. Moreover, bot* compounds also showed inhibitory effects on prostate and seminal vesid growth. The above data dearly demonstrated that both finasteride and flutamid exerted significant systemic effects after the topical administration, and thus ar unsuitable for topical treatment. in order to bearch an ideal compound as a topical antiandrogen, a series a compounds synthesized by our medicina1 chemistry division were tested i; vitro and in vivo models. Among them, EM-250, one of l7B-alkynyl substitutec T derivatives, inhibits the stimulatory effect of DHT on the growth of androgen sensitive Shionogi cells in vitro (Ki= 96 nM) at the concentrations comparabl to those of hydroxyflutamide (Ki= 60 nM). This campound had no inhibitor effect on the activities of human type I and II SP-HSD, human type 1 5a reduciase gene transfected celis, or 17i3-HSD and aromatase of human placent2 microsomes. ln vivo, administration of EM-2%) at the doses of 30,100, and 30 twice daily for 4 weeks on the Syrian hamster's right flank organs and ri@ ears si@cantly reduced the right flank organ area, underlying sebaceous glani volume, seboqte proliferation, and right ear sebaceous gland size in a dose dependent manner without affecting the left control side. The same treatmer did not influence the prostate weight and serum T and DHT concentration! Furthemore, systernic administration of EM-250 in the rats had no effect on th androgen sensitive parameters, namely PBP-mRNA levels, prostate weight, an( senim LH concentration while injection of EM-250 to OVX female rats did nc change the uterus weight. These data clearly showed that EM-250 is a pur topicai antiandrogen without any other hormonal and antihormonal effects.

We have also examined a series of 17f3-(N-alkyl/arylformamido)- and 17G-[(K ar kyl /ary 1) alky/arylamido]-4-methyl-4-aza-3~xo-5a-androstan-3~nesas 5a reductase inhibitors in the hamster flank organ in the saeening process. Amon, al1 the compounds, EM-401, EM-402, and EM-540 showed the most poten inhibitory effects on the flank organ. Fwther study was performed with topicé treatment of increasing doses of these three compounds on the right flan organs and right ears. EM-402 turned out to be a potent inhibitor of Sa reductase, which remarkably decreased the size of flank organs, the size a sebaceous glands, and the activities of Sa-reductase of right flank organs and ear after topical treatment without affecting the prostate and seminal vesicl weights while EM401 and EM-YLO exerted some systemic effects. Thus, the dat of EM-250 and EM-402 should encourage additional animal studies and result ii carefdly contded clinid trials in humans in which these products are appliec to skin disordm asdociated with excess andtogem.

We have next investigated the influence of tumor size on the responsiveness ti androgen deprivatiofi The animais bearing SC-115 tumors ranging from 0.1 ti 1.8 an in diameters were divided into different groups according to the tumo size and treated with flutamide f mg twice daily. The data have demonstratec that flutarnide inhibited tumor growth by 32% to 57% in the groups with smal tumors (0.1-1.0 cm at the day 0) while no significant inhibitory effect wai obse~edin large tumor groups (1.0-1.8 cm in diameters at the day O). In thc same experiments, flutamide decreased the rnouse kidney ODC activity to 1%O the intact control value and inhibited the prostate growth with no differencr among al1 of the groups. This demonstrates the high sensitivity of small tumori to androgen deprivation.

FinaUy, we have studied the effects of the treatment with flutamide anc finasteride alone or in combination (1 mg, twice daily) in Shionogi mice bearinl SC115 tumors. The inhibitory effects d flutamide and finasteride on thc Shionogi tumor and prostate growth are additive. The results frorn the abovc two experiments suggest that the combination of an antiandrogen and a Sa reductase inhibitor could provide the basis for further improvement of thc endocrine therapy of prostate cancer and that this treatment shodd be given a the earliest stage whenever possible. REFERENCES

Abalain, J., Quemener, E., Carre, J., Sinon, B., Amet, Y., Nangîn, P., and Fioch, H. (1988) J Steroid Biochem, 34: 467-470. Abd-Hajj, Y.J. (1975) Steroids, 26: 488-50. Adachi, K. (1974) Journal of Invest Dermatol, 62: 217-223. Adachi, K., and Kano, M. (1972b) Steroids, 19: 567-574. Adachi, K., Takayasu, S., and Takashima, 1. (1970) Soc. Cosmet. Chem., 21: 91-99. Adams, J. (1985) Mol. ceiI Endocrinol., 45: 1-17. Aksoy, I., Wood, T., and Weinshilboum, R. (1994) Biochem Biophys Res Commun, 200: 1621-1629. Anderson, KM.,and Liao, S. (1968) Nature, 219: 277-279. Andersson, S., Bergman, D.M.,Jenkuis, E.P., and Russel, D.W. (1991) Nature, 354: 159-161. Andersson, S., Bishop, R.W., and Russell, D.W. (1989) J. Biol. Chem., 261: 16249- 16255. Andersson, S., and Russel, D.W. (1990) Proceedings of the National Academy of Sciences of the USA, 87: 3640-3644. Andriole, G., Block, N., Boake, R., Cox, C., Crawford, D., Fradet, Y., Kadmon, J., Dedeernion, J., Lieber, M., Middleton, R., Patterson, J., Perreault, J., Rajfer, J., Ramsey, E., Schrodeer, F., Soloway, M., Smith, J., Trachtenberg, J., Ferguson, ]ND.,Gormley, GJ. (1994) J Ur01 151:435A, abstr. 832 Artega, C.L., and Osborne, C.K. (1991) In: Lippman, M.E. and Dickson, RB. (eds.), Regdatory mechanims in breast cancer, pp. 289 p. Boston: Kluwer Academic Press. Asselin, J., Labrie, F., Gourdeau, Y., Bonne, C., and Rapud, J.F. (1976) Sterhds, 28: 449459.

Auchair, C., Kelly, P., Cony, D., Schally, A, and Labrie, F. (1977) Endoqhol, 101: 1090-1893.

Azadian-Bodanger, G., and Bonne, C. (1974) J Steroid Biodiem, 5: 352. Bailey-Pridham, D., and Sanfilippo, J. (1989) Pediatr. ChNorth Am., 36: 581 599.

Baillie, A., Caban, K., and Milne, J. (1970) Br. Dermatol., 77: 610616. Baillie, AH., Thomson, J., and Milne, J.k(1966) Br. J. Dermatol., 78: 451457.

Bain, P.A., Yoo, M., Clark, T., Hamrnond, S.H., and Payne, A.H. (1991) Pro< Nati Acad. Sci. USA, 88: 8870-8874. Baird, B.T., Worton, R, Longcope, C., and Tait, J.F. (1969) Rec. Progr. Horm. Res 26: 611464. Baker, J., Bahan, G., Schuamcher, L, Roman, D., and Tharp, A. (1967) J Mec Chem) 10: 93-95 Baker, M. (1991) Steroids, 56: 354-360. Balasubramanian, A. (1976) Indian 1Biochem Biophys, 13: 325-330. Barrett-Connor, E., Khaw, KT.,and Yen, S.S.C. (1986) N. Engl. J. Med., 315: 1519 1524. Bartsch, W., Greeve, J., and Voigt, K. (1987) J Steroid Biochem, 28: 3W.

Bates, G. (1981) Pediatr. Clin. North Am., 28: 513-530. Baulieu, E., Corpechot, C., Dray, F., Emiliozzi, R., iebeau, M., Mauvais-Jarvis, P, and Robei, P. (1965) Recent Prog Hom Res, 21: 411. Baulieu, E.E. (1960) CR Acad. Sci. Paris, 251: 1421-1425.

Beato, M. (1987) Biochim. Biophys. Acta., 910: 95-102. Bégin, D., Luthy, I.A., and Labrie, F. (1988) Mol. Cd.Endocrinol., 58: 213219. Bélanger, A., Broch, M., and Clidie, J. (1986) J. Clin Endocrinol. Metab., 62 812-815. Bélanger, A., Candas, B., Dupont, A., Cusan, L., Diamond, P., Gomez, J.L., anc Labrie, F. (1994) J. ChEndocrinol. Metab., 79: 1-1090. Bélanger, A., Caron, S., and Picard, V. (1980) J. Steroid Biochem., 13: 185-190.

Manger, A., Couture, J., Caron, S., and Roy, R. (1990) In: Bradlow, H. Castagnetta, L., d'Aquino, S. and Labrie, F. (eds.), Steroid Formation Degradation and Action in Pezipheral, Normal, and Neoplastic Tissue, pp 251-259. Am.N.Y. Acad. Sci. Bélanger, A., LeGoff, J.M.,Proulx, L., Caron, S, and Labrio, F. (1985) Cancer Ra 45: 62934295.

Bélanger, B., Bélanger, A., Labrie, F., Dupont, A., Cusan, L., and Monktte, ( (1989) J. Steroid Biochem., 32: 695-698. Berg, O. (1958a) Acta Endocrinol, 129. Berg, O. (1958b) Acta Endocrinol, 27: 140. Bernier, F., Lopez-Solache, I., Labrie, F., and Luu-The, V. (1994) Mol. Ce1 EndOcnn01., 99:

Mes,P., Edman, C., and Karavolas, H. (2984) J Bi01 Chem, 259: 107-111. Bingham, ICD., and Shaw, D.A. (1973) J. EndOctinoI., 57: 111-121. Bloch, B. (1931) Br. J. Dermatol., 43: 61. Bloh, T., Metcalf, B., Laughlin, M., Sjoerdsma, A., and Schatzman, G. (1981 Biochem Biophys Res Commun, 95: 273-280. BoneUi, M., and Alessi, E. (1970) Semana Médica, 136: 1052-1063. Boring, C.C., Squire, T.S., and Tong, T. (1993) CA Cancer J. Clin., 43: 7-26. Bouton, MM.,Lecaque, D., Secchi, J., and Tournemine, C. (1986) J. Inves Dermatol., 86: 163467. Bradford, M. (1976) Anal. Biochem., 72: 248-254. Bringham, K., and Shaw, D. (1973) J Endoainol, 57: 111-121. Brochu, M., and Bélanger, A. (1987) J. Clin. Endocrinol., 64: 1283-1287. Brodiu, M,Bélanger, A., Dupont, A., Cusan, L., and hbrie, F. (1987) J. Steroi Biochem., 28: 619-622. Boris, A. (1965) Endoainol, 76: 1062-1067. Boris, A., and Stevenson, R. (1966) Endocrinol, 78: 549-555. Brooks, J.R. Baptîsta, E.M., Berman, C., Ham, E.A., Hidrens, M., Johnson, D.E Primlsa, RL., Rasmusson, G.H., Reynolds, G.F., Schmiff, S.M., and Arth, G.1 (1981) Endocridogy, 109: 830-836. Brooks, J.R, Berman, C., Nguyen, H., Prahahda, S., Primka, R.L, Rasmusso~ G.H, and Slater, E.E. (1991a) The Rostate, 18: 215227. Brooks, J.R., Primka, R., Berman, C., Krupa, D.A., Reynolds, G.F., and Rasmusson, GX. (1991b) Steroids, 56: 42û-433. Brown, M.S., Kovanen, P.T., and Goldstein, J.L. (1979) Rec Progr Horm Res, 35: 215-257.

Bruchuvsky, N. (1972) Biochem. J., 127: 561-565. Bruchovsky, N., and Meakin, J. (1973a) Cancer Res, 33: 1689-1695. Bmchovsky, N., and Meakin, J.W. (1973b) Cancer Res., 33: 1689-1695. Bruchovsky, N., and Wilson, J. (1968) J: Biol. Chem., 243: 2012-2021. Buckley, D., and Ramachandran, J. (1981) Proc Nati Acad Sci USA, 78: 7431-7435. Bdlock, L., and Bardin, C. (1970) J Clin Endocrin01 Metab, 31: 113-117.

Burdick, K.H, and Hill, R. (1970) Br. J. Dermatol., 82 (Suppl. 6): 19-25. Burton, J.L., Johnson, G., Libman, S., and Shuster, S. (1972) J. Endoîiinol., 53: 349-354.

Burton, J.L., laschet, U., and Shuster, S. (1973) Br. J. Dermatol., 89: 487490.

Cameron, E.H., Baillie, AH., Grant, J.K, Milne, J.A., and Thomson, J. (1966: Transformation in vifro of [7a-3H] dehydroepiandrosterone to [3H] testosterone by skin from men. Proceedings of the lOlst Meeting of th6 Society for Endocrinology, London, p. 19 (abst.). Carpenter, T., hperato-McGiniey, J., Bouiware, S., and et al. (1990) J Ch Endocrinol Metab, 71: 318. Cella, J., and Tweit, R (1959) J Org Chem, 24: 1109-1110. Cella, J., and Tweit, R (1961) U.S. Patent 3,013,012,

Chamberlain, J., Jagarinec, N., and Oh,P. (1966) Biochem j, 99: 616.

Chen, C., Puy, L., Simard, J., Li, X., Singh, S., and Librie, F. (1995) J. Ivest. Derma tol., 105:678-682. Cheng, Y.C., and Prusoff, W.H. (1973) Biachem. Pharmacol., 22: 343993108. Chishoim, G.D. (1981) Rec Res. Cancer Re., 78.17Hû4. Chrousos, G., Loriaux, L, Mann, D., and Cutler, G.J. (1982) Am. intem. Med. 96: 143448. Chu, J., and Kimura, T. (1973) J Bi01 Chem, 248: 2089-2094. Coleman, D.L., Schwizer, RW., and Leiter, E.H. (1983) Diabetes, 33: 26-33. Corner, Ki, Fdany, J.,and Falany, C. (1993) Biochem J0 289: 233-240. Connolly, P., and Resko, J. (1989) J Steroid Biochem, 33: 1013-1018. Couture, P., Thériault, CC,Simard, J., and Labrie, F. (1993) Endoahology, 132 179-185. Covey, D., and Robinson, C. (1976) J Am Chem Soc, 98: 503û-5040. Covol, P., Michaud, A., Mevard, J., Freifeld, M., and Mahoudeau, J. (1975 Endocrinol, 97: 52-58. Crawford, D., Eisenberger, MA-,McLmod, D.G.,Spaulding, J.T., Benson, R, Dm F-A., Blumenstein, D.A., Davis, M.A., and Goodman, P.J. (1989) New Engl J. Med., 321: 419424. Cunliffe, W., Clayden, A., Godd, D., and Simpson, N. (1981) Clin. Experimeta Dermatol., 6: 461-469. Cunliffe, W.J.,and Shuster, S. (1969b) The Lancet, 1: 685-687.

Cunliffe, W.J., Shuster, S., and Cassels-Smith, A.J. (1969) British Journal O Dennatology, 81: 200-201. Cunliffe, W.J., and Simpson, N.B. (1980) In: Mauvais-Jarvis, P. (eds.) Percutaneous Absorption of Steroids, pp. 149-154. New York: Academh press. Cusm, L., Dupont, A., Manger, A., Tremblay, R.R, Manhès, G., and Labrie, F (1990) J. Am. Acad. Dermatol., 23.462469. Cusan, L., Dupont, A., Cossette, M., and Labrie, F. (1993) Can. J. Dermatol., 5 42147. Cusan, L., Dupont, A., Gomez, J.L., Tremblay, RR, and Labrie, F. (1994) Fertil Steril., in press. Darley, C.R, Kirby, J.D., Besser, G.M.,Munro, D.D., Edwards, CRW., and Rees LH. (1982) Br. J. Dermatol., 106:517-522. De Launoit, Y., Simard, J., Durocher? F., and Labrie, F. (1992) Endocrhology, 130 553-555. De Schdpper, P., Imperato-McGinley, J., Hedken, A., De Lepeleire, L, Buntinx, A. Carlin, J., Gressi, M., and Stoner, E. (1991) Steroids, 56: 469-471. Debeljuk, L., Atimwa, A., and Schaily, A. (1972) Endocrinol, 90: 157û-1581. Demyan, W., Song, C., Kim, D., Her, S., Gall-Witz, W., Rao, T., Slomczynska, M. Chatte*, B., and Roy, A. (1992) Mol Endocrinol, 6: 589697. Denis, L., Carneiro de Moura, J.L., Bono, A., Sylvester, R., Wheeton, R. NewIing, D., and Pauno, M.D. (1993) Urulogy, 42: 119-129. Depre, M., Meeter, C., Hecken, A., van, L., OL, Buntinx, A., and Schepper, P.d (1992) Qin Pharmacol Therap, 52: 409412. Desmond Jr, W.J., Wolbers, S.J., and Sato, G. (1976) Ceil, 8: 79-86. Deutsch, S., Benjamin, F., SeItzer, V., Tafreshi, M., Kocheril, G., and Frank, A (1987) Int. J. Gynecol. Obstet., 25: 217-222. Dexter, D.L., and Cdabresi, P. (1982) Biochem. Biophys. Acta, 695: 97-112. Diabbelt, L., and Kuss, E. (1991) Bi01 Chem Hoppe-Seyler, 372: 173-185. Diani, A., Muiholland, M., Shull, K., Kubicek, M., Johnson, G., Schostarez, H. Brunden, M., and Buhl, A. (1992) J. Clin. Endocrinol. Metab., 74: 345-350. Djoseland, O., Bruchovsky, N., Rennis, P., Otal, N., and Hoglo, S. (1983) Acti Endocrinol, (Copenh) 8: 273-281. Downing, D.T., Strauss, J.S., Ramasastry, P., Abel, M., C.W., L., and Pochi, P.E (1975) J hvest Dermatol, 64: 215-219. DriscoU, W., Martin, B., Chen, H., and Strott, C. (1993) J Bi01 Chem, 268: 23496 23503. Drucker, W., Blumberg, J., Gandy, H., David, R, and Verde, A. (1972) Steroids 18: 11.

Dumont, M., Luu-The, V., Dupont, E., Pelletier, G., and Cabrie, F. (1992) J Invest. Dermatol., 99: 415421. Dunn, J., Goldstein, J., and Wilson, J. (1973) J Bi01 Chem, 248: 7819-7825. Dupuy, G., KD, R, Bleau, G., and Chapdelain, A (1978) J Steroid Biochem, 9 1043-1047- Durocher, F., Sanchez, R, Laudet, V., Labrie, Y., Samson, C., Tremblay, Y., Labric F., and Simard, J. (1994) Struchual and functional characterization of th guinea pig 3$-hydroxysteroid dehydrogenase/~5-~4isomerase expressed i the adrenal gland, and gonads. Ptoc. 76th Ann. Meet. Endocrine Society, 1 Abst. No 1192, p. 498. Ebling, F. J. (1948) J. Endoclinol., 5: 297-302. Ebhg, F.J. (1963) ln: Advances in Biology of Skin. The Sebaceous Glands, pl 200-219. Oxford: Pergamon Press. Ebling, F.J. (1973) Acta Endocrinol. (Copenh), 72: 361-365. Ebling, F.J. (1974) J hvest Dermatol, 62: 161-171. Ebling, F.J., RandaU, V.A., and Skinner, J. (1981) J. Invest. Dermatol., 458-4G Edwards, E., Hamilton, J., Duntley, S., and Hubert, G. (1941) Endoainol, 29: 852 853. Ehrmam, D.A., and Rosenfield, R. (1990) J. Clin, Endoainol. Metab., 71: 1-4. Eisenberger, M.A., Crawford, D.E., Wolf, M., Blumenstein, B., McLeod, D.G Benson, R., Dorr, F.A., Benson, M., and Spaulding, J.T. (1994) Semir Oncol., 21: 613-619. Ellis, R.A. (1958) ln: Montagna, W. and Ellis, R.A. (eds.), The Biology of Haî Growth, pp. 469-485.New York: Academic Press. Emanuel, S. (1936) Acta Derm Venereol, 17: 444-456. Eppenberger, U., and Hsia, S. (1972) J Bi01 Chem, 247: 5463-5469. Erickson, G., Magoffin, D., Dyer, C., and Hofeditz, C. (1985) Endocr Rev, 6: 371 399. Eriksson, L., Nilsson, B., carlstrom, K, Oscarsson, J., Eden, S., and Von Schoulti B. (1989) J steroid Biochem, 33: 413416. Ewald, W., Werbin, H., and Chaikoff, 1. (1965) Biochim Biophys Acta, 111: 306 312. Ewing, L, Berry, S., and Higginbottom, E. (1983) Endoainol, 113: 2004. Falany, C., Vazquez, M., and Kalb, J. (1989) Biochem J, 260: 641446. Fansworth, W.,and Brown, J. (1963) J Am. Med. Assoc., 183: 436439. Farsworth, WE., Brown, JR.(1963) J AM Med Assoc 183: 436-439 Fazekas, A., and Lanthier, A. (1971) Steroids, 18: 367-379. Fazekas, A., and Sandor, T. (1973) Froc Intl Congr Endodocrinol 4th Washington, D.C., p.80. Fearon, E.R, and Vogelstein, B. (1990) CeU, 61: 759-767. Feinberg, A.P., and Vogelstein, B. (1983) Anal. Biochem., 132: 6-13. Ferreri, R, Chakrabarty, K., Beyler, A., and Wiland, J. (1978) J Invat Dmnatol, 71: 320-323. Fichman, L.F., Nyberg, L.M.,Bujnovsky, P. (1981) J. Clin. Endocrinol. Metab., 52: 919-923. Freinkel, R.K. (1969) New Engl. J. Med, 280: 1161-1163. French, F., and Ritzén, E. (1973) Endocrinol, 93: 88-95.. Frost, P., Giegel, J.L., Weinstein, G.D., and Gomez, E.C. (1973) J Invest Dermatol, 61: 159-167.

Frost, P., and Gomez, E. (1972) Adv. Bi01 Skin, 12: 403-413. Frost, P., Reed, M., and James, V. (1980) J Steoid Biochem, 13: 1427-1431.

Futterweit, W., Dunaif, A., Yeh, H.C., and Kingsley, P. (1988) J. Am. Acad, Dermatol., 19: 831-836. Garren, L., GU, G., Masui, H., and Walton, G. (1971) Rec Progr Hom Res, 27: 433-478. Gehring, U., Tomkins, G., and Ohno, S. (1971) Nature (London), New Biol, 232: 106-107. Geller, J., Albert, J., Lopez, D., Geller, S., and Nishiyama, G. (1976) J Clin Endocrinol Metab, 43: Geller, J., Albert, J.D., Nachtsheom, D.A., and Loza, D.C. (1984) J. Urol., 132: 693 696. Gelier, J., Fishman, J., and Cantor, T. (1975) J Steroid Biochem, 6: 837443- Geîler, J., and Franson, A. (1989) Effea of MK-906 on prostate tissue androgem and prostate specific antigen. 71th annuai meeting of the endocrine Society, Srattle, Washington, p. Abstract No 1640. Giolgi, E., Stewart, J., Grant, J., and Scott, R (1971) BiochemJ., 123: 41-55. Glashan, RW., and Robinson, M.RG. (1981) Br. J. Urol., 53: 624-650. Glenn, E.M., and Gray, J. (1965) Endocrinol, 76: 1115-1123. Gloyna, R, Siiteeri, P., and Wilson, J. (1970) J. Clin. Invest., 49: 174901753.

Goff, J., and Beianger, B. (1984) Steroids, 44: 207-216. Gomez, E.C., Liewellyn, A., and Frost, P. (1974) J. Invest. Dermatol., 63: 383-387. Gmdfellow, A., Alaghband-Zadeh, J., Carter, G., Cream, J.J., Holland, S., J.Scullj and Wise, P. (19û4) Br. J. Dermatol., 111: 209-214. Gormley, G., Stoner, E., Rithaster, R., Gregg, H., Thompson, D., Lasseter, K Vaisses, P., and EA, S. (1990) J Clin Endocrinol Metab, 70: 1136-1141. Gormley, GJ.(1991) Ur01 Clin North Am, 18: 93-98 Goss, G.M.A.A.,Wirtz, P., and Vormorken, H. J.M.(1982) Dermatol. Res., 272 333-341.

Greenwood, R., Brummitt, L., Burke, B., and Cunliffe, W.J.(1985) Br. Med. J 291: 1231-1235. Grino, P., Griffin, J., and Wilson, J. (1990) Endocrinol, 126: 1165-1172. Gwynne, J.T., and Strauss, 1. (1982) Endocr Rev, 3: 299-329. Halkerston, I., Eichhom, J., and Hechter, 0. (1961) J Bi01 Chem, 236: 374-380. Hamilton, J. (1942) Am J Anat, 71: 451-480.

Hamilton, J. (1951) AM. NY Acad. Sci., 53: 712. Hamilton, J., and Mestler, G. (1963) Proc Soc Exp Bi01 Med, 112: 374-378. Hamilton, J.B., and Montagna, W. (1950) Am. J. Anat., 86: 191-233. Hanunerstein, J., Meckies, J., bRossberg, I., Molb, L., and Zielske, F. (1975) Steroid Biochem, 6: 827-836. Hammond, M., Talbert, L., and Groff, T. (1986) Postgrad. Med., 79: 107-113. Haning, RJ., nâckett, R, Boothroid, R, and Canick, J. (1990) J Steroid Biochem 36.175179. Hansson, V., Trgstand, O., French, F., McLean, W., Smith, A., Tindall, D., Weddington, S., Petrusz, P., and Nayfeh, S., (1974) Nature, 250: 387-391. Hansson, V., Tveter, K., Unhjem, O., and Djoseland, 0. (1972) J Steroid Biochem, 3: 427-431.

Harper, M., Pike, A., Peeiing, W., and Griffiths, K. (1974) J. Endocrinol., 60: 117. 125. Harris, G., Azzolina, B., Baginsky, W., Gimis, G., Rasmusson, G., Tolman, R, Raetz, L., and Ellsworth, K. (1992) Proc. Natl. Acad. Sei. U.S.A., 89: 10787. 10791. Hay, J.B., and Hocigins, M.B. (1978) J. Endocrinol., 79: 29-39. Hechter, O., Solornon, M.M., Zaffaroni, A., and Pincus, G. (1953) Arch Biochem Biophys, 46: 201-214. Herkner, K., Swoboda, W., Hoeller, B., and Goedl, U. (1986) J Steroid Biochem, 24: 239-243. Hirsch, K., Jones, C., Audia, J., Andersson, S., McQuaid, L., Stamm, N., Neubauer, B., Penning, P., Toomey, R, and Russell, D. (1993) Proc Nali Acad Sci, 90: 5277-5281. Hisaoka, H., Idota, R., Seki, T., and Adachi, K. (1991) Arch Dermatol Res, 283: 269-273.

Hobkrik, R (1985) Can J Biochem Cell Biol, 63: 1127-1144. Horton, R, (1984) J Am Geriatr, 32: 380. Horton, R., Hsieh, P., Barberia, J., Pages, L., and Cosgrove, M. (1975) J Ciin Endocrinol Metab, 41: 793-796. Houston, B., Chisholm, G., and Habib, F. (1985) J Steroid Biochem, 22: 461467. Houston, B., Chisholm, G., and Habib, F. (1987) Steroids, 49: 355-369. Hsueh, A., Adashi, E., Jones, P., and Weish, J.T. (1984) Endocr Res, 5: 76-127. Hudson, R (1987) J Steroid Biochem, 26: 349-353. Huggins, C., and Hodges, C.V. (1941) Cancer Res., 1: 293-307. ImperateMcCinley, J*, Guerrero, L., Gautier, T., and Peterscin, R (1974) Science, 186: 1213-1215. Imperato-McGinley, J., Peterson, R, Gautier, T., and SturIa, E. (1979) J. Steroj Biochem., 11: 637-645. Imperato-Mcginiey, J., Sanchez, R, and Spencer, J. (1992) Endocrinol., 131: 114 1156.

Imperato~McGinley,J., Shackleton, Ce0 Orlic, S., and Stoner, E. (1990) J. Cli: Endocrinol. Metab., 70: 777-782. hnperato-MGinley, J., Shackîeton, c., and Stoner, E. (1989) Cornparison phsrna and urinary Cl9 and C215mreductase metabolites in subjects treate with the 5a-reductase deficiency. Proc 71st AM Meet Endw Soc, p. P33 Abst 1639, lsaacs, J. (1990) Prostate, 3 (suppl): 1-7. Issacs, J., Brebdler, C., and Walsh, P. (1983) J Clin Endocrinol Metab, 56: 139-146$ Jagarinec, N., Chamberlain, J., Ofner, P.., (1967) Biochim Biophys Acta, 144: 47! 481 Janknegt, RA., Abbou, C.C., Bartoletti, R., Bernstein-Hahn, L., Bracken, E Brisset, J.M., Silva, F.C.d., KnonageI, H., and Vemer, P. (1993) J. Urolog 149: 77-83. Jarrett, A. (1955) Br J Dermatol, 67: 166-179. Jenkins, E.P., Hsieh, C.L., Milatovich, A., Normington, K., Berman, D.M Francke, U., and Russel, D.W. (1991) Genomics, 11: 1102-1112. Jones, C., Audia, J., Lawhorn, D., 'LA, M., Neubauer, B.# Pike, A., Pennington, E Stamm, N., Toomey, R, and Hirsch, K. (1993) J Med Chem, 36: 42143. Josko, J., Fajkos, J., and Som, F. (1960) Collect Czech Chem Commun, 25: 108( 1090. Jungblut, P., Hughes, S., Gorlich, Lm0 Gowers, U., and Wanger, R (1971) Hoppc Seyler's Z Physiol Chem, 352: 1603-1608.

Kadohama, N., Karr, C., Murphy, G., and Sandberg, A. (1984) Cancer Res, 44 49474954.

Kamiyama, H., Hirato, K., Akamatsu, T., and al, e. (1992) Nippon Sank Fujinka Gakkai Zasshi, 44: 419-426. Kassenaar, A., Querido, A., and Schaler, H. (1960) Acta Endoaino1 (Suppl), 5: 859. Katchen, B., Dancik, S., and Millington, G. (1975) Clin Endoaino1 Metab, 41: 373-379. Katchen, B., Dandk, S., and Millington, G. (1976) J Invest Dermatol, 66: 379-382. Kato, J., Atsumi, Y., and kiaba, M. (1974) EndocrinoI, 94: 309-317.

Kato, Je, and Onouchi, T. (1973) Endocriml Japon, 20: 429-432.

Kaufman, M., Pinsky, L., Trifiro, M., Lumbroso, S., Sabbaghian, N., and Gottlieb, B. (1993) J Steroid Biochem. Molec. Biol., 45: 467476. Kawano, J., Kotani, Te, Ohtaki, S., Minamino, N., Matsuo, He,Oinuma, Te, and Aütawa, E. (1989) Biochem Biophys Acta, 997: 199-205. Keahey, N., Matbinez, L., Blasco, K.J., and McGinley Leyden, J.J. (1983) J. Invest. Dermatol., 80: 359 (Abst). Kent, S. (1982) Geriatrics, 37: 157-161. Kerr, J.F., y,A., d Currie, A.R (1972) Br. J. Cancer, 26: 239-257. Kitamura, Y., Okamoto, S., Uchida, N., Yamaguchi, K., and Matsumoto, K. (1978) Cancer Res., 38: 47114716. Klein, Ho,Molwitz, Te,and Bartsch, W. (1989) J Steroid Biochem, 33: 195200. Kniewald, Z., Massa, R., and Martini, L. (1969) Excepta Med, Sect 3 23: 59-63. Kochakian, Ce,and Stidworthy, G. (1954) J. Biol. Chem., 210: 933-939. Kochakian, C., Arimasa, N. (1976) Handbuch der Experemtellen Pharmakologie 43: 287-359 Kramer, C.Y. (1956) Biometrics, 12: 307-310.

Kneger, N.R, Scott, RG., and Jurman, M.G. (1983) Je Neurochem., 40: 1460-1461. Krozowski, 2. (1992) Molec Cellu Endocrinol, 84: C25€31. Kupperman, H. (1944) Anat. Rec., 88: 26. Kutten, Fe,Mowszowicz, L, Wright, Fe, Baudot, N., Jaffiol, C., Robin, M., and Mauvais-Jawis, P. (1979) J Clin Endodnol, 49: 861-865.

Kuttenn, F, Mowszowiu, L, Schaison, G., and Mauvais-Jarvis, P. (1977) J. Endoainol., 75: 83-91. La Vecchia, C., Franceschi, S., Decarli, A., Gallus, G., and Tognoni, G. (1984) Natl. Cancer Inst., 73: 667-671. Labrie, C., Manger, A., and Labrie, F. (1988a) Endocrinology, 123: 14124417. Labrie, C., Cusan, L., Plante, M., Lapointe, S., and Labrie, F. (1987) J. Steroi Biochem., 28: 379-384. Labrie, C., Cusan, L., Plante, M., Lapointe, S., and Labrie, F. (1989a) J. Steroi Biochen, 28: 379-384. Labrie, C., Simard, J., Zhao, H.F., Bélanger, A., Pelletier, G., and Labrie, F. (1989k Endocrinology, 124: 2745-2754. Labrie, C., Simard, J., Zhao, H.F.,Pelletier, G., and Labrie, F. (1990a) Mol. Ce1 Endocrinol., 68: 169-179. Labrie, C., Trudel, C., Li, S., Martel, C., Couet, J., and Labrie, F, (1993 Endocrinology, 129: 566-568. Labrie, F. (1991a) Endoainol. Metab. Ch. North Am, 20: 845-872. Labrie, F. (1991b) Mol. Celi. Endocrinol., 78: C113-C118. Labrie, F. (1993) Cancer, 72: 38163827. Labrie, F. (1995) J. Clin. Endoainol. Metab., 80.1066-1071. Labrie, F., Bélanger, A., Cusan, L., Séguin, C., Pelletier, G., Keliy, P.A., Reeves, J.J Lefebvre, F.A., Lemay, A., and Ranaud, J.P. (1980) J. Androl., 1: 209-228. Labrie, F., Bélanger, A., Dupont, A., Luu-The, V., Simard, J., and Labrie, C (1993a) Clin. Invest. Med., 16: 487-504. Labrie, F., Bélanger, A., Simard, J-, Labrie, C., and Dupont, A. (1993b) Cancer, 73 1059-1067. . Labrie, F., Cusan, L., Gomez, J.L., Diamond, P., and Candas, B. (1995) J. CLir Endocrinol. Metab., 80: 2002-2013. Labrie, F., Cusan, L., Gomez, J.L., Diamond, P., Suburu, R., Lemay, M., Tetu, B Fradet, Y., and Candas, B. (1994) Urology, 44: 29-37. Labrie, F., Dupont, A, and Bélanger, A. (1985) In: de Vita, V.T., Hellman, S. am Rosenberg, S.A. (eds.), Important Advances in Oncology, pp. 193-211 Philadelphia: J.B. Lippincott. Labrie, F., Dupont, A., Bélanger, A., Cusan, L., Lacourcière, Y., Monfette, G., Laberge, J.G., Emond, J., Fazekas, A.T.A., Raynaud, J.P., and Husson, J.M. (1982) J ChInvest Med, 5: 267-275. Labrie, F., Dupont, A., Bélanger, A., Poyet, P., Giguere, M., Lacoura&re,Y., Emond, J., Monfette, G., and Borsanyi, J.P. (19ffi) Lancet, 49. Labrie, F., Dupont, A., Bélanger, A., St-Arnaud, R, Giguere, M., Lacoura&re,Y., Emond, J., and Monfette, G. (1986b) Endocr. Rev., 7; 67-74. Labrie, F, Dupont, A., Cusan, L., Gomez, J., Emond, J., and Monfette, G. (1990b) J. Steroid Biochem. Mol. Biol., 37: 943-950. Labrie, P., Dupont, A., Cusan, L, Cornez, J.L., and Diamond, P. (1993~) Ch. Invest Med., 16: 493-498. Labrie, F., Dupont, A., Cusan, L., Cornez, J.L., Diamond, P., Koutsilieris, M., Suburu, R., Fradet, Y., Lemay, M., Têtu, B., Emond, J., and Candas, B. (1993d) Clin. Invest. Med., 16: 511-521.

Labrie, F., Dupont, A., Simard, J., Luu-The, V., and Bélanger, A. (1993e) Eus. Urol., 24 (Suppl. 2): 94-105. Labrie, F., Simard, J., Luu-The, V., Belanger, A., and -Pelletier, G. (1992a) J. Steroid Biochem. Mol. Biol., 43: 805-826. Labrie, F., Simard, J., Luu-The, V., Pelletier, G., Bélanger, A., Ladiance, Y., Zhao, H.F., Labrie, C., Breton, N., de Launoit, Y., Dumont, M., Dupont, E., RMaume, E., Martel, C., Couet, J., and Trudel, C. (1992b) J. Steroid Biochem. Mol. Biol., 41: 421-435.

Labrie, F., Sugimoto, Y., Luu-The, V., Simard, J., Lachance, Y., Badivarov, D., Leblanc, G., Durocher, F., and Paquet, N. (1992~)Endocrinology, 131: 1571- 1573. Labrie, F., Sugimoto, Y., Luu-The, V., Simard, J., Lachance, Y., Bachvarov, D., Leblanc, G., Durocher, F., and Paquet, N. (1992d) Endoainology, 131: 1571- 1573. Labrie, F., and Veilleux, R (1986) The Prostate, 8: 293-300. Labrie, F., and Veillew, R (1988) J. Natl. Cancer Inst., 80: 966-970. Labrie, F., Veillew, R, and Fournier, A (1988b) Mol. Celi. Endoctinol., 58: 207- 211. Labrie, F., Veillew, R, and Fournier, A. (1988~) J. Natl. Cancer bt., 80: 1138- 1147. Labrie, Y., Duracher, F., Lachance, Y., Leblanc, G., Leblanc, J.F., Simard, J., an Labrie, F. (1994~)Submitted. Lachance, Y., Luu-The, V., Labrie, C., Simard, J., Dumont, M., Launoit, Y.d Gukrin, S., Leblanc, G., and Labrie, F. (1990) J. Biol. Chem., 265: 20469-20475. Lachance, Y., Luu-The,'V., Verreault, H., Dumont, M., Rhéaume, E., Leblanc, G and Labrie, F. (1991) DNA Cell Biol, 10: 701-711. Lacoste, D, Manger, A., and Labrie, F. (1990) In: Bradlow, EL, Castagnetta, L d'Aquino, S. and Labrie, F. (eds.), Steroid Formation, Degradation, an1 Action in Peripheral, Normal, and Neoplastic Tissue, pp. 389-391. Ann. NA Acad. Sci. Lakshmi, S., and Balasubramanian, A. (1981) J Neurodiem, 37: 358-362. Lamb, J., English, H., Levandoski, P., Rhodes, G., Johnson, R., and Isaacs, : (19%) Endoainol, 13: 685-694- Lamb, J., kvy, M., Johnson, R, and Isaacs, J. (1992b) Prostate, 21: 15-34. Lephart, E., Simpson, E., and Trzeeiak, W. (1991) J Mol Endoc~ol,6: 163-170. Lerner, L., Bianchi, A., and Borman, A. (1960) Proc Soc Exp Bi01 Med, 103: 171 175. Leshin, M., and Wilson, J. (1982) J Steroid Biochem, 17: 245-250.

Leung, A., and Kiefer, G. (1989) J. Natl. Med. Assoc., 81: 65-67. Levin, S. (1965) J Med Chem, 8: 537-539. Levin, S., and Diassi, P. (1965) J Org Chem, 30: 1325-1327. Levy, H., and Maloney, P. (1961) Biochem Biophys Acta, 57: 149-150. Levy, M., Brandt, M., and Greway, A. (1990) Biochem., 29: 2808-2815. Levy, M., Metcalf, B., Brandt, M., Erb, J., Oh, H., Heaship, J., Yen, H., Rozamus, L and HoIt, D. (1991) Bioorgan Chem, 19: 245-260.

Li, X., Shgh, S., and fibrie, F. (1995) J. Med. Chem., 38: 1158-1173. Liang, T., Cascieri, M., Cheung, A., Reynolds, G., and Rasmusson, G. (1985 Endodnol, 117: 571-597. Liang, T., Heieirs, C., Cheung. A., Reynolds, G, and Rasmusson, G. (1984) J Bic Chem, 259: 734-739. Liang, T., and Hesis, CI (1981) J Bi01 Chem, 256: 799û-ûûû5. Liang, T., Ramusson, G., and Brooks, J. (1983) J. Steroid .Biochem., 19: 385-390. Liao, S., Lin, A., and Tymodco, J. (1971) Biodiim Biophys Acta, 230: 535-538. Lisboa, BP., Drosse, I., Breuer, H. (1965) Z Physio Chem 342: 123-131 Lloyd, J., JA, T., and Mawhinney, M. (1975) hvest Urol, 13: 220. Lookingbill, D.P., Abrams, B.B., Ellis, C.N., Jegasothy, B.V., Lucky, A.W., Urtiz. Ferrer, L.C., Savin, R.C., Shupack, J.L., Stiller, M.J., Zone, J., Landis, J.R Ramaswamy, R, Cherill, RJ., and Pochi, P.E. (1992) Arch. Dermatol., 128 1197-1200. Lorence, M., Muny, B., Trant, J., and Mason, J. (1990a) Endocrinol, 126: 2493 2498. Lorence, MC.,Murry, B.A., Trant, J.M., and Mason, J.I. (2990b) Endcxrinology, 126: 2493-2498.

Luderschmidt, C., Bidlingmaier, F., and Plewig, G. (1982) J. Invest. Dermatol., 78: 253-255. Luderschmidt, C., Eiermam, W., Jawny, J., Bidlingmaier, F., and Ring, J. (1984a; Nauynyn Sduniedebergs &ch. Pharmacol., 328: 214-218. Luderschmidt, C., Hoffman, K., and Bidlingmaier, F. (1984b) J Invest Dermatol, 83: 157-160. Ludwig, E. (19n) Br. J. Dermatol., 97: 247-254. Lunde, O., and Djseland, 0. (1987) J. Steroid Biochem., 28: 161-165. Luthy, I., Begin, D., and Labrie, F. (1988) J. Steroid Biochem., 31: 845462.

Lutsky, B.N., Budak, M., Koziol, P., Monahan, M., and Neri, R.O. (1975) J, Invest. Dermatol., 64: 412-417. Luu-The, V., Labrie, C., Simard, J., Lachance, Y., Zhao, H.F., Couët, J., Lablanc, G., and Labrie, F. (1990a) Mol. Endocrinol., 4: 268-275. Luu-The, V., Labrie, C., Zhao, H.F., Couet, J., Lachance, Y., Simard, J., CM, J., ïeblanc, J., Lagac4, L, Bérubé, O., Gagné, R, and iabrie, F. (1990b) AM. N.Y. Acad. Sci., 595: 40-52. Luu-The, V., Labrie, C., Zhao, H.F., Couët, J., Lachance, Y., Simard, J., Leblanc, G CM, J., Bérubé, D., Gagné, R, and Labrie, F. (1989a) Mol. Endocrhol., I 1301-1309. Luu-The, V., Lachance, Y., Labrie, C., Leblanc, G., Thomas, J.L., Strickler, RC and Labrie, F. (1989b) Mol. Endoairtol., 3: 1310-1312 Luu-The, V., Sugimoto, Y., Puy, L., Labrie, Y., Lopez, I., Singh, M., and Labrie, 1 (1994) J. Invest. Demtol., 102: 221-226. MacIndoe, J. (1988) Endocrinol, 123: 1281-1287. Mainwaring, W., and Mangan, F. (1973) J Endocrinol, 57: 371-184. Marchetti, B., and Labrie, F. (1988) J. stkidBiochem., 29: 691498. Martel, C., Labrie, C., Couët, J., Dupont, E., Trudel, C., Luu-The, V., Takahash M., Pelletier, G., and Labrie, F. (1990a) Mol. Ceii. Endocrinol., 72:R7-R13. Martel, C., tabrie, C., Dupont, E-,Couët, J., Trudel, C., Rheaume, E., Simard, ] Luu-The,V., and Pelletier, G. (1990b) EndocrinoIogy, 127: 2726-2737. Martel, C., Melner, M.H., Gagné, D., Simard, J., and Labrie, F. (1994) Mol. Ce1 Endocrinol., 104: 103-111. Martel, C., Rhéaume, E., Takahashi, M., Trudel, C., Couet, J., Luu-The, V Simard, J., and tabrie, F. (1992) J. Steroid Biochem. Molec. Biol., 41: 597-603 Martel, C., Trudel, C., Couet, J., Labrie, C., Bélanger, A., and Labrie, F. (199; Mol. Cell. Endocrinol., 91: 43-49. Martini, L.Z., S, and Motta, M. (1986) J Steroid Biochem, 24: 177-182. Mather, J., Liotta, A., Berker, R., Krieger, D., and Bardin, C. (1983) J Steroi Biochem, 19: 41-51. Matias, J., Malloy, V., and Orentreich, NI (1988) J Invest Dermatol, 91: 429-433. Matias, J.R, and Orentreich, N. (1983) J. Xnvest. Dermatol., 81: 43-46. Mavcais-Jarvis, P., Bercovici, J., and Gauthier, F. (1969) J Clin Endocrino1 Metal 29: 41741. Meikle, A., Burton, G., Stringham, J., and et al. (1979) J Bio Chem, 235: 3112. Meikle, A, Collier, E., Middleton, R, and Fang, S. (1980) J. Clin. Endocrina Metab., 51: 945-947. Meikle, A., Collier, E., Stringham, J., and et al. (1981) J Steroid Biochem, 14: 331. Mellin, T., Busch, R, and Rasmusson, G. (1993) J.Steroid Bidem. Molec. Biol., 44: 121-131. Messina, M., Manieri, C., Rizzi, G., Gentile, L., and Milani, P. (1985) Cuir. Ther. Res., 38: 264-282. Metcalf, B., Holt, D., Levy, M., Erb, J., Heaslip, J., Brandt, M., and Oh, H. (1989) Bioorgan Chem, 17: 372-376. Metcalf, B., Jund, K., and Burkhart, J. (1980) Tetrahedron Lett, 21: 15-18. Mettlin, C., Natarajan, N., and Murphy, G.P. (1982) kit. Adv. Surg. Oncol., 5: 277-321.

Migeon, C.J., Kelier, A.R., Lawrence, B., and Shepart I[, T.H. (1957) J. Clin. Endocrinol. Metab., 17: 1051-1062. Milewich, L., Shaw, C.E., Doody, KM., Rainey, W.E., Mason, J.I., and Cam, B.R. (1991) J. Ch.Endoainol. Metab., 73: 1134-1140. Milewich, L., Winters, A., Stephens, P., and MacDonald, P. (1977) J Steroid Biochem, 8: 823-822. Miller, J.A., Wojnarowska, F.T., and Dowd, P.M. (1986) Br. J. Dermatol., 114: 705-716. Miller, W. (1991) J Steroid Biochem Mol Biol, 39: 783-790. mer,W., Coit, D., Baxter, J., and Martial, J. (1981) DNA, 1: 37-50. Miller, W.L. (1988) Endocr. Rev., 9: 295-318. Milne, J. (1969) Br J Dermatol, 81(Suppl2): 23-28. Minesita, T., and Yamaguchi, K. (1964) Steroids, 4: 815430. Minesita, T., and Yamaguchi, K. (1965) Cancer Res., 25: 1168-1175. Moghissi, E., Ablan, F., and Horton, R (1984) 1. Ch.EndocMol. Metab., 59: 417- 421. Moguilewsky, M., and Bouton, M. (1988) J Steroid Biochem, 31: 699-710. Monslave, A., Blaquier, JA. (1977) Steroids 30: 41-51 Montagna, W., and Parakkal, P.F. (1974) The structure and function of skix~3rd ed. New York, Academic Press Moore, R (1943) J Uml, 50: 680. Moore, R (1944) Surgery, 16: 152. Moore, R., Gazak, J., and Quebbeman, J. (1979) J Clin Invest, 64: 1003. Moore, R, and Wilson, J. (1976) J. Biol. Chem., 251: 5895-5900. Morimoto, L, Edmiston, A., and Horton, R (1980) J Ciin Invest, 66: 612415. Morimoto, t, Eto, S., houe, S., IzuLni, M., Nagataki, S., Saito, Y., and Hasa, 1 (1991) J stemid Biochen Molec. Biol., 38: 227-231. Morishige, K., Kuriche, H., Amemiya, K., Fujita, Y., Yamamoto, T., Miyake, A and Tanizawa, 0. (1991) Cancer Res., 51: 5322-5328. Mowszowicz, I., Riahi, M., and Wright, F. (1981) J. Ciin. Endoainol. Metab., 5: 338-344. Muhieniam, M.F., Carter, G.D., Gream, J.J., and Wise, P. (1986) Br. J. Dermatol 115: 227-232.

Nakai, H., Terashima, H., and &ai, Y. (1988) Eur Pat Appl No EP O 291 245 A: No EP O 291 245 A2 Nakayama, O., Arakawa, H., Yagi, M., Tanaka, M., Kiyoto, S., Okuhara, S., an1 Kohsaka, M. (1989a) J Antibiotics, XLII: 1230-1234. Nakayama, O., Yagi, M., Tanaka, M., Kiyoto, S., Okuhara, S., and Kohsaka, M (1989b) J Antibiotics, XLII: 1221-1229. Nassaro-Porro, M., Psaai, S., Picardo, M., Breathnach, A., Clayton, R., and G. (1983) Br J Dermatol, 109: 45-48. Naviile, D., Keeney, D., Jenkin, G., Murry, B., Head, J., adJI, M. (1991) Mol Endocrinol., 5: 1-1100. Neri, R, Fiorance, K., Koziol, P., and Cleave, S.V. (1972) Endocrinology, 91: 427 437.

Neri, R, Monab, M.D.,Meyer, J.G., Afonso, B.A., and Tabachnick, LA. (1x7 Eur. J. Pharmacol., 1: 4M. Neri, R, and Peets, E. (1972) Invest Urol, 10: 123-130. Neri, R, and Peets, E. (1975) J Steroid Biochern, 6: 815-819. Neri, R, Pm,E., and Watnick, A (1979) Biochem Soc Trans, 7: 565-569. Neumann, F., and Topert, M. (1986) J Steroid Biodrem, 25: 855-895. Neumann, Fa,Elger, W., Steinbedc, H., and von Berswordt-Wallrabe, R (1968) Symep Deut Ces Endokrinol, 13: 73-101. Neuinann, F., and Kramer, M. (1964) Endoainol, 75: 428-433.

Neuma~,F., and Steinback, Ho (1991) In: Eider, O., Farh, A., Herken, H. and Weleh, A-D. (eds.), Berlin Heidelberg, pp. 236. New York: Springer-Verlag. Neumann, F., von Berswordt-Wallrabe, R, Elger, W., and Steinbeck, H. (1967) Mosbadher KoUoq Ges Bi01 Chem, 18: 21û-248. New, M., Lorenzen, F., Lerner, A., Kohn, B., et al., (1983) J. Clih endoainol. Metab, 57: 320.326. Nielsen, P.G. (1983) Derma tologica, 166: 275-276. Noguchi, S., Nishizawa, Y., Uchida, N., Yamaguchi, K, sato, B., Kitamura, Y., and Matsumoto, K. (1985) Cancer Res.: 5746-5750. Nordin, B.E.C., Robertson, A., Seamark, RF., Bridges, A., Philcox, J.C., Need, A.Gw,Horowitz, M., Morris, H.A., and Deam, S. (1985) J. Clin Endocrinol. Metab., 60: 651457. Norrnington, K, and Russell, D. (1992) J.Bio1. Chem.,267: 19548-19554.

Nowell, P.C. (1976) Saence, 194: 23-28.

Oeda, Ta, Lee, Y., Driscoll, W., Chen, H., and Strott, C. (1992) Mol Endocrinol, 6: 1216-1226. Ogura, K., Kajita, J., Norihata, H.,Watanabe, T., Ozawa, S., Nagata, K, Yamazoe, Y., and Kato, R (1990) Biochem Biophy Res Commun, 166: 1494-1500. Orentreich, N., Brind, J.L., Rizer, R.L., and Vogelman, J.H. (1984) J. Clin. Endoctinol. Metab., 59: 551-555. Pajunen, A.E.I., Isomaa, V.V., Ja~e,O.A., and Bardin, C.W. (1982) J. Biol. Chem.,257: 8190.8198. Parker, LN., and et al (1980) Endocr. Rev., 4: 392410.

Paqu,J., çchatz, B., Varin, Ce,and Nguyen, B. (1992) J Steroid Biochem, 41: 323-329. Peariman, W., and Pearlman, M. (1961) J Biol Chern, 236: 1321-1327. Peirone, F., and Femgho, S. (1965) Venereol Sessuol, 33: 455466. Pelletier, G., Dupont, E., Simard, J., Luu-The, V., Bélanger, A., and Labrie, i (1992) J. Steroid Biochem. Mol. Biol., 43: 451467. Pelletier, G., Labrie, C., Simard, J., Duval, M., Martinoli, MG., Zhao, H.F.,an Labrie, F. (1988) J. Mol. Endocrinol., 1: 213-223. Peltoketo, H., Isumaa, V., Maentausta, O., and Vihko, R (1988) FEBS Lett., 23' 73-77. Peltoketo, Id, Isomaa, V., and Vihko, R (1992) Eur. J. Biochem., 209: 459-466. Perel, E., Wilkins, D., and Killinger, D. (1980) J. Steroid Biochem., 13: 89-94. Peter, R, hperto-MGinley, J., Gautier, T., and Sturla, E. (1977) Am J Med, 62: Picha, G., Saunders, F., and Green, D. (1952) Science, 118: 704-705. Pita, J., Lippman, M., Thompson, E., and Loriaux, D. (1975) Endoainol, 97: 152: 1527. Plante, MGLapointe, S., and Labrie, F. (1988) J. Steroid Biochem., 31: 61-64. Plewig, G. (1974) Br. J. Dermatol., 90: 623-630. Plewig, G., Christophers, E., and Braun-Falco, O. (1971a) Acta Derm Venera (Stockh), 51: 423-428. Plewig, G., Christophers, E., and Braun-Falco, O. (1971b) Acta Derm Venera (Stockh), 51: 413422. Plewig, G., and Luderschmidt, C. (1977) J ïnvest Dermatol, 68: 171-176. Plewig, G., and Luderschmidt, C. (1982) J. Invest. Dermatol., 68: 171-176. Plewig, G., and Luderschmidt, C. (1991) In: Hammerstein, J. and Lachnia-Fixoi U. (eds.), Androgenization in women, pp. 96-102. Amsterdam Oxforc Excerpta Medica. PoM, P., Strauss, J., and Mescon, H. (1962) J hvest Dermatol, 39: 47M. Pochi, PX., and Straus, JS. (1964) J. Invest. Dermatol., 43: 383-388. Pochi, P.E., and Strauss, J.S. (1969) J. ïnvest. Dermatol., 52: 32-36. Pochi, P.E., and Strauss, J.S. (1974) J. Invest. Dermatol., 62: 191401. Poulin, R, and Labrie, F. (1986) Cancer Res., 46: 493314937. Poyet, P., and iabrie, F. (1985) Mol. CeiL Endoainol., 42: 283-288. Presti, JC.,Fair, WR, Andriole, G., Sogani, PC., Seidmon, EJ., Ferguson, D., Ng, J, Gormely, GJ. (1992)J Ur01 148: 1202-1204 Ph, SIDI,Greenblatt, RB., and Mahesh, V.B. (1969) J. ïnvest. Dermatol., 52: 348 350. Price, V.H. (1975) Arch. Dermatol., 111: 1496-1502. Prost, O., Ottignon, Y., Remy-martin, A., Vuitton, D., Miguet, P., and Adessi, G (1984) Steroids, & 189-199. Radominska, A., Corner, K., Zimniak, P., Falany, J., Iscan, M., and Falany, C (1990) Biochem J, 272: 597604. Ramasastry, P., Downing, D., Pochi, P., and Strauss, J. (1970) J. Invest. Dermatol, 54: 193.

Randail, L, and Selitto, j. (1958) Endocrinol, 62: 693-695. Rasmusson, G., Reynolds, G., Steinberg, N., Walton, E., Patel, G., Liang, T, Cascieri, M., Cheung, A., Brooks, J., and Berman, C. (1986) J Med Chem, 25 2298-2315. Rasmusson, G., Reynolds, G., Utne, T., Jobson, R., Primka, R., Berman, C., anc Brooks, J. (1984) J Med Chem, 27: 1690-1701. Raynaud, J., Bonne, C., and Bouton, M. (1979) J Steroid Biochem, 11: 93-99. Regelson, W., Loria, R., and Kalimi, M. (1988) Ann. New York Acad. Sci., 521 260-273. Rhéaume, E., Lachance, Y., Zhao, H.F., Breton, N., de Launoit, Y., Trudel, C Luu-The, V., Simard, J., and Labrie, F. (1991a) Mol. Endocrinol., 5: 1147415 Rhdaume, E., Simard, J., Morel, Y., Mebarki, F., Zachma~,M., Forest, M., New ML, and Labrie, F. (1992) Nature Genet., 1: 239-245. Rhéaume, E., Sirois, I., Labrie, F., and Simard, J. (1991b) Nudeic Acids Res, 1s 6060.

Ri-&, J., Jahnsen, Tm,Tedin, L., titka, J., Ratoosh, S., Durica, J., and Goldriq N. (1987) Rec Prûgr Honn W,43: 231-270. Rittmaster, R (1988) Clin. Dermatol., 6: 122-128. Rittmaster, R, Vno, H., Povar, M., Meilin, T., and Loriaux, D. (1987) J. Clin. Endocrinol. Metab., 65: 1û8-193. Ritzén, E., Nayfeh, S., French, F., and Aronin, P. (1972) Endocrinal, 91: 116-124. Ritzén, E., Nayfeh, S., French, F., and Dobbins, M. (1971) Endoainol, 89: 143451. Roballe, Bv Covey, D., Robinson, C., and Ewing, L (1977) J Steroid Biochem, 8: 307-310. Rodbard, D., and Lewald, J.E. (1970) Cornputer analysis of radioligand assay and tadioimmunoassay data. Second Karolinska Symposium on Research Methods in Reproductive Endoainology, p. 79-103. Rony, H., and Zakon, S. (1943) Arch Dermatol, 48: 601-603. Rose, D.P., Stauber, P., Thiel, A., Crowley, J.J., and Miibrath, J.R. (1977) Eur. J, Cancer, 13: 43-47.

Roseiii, C., and Resko, J. (1989) Bi01 Reprod, 40: 929-934. Rosenfield, J. (1986) Clin EndOctinoI-and Metab., 45: 341-362. Roy, A. (1992) Porc Soc Exp Bi01 Med, 199: 265272-

Rubin, B., and Dorfman, R (1965) Proc. Soc. Exp. Bi01 ( N'Y), 91: 585-586. Ruoff, B., and Daniel, W. (1991) Comp Biochem Physiol, 98B: 313-322. Sansone, G.,and Reisner, R (1971) J. Invest. Dermatol., 56: 366-372. Sar, M., and Stumpf, W. (1972) Experientia, 28: 1364-1366. Sar, M., and Stumpf, W. (1973) EndOczinool, 92: 251-256. Sawaya, M.E.,and Pemeys, N.S. (1991) J. Cutan. Pathol., 19: 309-314. Schroder, HG.,Ziegler, M., Nidcisch, K,Kaufmann, J., and Etreby, F.E. (1989) J, Invest. Dermatol., 92: 769-773. Schwartz, A.G. (1979) Cancer Res., 39: 1129-1132. Schwartz, A.G., Whitcomb, J.M., Nyce, J.W.,Lewbart, ML., and Pashko, L.L. (1988) Adv. Cancer.Res.,51: 391424. Schwartz, F., and Flink, E. (1985) Postgrad. MED., 77: 81-93. Schweikert, EU., and Wilson, J.D. (1974) J. Clin. EndOQiilOl. Metab., 38: 811419, Sciarra, F., Toscano, V., Concolino, G., and Di Silveio (1990) J Steroid Biocherr Molec Biol, 37: 349-362. Scott, W., and Wade, J- (1969) J Urol, 101: 81-85. Scrable, H.J., Sapienza, C., and Cavenee, W.K. (1990) Adv. Cancer Res., 54: 25-62. Seki, T., Ideta, R, Shibuya, M., and Adachi, K. (1991) J hvest Dennatol, 96: 926. 931. Serafini, P., and Lnbof RA. (1985) Fertil. Sied., 43: 74-78. Shida, K, Yamanaka, H.,Koya, A., Moi, H., and Shibata, K. (1980) EndOCriCLO Japon, 27: 69-76. Süteri, P., and Wilson, J. (1974) J Clin Endocriml Metab, 38: 1101125. Süteri, P., Wilson, J., and Mayfield (1970) J. Clin. Invest., 49: 17374745. Simard, J., Couët, J., Durocher, F., Labrie, Y., Sanchez, R, Breton, N., Turgeon, C. and Labrie, P. (1993a) J. Biol. Chem., 268: 19569-19668. Simard, J., Dauvois, S., Haagensen, D.E., Lévesque, C., Mérand, Y., and Labrie, F (1990) End~ology,126: 3223-3231. Simard, J., de Lnunoit, Y., and Labrie, F. (1991a) J. Biol. Chem., 266: 14824-14&45. Simard, J., and Labrie, F. (1987) J. Steroid Biochem., 26: 539-546.

Simard, Je, Luthy, I., Guay, J., Bélanger, A., and Labrie, F. (1986) Mol. CeU Endoainol., 44: 261-270. Simard, J., Melner, M.H.,Breton, N., Low, KG., Zhao, H.F., Periman, LM.,ant Labrie, F. (1991b) Mol. Cell. Endoainol., 75: 101-110. Simard, J., Rhéaume, E., Leblanc, J.F., Wallis, S.C., Joplin, G.F., Gilbey, S. manson, J., Mettler, G., Bettendorf, M., Heinrich, U., and Labrie, F. (1994 Hum. Mol. Genet., 3: 327-330. Simard, J., Rhgaume, E., Sanchez, R., Laflamme, N., de Launoit, Y., Van Seters A.P., Gordon, R.D., Bettendorf, M., Heinrich, U., Moshang, T., New, M.I. and Labrie, F. (1993b) Mol. Endaxinol., 7: 716-728. Simard, J., Vincent, A., Duchesne, R, and Labrie, F. (1988) Mol. Ce11 Endocrinol., 55: 233-242. Smith, J., and Brunot, F. (1%1) Acta Oerm Venereol, 41: 6145. Smith, ME., and Jenkirwn, D.M. (1975) J Agric Sci, 84: 57-61. St-Arnaud, R, Landhance, R., Kelly, S., Belanger, A., Dupont, A, and Labrie, E (1986) Clin Endo&l, 24: 21-30. Stamatiadis, D., Bulteau-Portois, M., and Mowszowicz, 1. (1987) Nouvelle Demtol, 6: 601403. Stamatiadis, D., Bulteau-Portois, M., anci Mowszowicz, 1. (1988) Br J Dermatol 119: 62742. Stern, J., and Eisenfeld, A. (1969) Science, 166: 233-235. Stinson, S. (1992) Chem Eng News, 29: 7-8. Stoner, E. (1990) J. Steroid Biochem. Mol. Biol., 37: 375-378. Strauss, J., Kligman, A., and Pochi, P. (1962) J Invest Dermatol, 19: Strauss, J.S., Pochi, P.E., and Downing, D.T. (1976) J. Invest. Dermatol., 67: 90-97, Sudo, K.,Yoshida, K., Akinaga, Y., and Nakayama, R (1981) Steroids, 38: 55-71. Sudo, K., Yoshida, K., and Nakayama, R (1979) Acta Endocrinol, 92 (suppl229: 82-99.

Suzuki, K., and Tamaoki, B. (1973) Steroids Lipids Tes., 4: 266-276.

Suzuki, K., and Tomaoki, B. (1974) J. Steroid Biochem., 5: 249-256.

Svensson, J., and Snodiowski, M. (1979) J. Clin. Endocrinol. Metab., 49: 340-345. Swyer, G. (1944) J Anat (Lond), 78: 130. Taga, S., Yoshida, N., and Sekiba, K. (1990) J Steroid Biochem Mol Biol, 37: 741 745.

Takahashi, M., Luu-the, V., and Labrie, F. (1990) J Steroid Biochem Molec Bi01 37: 231-236. Takayasu, S., and Adachi, K. (1970) Takayasu, S., and Adachi, K. (1972) J. Clin. Endacrinol. Metab., 34: 109û-1101. Takayasu, S., and Itami, S. (1982) Endoaiml Japon, 29: 95-98. .. Takayasu, S., Wakimoto, H., Itarni, S., and et ai (1980) J. Invest. Dennatol., 74 187-191. Takezawa, Y., Fuicabore, Y., Yamanaka, H., Mieda, M., Honma, S., Kushitani, M and Hamataki, N. (1992) Prostate, 21: 315329. Tenover, J., Zeitner, M., and Plymate, S. (1989)Effects of. 71th annual meeting c the endobine wciety, Seattle, Washington, p. Abstrad No 583. Theriault, C., and Labrie, F. (1991) J. Stemid Biochern. Molec. Biol., 38: 155164. Thoman, M., and Weigle, W. (1989) Adv. Immununol., 46: 221-260. Thomas, J.P., and Oake, RJ. (1974) J. Clin. Endoainol. Metab., 38: 19-22. ThdaU, D., Chang, C., Lobl, T., and Cunningham, G. (1984) Phannac Ther, 24 367400. Toscano, V. (1991) J Endocrinol Invest, 153-170. Traupe, H., Muhlendal, K., Bramswig, J., and et al. (1988) J. Invest. Derrnatol 74: 21-25. Trcker, H., 'JW,C., and Chesterson, G. (1988) J Med Chem, 31: 954-959. Tremblay, Y., and Beaudoin, C. (1993) Molec Endocrinol, 7: 355-364. Tveter, IC, 'and Akag, A. (1969) Endocrinol, 85: 68349. Vaccaro, A., Salvioli, R., Muscillo, M., and Renola, L. (1987) Enzyme, 37: 11: 126. Valavaara, R (1990) Breast Cancer Res. Treat., 16 (Suppl.): 531-535. Vaux, D.L., Haecker, G., and Strasser, A. (1994) Cell, 76: 777-779. Venning, V.A., and Dawber, RP.P. (1988) J. Am. Acad. Dermatol., 97: 1073-1077 Vermeulen, A., Giagulli, V., De Schepper, P., Buntinx, A., and Stoner, E. (1985 Prostate, 14: 45-53. Vermorken, A., Goos, C., and Roelofs, H. (1980) Br. J. Derrnatol., 102: 455-460. Vermorken, A., Goos, C., and Wirtz, P. (1982) Br. J. Dermatol., 106: 99-101. Veterans Administration Cooperative Urological Research Group (1967) Sur4 G-1. Obstet., 124: 1011-1017. Viguier-Martinez, M., Hocheseau de Reviers, M., Barenton, B., and Perreau, C (1983) Acta End-1, 102: 299,306. Voigt, W., Fernandez, E., and Hsia, S. (1970) J. Biol. Chem., 245: 5594-5599. Voigt, W.,and Hsia, S. (1973a) J Bi01 Chem, 248: 42MX285. Voigt, W., and Hsia, S.L. (1973b) Endocfinology, 92: 1216222. Walsh, P., Madden, J., Hatrod, M., Goldstein, J., MacDonald, P., and Wilson, J4 (1974) New Engl. J. Med., 291: 944-949. Walsh, P., and Wilson, J. (1976) J Ciin Invest, 57: 1093. Walton, S.# Cdfe, W.J., Lookingbill, O.P., and Keezkes, K. (1986) Br. Jc Dermatol., 114: 261-269. Weichert, R (1966) U.S.Patent 3,234#93, Weisenborn, F., and Applegate, N. (1959) J Am Chem Soc, 81: 1960-1969.

Weissman, A., Bowden, J., Frank, B.L., Horwitz, SN., and Frost, P. (1984) J, Inves t. Dermatol., 82: 522-525. Weissmam, A., Bowden, J., Frank, B.L., Horwitz, S.N.,and Frost, P. (1985: Arch. Dermatol., 121: 57-62. West, N., Roselli, C., Resko, J., Greene, G.,and Brenner, R. (1988) Endocrino! 123: 2312-2322. Wiiborn, T., corner, K., Dooley, T., Reardon, I., Heinrikson, R., and Falany, C (1992) Mol Pharmacol, 43: 70-77. Wiison, J. (1980) Am. J. Med., 6û: 745-756. Wilson, J., and Walker, J. (1969) J. Clin. Invest., 48: 371-379. Wing, LX., Garrett, W.M.,and Brodie, A.M.H. (1985) Cancer Res., 45: 24252428. Wingo, P.A., Tong, T., and Bolden, S. (1995) CA Cancer J. Clin., 45: 8-30. Winkier, K., and Schaefer, H. (1973) Archiv fur Dermatologische Forschung, 247: 259-264. Wu, L., Einstein, M., Geissler, W.M.,Chan, K.H.,Elliston, K.O., and Andersson, S. (1993) J. Biol. Chem, 268: 12964-12969. Yamaguchi, K., Minesta, T., Kasai, K., Kurachi, K., and Matsumoto, K. (19ïlI Steroids, 17: 345455. Yates, E., and Urquhart, J. (1962) Physiol Iies, 42: 35941. Zarate, A., Mahesh, V.B., and Greenblatt, RB. (1966) J. Clin Endoainol. Metab., 26: 1394-1398. Zhao, H.F., Labrie, C., Simard, J., de Lamit, Y., Trudel, C., Martel, C., Rhéa- E., Dupont, E., Luu-The, V., PeUetier, G., and Labrie, F. (1991) J. Biol. Chem 266: 5û3-593. Zhao, H.F., Rhéaume, E., Trudel, C., Cou& J., tabrie, F., and Simard, J. (199( Endocrinology, 127: 323703239.

Zyirek, M., Flood, C., and Longcope, C. (1987) Proc SOC Exp Bi01 Med, 186: 1% 138. mi ri^ .w- bvtrnbwr-u WVII TEST TARGET (QA-3)

APPLIED IMAGE. lnc - 1653 East Main Street --= --= - Rochester, NY 14609 USA - Phone: 71 6148%Wûû ------Fax 7161288-5989

O 1993. Applbd image. IE, An Righis Reserwd